TRACE SW 4024 - True Sine Wave Inverter
John A. Stanley
ricky...@aol.com (RickyB1111) wrote:
>I have reciently been told that the TRACE Sine Wave model SW4024 is a true
>sine wave and not an approximation like the lower end TRACE DR2424. Is
>this true ? What are the facts around this ?
The waveform from the SW4024 approximates a sine wave with what
Home Power Magazine calls a "Mayan temple" waveform. The 4024
uses between 34 and 52 steps per cycle depending on battery
voltage and/or load.
I upgraded to the 4024 from a 2624, and the difference is
astonishing. Electric motors start more easily and have more
power. There are no squiggly lines in the TV, and the inverter
noise in my stereo system is so low that I can only hear it when
no audio signal is present. It has no problem with loads in my
house like the dishwasher, washing machine, and even the
submersible cistern pump. It also automatically starts my
back-up generator when needed and diverts the wind power to an
alternative load when the battery is full.
The Trace SW4024 is quite simply, IMHO, the best inverter ever
made.
--
John A. Stanley jsta...@gate.net
Bruce Simpson
>Questions:
>1) What shuts the backup generator off again ? Is this
> a programmable time interval or other arrangement ?
>2) What are you using for an alternative load ? Water
> heater perhaps ?
Gasp! Nobody using solar, wind or other low-output electrical generation
should be using a water heater unless it's a heat-pump. Heating water with
valuable electricity is such a waste!
>3) Does the startup/shutdown of the generator create
> any noticeable glitches or voltage fluctuations ?
>>The Trace SW4024 is quite simply, IMHO, the best inverter ever
>>made.
>Sounds tempting
My wind & solar power system (currently under design, hopefully implemented
next year when I move house) doesn't require an inverter. It seems such a
waste of money to store electricity in 12V or 24V batteries and then have
to run it through an inverter to convert 12/24->110 (or in my case 230VAC).
My system uses a rewound alternator which gives me 200VDC which I use to
charge a bank of 13 12V batteries wired in *series* such that I have about
180Volts on tap. That 180VDC is then reticulated throughout the house and
at each power-point there is a small FET-bridge which acts as a pulse-width
modulator to give me a very accurate 230VAC sine-wave output. By doing
things this way you get the following benefits:
1. Lighter wiring can be used from the generator and the batteries since
I2R losses are dramatically reduced thanks to the much lower currents.
2. No inverter is required, just a simple FET bridge and driver circuit.
3. Much higher powered appliances can be run without the need for a huge
inverter which would most likely lose significant efficiency under light
loads.
4. Since the FET bridges are so simple and cheap to produce, and since you
have multiple units around the house you're not going to be adversely
affected by a failure in the way that you would if you were using a single
inverter to run all your appliances.
5. The FET bridge can be designed to operate at quite high pulse-rates
(5Khz+) which makes filtering the output wave very easy with the result
that RFI is significantly reduced.
I was thinking of going commercial with these things if there was enough
interest.
*=[FREE FAX GATEWAY INTO THE AUCKLAND, NZ TOLL-FREE AREA]=*
* sponsored by FaxMail Technologies *
* Send email to in...@faxmail.co.nz for details *
*=========================================================*
John A. Stanley
br...@faxmail.co.nz (Bruce Simpson) wrote:
>ro...@hal.COM (Robert Kleinschmidt) wrote:
>
>>2) What are you using for an alternative load ? Water
>> heater perhaps ?
>
>Gasp! Nobody using solar, wind or other low-output electrical generation
>should be using a water heater unless it's a heat-pump. Heating water with
>valuable electricity is such a waste!
The electricity gets diverted to heating water only when the
battery bank is full. Wind turbines must always see a load
otherwise they fly apart, so when the battery can no longer be
the load the power must be diverted to an alternative load such
as a water heater.
If you were suggesting that the alternative load should be a
heat pump instead of resistance heaters, I can guarantee that it
wouldn't work because when the wind power is diverted from the
battery the battery voltage drops. When it drops the relay lets
the wind power flow to the battery again. When the battery is
full that relay cycles on and off every few seconds, sending
intermittant pulses of power to the diversionary load, and if
the load were a motor running a compressor it would likely
destroy the compressor. And even if one were to program some
hysteresis into the relay's set point (which the Trace can do),
the cycling times would still very likely be too short to safely
run a compressor.
John A. Stanley
ro...@hal.COM (Robert Kleinschmidt) wrote:
>In article <VgJxwcGU...@gate.net>,
>John A. Stanley <jsta...@gate.net> wrote:
>>
>>I upgraded to the 4024 from a 2624, .........
>>......................... It also automatically starts my
>>back-up generator when needed and diverts the wind power to an
>>alternative load when the battery is full.
>
>
>1) What shuts the backup generator off again ? Is this
> a programmable time interval or other arrangement ?
There is one relay that activates the genset's ignition and one
relay that activates the starter motor. I set mine to come on
when battery voltage drops to a certain voltage for ten minutes
(that eliminates genset starting when there is a voltage drop
from a large load's initial surge.) You can also program quiet
times when the genset won't come on unless the volatge drops to
an even lower voltage.
>2) What are you using for an alternative load ? Water
> heater perhaps ?
My wind energy comes to the house as 240 volt 3-phase AC where
it is stepped down and rectified. When the battery is full, the
Trace activates a relay that sends the 240vac to three electric
hot water elements in the solar thermal storage tank.
>3) Does the startup/shutdown of the generator create
> any noticeable glitches or voltage fluctuations ?
When the Trace transfers the house over to genset power and
starts charging the batteries there is a small glitch that
results in a small flicker of the lights. I have not experienced
any problems with electronics from the glitch.
Robert Kleinschmidt
>
>>2) What are you using for an alternative load ? Water
>> heater perhaps ?
>
>should be using a water heater unless it's a heat-pump. Heating water with
>valuable electricity is such a waste!
>
The question referred to excess electricity produced when the
batteries were in a state of full charge. (This is assuming that
somewhere on the planet there may be an alternative energy household
with a surplus of energy). If not water heating, then what ?
Pumping ?
While I completely agree that water heating is not a good primary
use for self-generated electric, seems as though an electrical
pre-heat tank for surplus electricity feeding to a tankless propane
unit would be a reasonable load management trick.
>My system uses a rewound alternator which gives me 200VDC which I use to
>charge a bank of 13 12V batteries wired in *series* such that I have about
>180Volts on tap. That 180VDC is then reticulated throughout the house and
>at each power-point there is a small FET-bridge which acts as a pulse-width
>modulator to give me a very accurate 230VAC sine-wave output.
This is certainly an interesting idea, although I would think
it might have the following drawbacks:
1) For a PV charged system, there is the question of how
to add incremental charging. The need to purchase 12 v.
PV modules in 15 panel increments would seem to be a
substantial problem.
2) To my mind, high voltage DC seems to have some safety
issues. (Arcing, need for special switches, etc.). I
have also heard it asserted that AC "lets go of you"
more readily than DC, should you complete a circuit
with your hands.
3) Each FET-bridge would have to be sized for max load,
and would have to have a safe failure mode. Possible
"cross-talk" problems as well ?
4) Inverter efficiency is steadily increasing, and
best of my recollection can stay above 90 % through
most of the load range. Is there a real advantage
to be had here ?
These are questions of genuine interest, and in no way intended
as flames. Looking forward to your response.
Rob Kleinschmidt
D.F.S.
: In article <4a6nmr$h...@news.express.co.nz>,: >
: >>2) What are you using for an alternative load ? Water
: >> heater perhaps ?
: >
: >Gasp! Nobody using solar, wind or other low-output electrical generation
: >should be using a water heater unless it's a heat-pump. Heating water with
: >valuable electricity is such a waste!
:
: battery bank is full. Wind turbines must always see a load
: otherwise they fly apart, so when the battery can no longer be
: the load the power must be diverted to an alternative load such
: as a water heater.
:
: If you were suggesting that the alternative load should be a
: heat pump instead of resistance heaters, I can guarantee that it
: wouldn't work because when the wind power is diverted from the
: battery the battery voltage drops. When it drops the relay lets
: the wind power flow to the battery again. When the battery is
: full that relay cycles on and off every few seconds, sending
: intermittant pulses of power to the diversionary load, and if
: the load were a motor running a compressor it would likely
: destroy the compressor. And even if one were to program some
: hysteresis into the relay's set point (which the Trace can do),
: the cycling times would still very likely be too short to safely
: run a compressor.
The simple answer to this is to put the compressor in the system
as a regular load just like anything else and switch it on to generate
a load.
It shouldn't need to be an either/or situation of batt or excess
load.
You could use a compressor with a DC motor and Pulse-width-modulate
the DC direct from the charging system to get a variable speed compressor
to allow you to keep a constant system load and keep the turbine
running at optimum speed and efficiency.
Marc Christensen
Bruce Simpson
>The question referred to excess electricity produced when the
>batteries were in a state of full charge. (This is assuming that
>somewhere on the planet there may be an alternative energy household
>with a surplus of energy). If not water heating, then what ?
>Pumping ?
I've actually wondered about this myself. There have been lots of "ideas"
for storing excess energy in situations like this but most of them (huge
basment flywheels, pumping water into a tower, etc) tend to be fairly
impractical for the average home user. About the best idea I can come up
with is using it to heat a 5000 gallon tank of water which has been
insulated and buried underground. The heat thus stored could be used
during the winter months for internal heating through radiators or via
heat-exchangers and ducted airflows. Even in this situation I'd still opt
for a heat-pump in preference to a resistive immersion heater. The
differences in efficiency are *huge*.
>>My system uses a rewound alternator which gives me 200VDC which I use to
>>charge a bank of 13 12V batteries wired in *series* such that I have about
>>180Volts on tap. That 180VDC is then reticulated throughout the house and
>>at each power-point there is a small FET-bridge which acts as a pulse-width
>>modulator to give me a very accurate 230VAC sine-wave output.
>This is certainly an interesting idea, although I would think
>it might have the following drawbacks:
>1) For a PV charged system, there is the question of how
> to add incremental charging. The need to purchase 12 v.
> PV modules in 15 panel increments would seem to be a
> substantial problem.
Maximum PV charging currents are significantly less than maximum load
currents so there's no reason why you can't use existing 12V/24V PV arrays
and run them through an inverter and rectifier to create the 120V charging
voltage. I2R losses are much lower on this side of the equation and you
can use a *much* smaller (cheaper/simpler) inverter for this purpose.
>2) To my mind, high voltage DC seems to have some safety
> issues. (Arcing, need for special switches, etc.). I
> have also heard it asserted that AC "lets go of you"
> more readily than DC, should you complete a circuit
> with your hands.
There is nowhere that you need to switch the DC voltages (using a switch or
relay) regularly. The Fet bridge uses (passes) no current when there's no
load and so the switch can be on the AC side. In this situation it's
exactly the same as switching normal 110VAC. As regards to the safety
issues of DC vs AC, I've read nothing to substantiate the "belief" that AC
"lets go of you". If you examine the situation: at 50Hz there are two
zero-voltage points and two periods when the voltage is "safe", say 50V or
less. Assuming that these two "safe" periods account for 50% of the cycle
we then find that the length of such a "safe" period is around 20mS. I
don't think that your muscles could relax quickly enough to let go of
anything "live" that you might have grabbed ahold of.
The other consideration is that the DC wiring should not be in a location
where it can be inadvertantly touched and all maintenance should be
performed *only* after disconnecting the batteries from the wiring by use
of a master switch.
>3) Each FET-bridge would have to be sized for max load,
> and would have to have a safe failure mode. Possible
> "cross-talk" problems as well ?
The FETs in a FET-bridge are only carrying 1/10th to 1/5th the current of
the switching transistors in a 12/24V to 110VAC inverter and whereas an
inverter may need to supply several outlets, each outlet can have its own
FET-bridge thus reducing the maximum loading per unit.
The safe failure-mode is very easy to arrange. There is a over-current
fuse on the DC side to protect against both FETs on one side of the bridge
going short-circuit. The output of the bridge is also monitored by a
simple circuit which immediately detects any open-circuit or
drive-circuitry failures and reacts to instantly shut the bridge down.
>4) Inverter efficiency is steadily increasing, and
> best of my recollection can stay above 90 % through
> most of the load range. Is there a real advantage
> to be had here ?
Yes, the advantages include:
- increased redundancy. Instead of all AC outlets being dependent on a
single inverter, each outlet has its own FET-bridge. The failure of one
bridge still leaves all other outlets fully operative.
- simplicity. Good sine-wave inverters are complex and expensive, FET
bridges are by comparison very simple and cheap.
- efficiency. If your inverter is located close to your batteries then I2R
losses won't be significant but what about your generation facilities?
This is especially true if you run a wind-generator which is frequently
located some distance (ie: more than 20 yards) away. If your generator
puts out 20A or more at 12V/24V (240-480 Watts+) then I2R losses can become
significant (even .01 ohms of line resistance wastes 16 watts when it's
carrying 40A) - mandating the use of heavier (more expensive) cabling. If
you increase the voltage by a factor of five then you reduce the current by
a factor of five which reduces I2R losses by a factor of 25. This is the
very reason that the national power grid runs voltages of over 100KV, to
keep I2R losses to an absolute minimum.
>These are questions of genuine interest, and in no way intended
>as flames. Looking forward to your response.
Fair enough. I'm not someone who likes wasting money (hence the desire to
be energy self sufficient) so I've weighed up the pro's and con's very
carefully before committing to this project. The main benefits to me are:
* simplicity. A FET bridge requires just a handful of components and
eliminates the need for special transformers, large cases and high current
devices. Anyone who has basic soldering abilities and can read will be
able to assemble a FET-bridge in kitset form.
* efficiency. A FET bridge based system will always be more efficient than
an inverter because it reduces I2R losses.
* cost. A FET bridge-based system is highly scalar. You can just keep
adding bridges as you add appliances or as you build your house. The
entry-level cost for a FET bridge system is significantly lower than that
of an inverter-based system.
* reliability. *good* inverters are sufficiently expensive that not
everyone can afford to have a standby unit available. I like the prospect
of multiple redundancy that guarantees I'll never be left without power
just because one $0.10 component has failed. Thanks to their low-cost I
can afford to keep a couple of spare FET bridges lying around "just in
case".
I hope that answers your queries.
*==[FREE-FAX GATEWAY INTO THE AUCKLAND NZ TOLL-FREE AREA]==*
* FREE listings in NZ's largest WWW Business Directory *
* http://www.actrix.gen.nz/biz/faxmail/faxmail.htm *
*==========================================================*
Bruce Simpson
> The electricity gets diverted to heating water only when the
> battery bank is full. Wind turbines must always see a load
> otherwise they fly apart, so when the battery can no longer be
> the load the power must be diverted to an alternative load such
> as a water heater.
Good grief what kind of wind generator are you using?
A "good" wind generator will have a variable pitch propeller, the pitch of
which can be either mechanically controlled by a centrifugal mechanism or
(in the case of a more sophisticated unit) by a microprocessor and small
servomechanisms.
Such a system not only ensures that the generator won't over-rev but also
enables optimal efficiency over a wider range of wind speeds (including
very light winds) where fixed-pitch vanes are often almost fully stalled
and very inefficient -- just when you need maximum efficiency.
> If you were suggesting that the alternative load should be a
> heat pump instead of resistance heaters, I can guarantee that it
> wouldn't work because when the wind power is diverted from the
> battery the battery voltage drops. When it drops the relay lets
> the wind power flow to the battery again. When the battery is
> full that relay cycles on and off every few seconds, sending
> intermittant pulses of power to the diversionary load, and if
> the load were a motor running a compressor it would likely
> destroy the compressor. And even if one were to program some
> hysteresis into the relay's set point (which the Trace can do),
> the cycling times would still very likely be too short to safely
> run a compressor.
Switching the load on the basis of floating battery voltage is not a
particularly good way to manage your energy resources. It is far smarter
to measure the energy being generated, subtract the energy being used and
ad a margin for battery efficiency before determining the excess. A simple
single-chip microprocessor and interface circuitry can achieve this quite
simply.
Hugh Piggott
>
>>1) What shuts the backup generator off again ? Is this
>> a programmable time interval or other arrangement ?
>
>> heater perhaps ?
>
>Gasp! Nobody using solar, wind or other low-output electrical generation
>should be using a water heater unless it's a heat-pump. Heating water with
>valuable electricity is such a waste!
Waste is all relative in this context. As a windpower user, I dump a lot of power as heat, but it is not every day. If I built a h=
eatpump, this would not be worth while unless it was in frequent, if not constant use. I would do better building another windmill =
than a heatpump. Then I would have more power during periods of light winds. Or maybe some photovoltaics. But when there is loads=
of wind, I just enjoy basking in the dump load heat, and soaking up nature's abundance of power...
>
>4. Since the FET bridges are so simple and cheap to produce, and since you
>have multiple units around the house you're not going to be adversely
>affected by a failure in the way that you would if you were using a single
>inverter to run all your appliances.
>
>5. The FET bridge can be designed to operate at quite high pulse-rates
>(5Khz+) which makes filtering the output wave very easy with the result
>that RFI is significantly reduced.
>
>I was thinking of going commercial with these things if there was enough
>interest.
Sounds a bit ambitious, but that's not really a criticism. 240 volts Dc is said to be pretty lethatl though..
******* FROM: Hugh Piggott**********
Off the grid, on windpower since 1978
Small scale wind energy specialist.
****Scoraig, Highlands of Scotland****
John A. Stanley
br...@faxmail.co.nz (Bruce Simpson) wrote:
>>John A. Stanley (jsta...@gate.net) wrote:
>
>> The electricity gets diverted to heating water only when the
>> battery bank is full. Wind turbines must always see a load
>> otherwise they fly apart, so when the battery can no longer be
>> the load the power must be diverted to an alternative load such
>> as a water heater.
>
>Good grief what kind of wind generator are you using?
A Whisper 1000.
>A "good" wind generator will have a variable pitch propeller, the pitch of
>which can be either mechanically controlled by a centrifugal mechanism or
>(in the case of a more sophisticated unit) by a microprocessor and small
>servomechanisms.
I'm not aware of any small wind generator with those features.
All the wind generators for individual AE systems I've seen use
fixed blades with a mechanism to tilt the unit back in high wind.
dun...@shack.punk.net
>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>
>>4. Since the FET bridges are so simple and cheap to produce, and since you
>>have multiple units around the house you're not going to be adversely
>>affected by a failure in the way that you would if you were using a single
>>inverter to run all your appliances.
>>
>>5. The FET bridge can be designed to operate at quite high pulse-rates
>>(5Khz+) which makes filtering the output wave very easy with the result
>>that RFI is significantly reduced.
>>
is said to be pretty lethatl though..
To get 230v(rms) from these units he'd need more than 240vdc input,
assuming he is aiming for a sine wave output.
I wouldn't worry about switching -- just avoid capacitors on the
input to the small units, or the contact arc will wipe out the
switches.
At lower voltages I would consider dc to be safer than a corresponding
ac voltage. The body responds particularly badly to frequencies in
the 50-60Hz range. Funny that this has become the world standard.
At the 300 or so volts that he will need, both are pretty lethal.
Use fuses rather than circuit breakers. A bank of batteries wired in
series can produce more current than a typical residential feed from
the local electric company. This can weld a circuit breaker together
with the circuit still closed, defeating the purpose of the breaker.
The devices are being "marketed" here as FET-bridges. Each bridge
will require a timing and control circuit. The outputs will beat
against each other, producing voltages between two outlets which may
be unexpected and may cause trouble with some older audio equipment,
which connects signal and chassis ground to one side of the line.
An earlier poster mentioned that the devices should have a safe
failure mode. Whenever I have had a similar idea I have been
reminded by the people I work with that "H-bridges go BOOM in the night".
A central inverter, on the other hand, can provide transformer
isolation between the dc and ac components. Failures should
damage only the inverter, rather than suddenly applying dc to
your various induction motors around the house. Any audible whine
from the inverter can be locked in the garage, rather than being
allowed to come from a box mounted behind each outlet. RF noise
can be filtered at the inverter and never distributed through the
house. (wishfull thinking)
Be careful.
Duncan. ( inverters keep the computers running... )
John A. Stanley
br...@faxmail.co.nz (Bruce Simpson) wrote:
>alternative energy solutions was how lacking the whole area was in the use
>of modern technology. Inverters, fixed-pitch blades, low primary voltages,
>no use of intelligent control systems (when microprocessors are under a
>dollar each in quantity).
>
>Perhaps it's just that those in the industry are "back to basics" types or
>maybe it's just that the market will buy whatever's offered - I don't know.
I'd venture that it's just a simple case of economics and KISS.
Why create a more elaborate and more expensive system when a
simpler and less expensive one can accomplish the same task?
Personally, I'm perfectly happy that there isn't pitch control
on my windmill; it's one less thing to break.
Bruce Simpson
>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>Gasp! Nobody using solar, wind or other low-output electrical generation
>>should be using a water heater unless it's a heat-pump. Heating water with
>>valuable electricity is such a waste!
> Waste is all relative in this context. As a windpower user, I dump a lot of
> power as heat, but it is not every day. If I built a heatpump, this would not
> be worth while unless it was in frequent, if not constant use. I would do
> better building another windmill than a heatpump. Then I would have more
> power during periods of light winds. Or maybe some photovoltaics. But
> when there is loads of wind, I just enjoy basking in the dump load heat,
> and soaking up nature's abundance of power...
But why consider "dumping" energy it when you can store it and use it later
or at least offset your other energy demands by pre-heating your hot water
supply.
>Sounds a bit ambitious, but that's not really a criticism.
>240 volts Dc is said to be pretty lethatl though..
50VDC can be lethal if you don't treat it with respect. Here in New
Zealand we have a mains voltage of 230VAC which is 320V peak-to-peak yet
very few people are electrocuted each year. In fact I suspect that if
you're energy self-sufficient, you're in more danger of dying in a
house-fire caused by high-resistance terminations overheating at the high
currents demanded at 12V/24VDC than you are from electrocution.
A properly designed and installed system should *never* expose anyone to
the DC voltages, even when maintenance is required. As regards the AC,
well people all over the world are happy to live with the dangers of
110-240VAC as the cost of reticulating reasonable amounts of electrical
energy at managable current levels.
Bruce Simpson
>To get 230v(rms) from these units he'd need more than 240vdc input,
>assuming he is aiming for a sine wave output.
No, you only need 180VDC to get 230V (RMS) AC through the FET bridge.
"Gasp" .. I hear you say, "how can that be?"
Simple.. the peak to peak voltage of 230VAC is around 325. It varies from
a maximum positive amplitude of +162V to a maximum negative amplitude of
-162V. The FET bridge works by effectively reversing the polarity of the
batteries across the output for each half cycle so the voltage goes from
+162V down to zero (then the battery is effectively reversed) and up to
-162V. The result is a peak to peak voltage of 325VAC and since it's a
sine-wave, the RMS voltage is around 230V.
For 110VAC (RMS), you only need a DC voltage of around 85VDC so seven 12V
batteries in series will do the job.
One could argue therefore that since the DC voltage is less than 70% of the
peak AC voltage that this system is safer than a mains-only setup.
(before I get flamed, please note that I don't have a calculator handy so
these figures are "rough" but certainly in the ballpark.)
>I wouldn't worry about switching -- just avoid capacitors on the
>input to the small units, or the contact arc will wipe out the
>switches.
You can perform "zero current" switching through the FET bridges anyway
which totally eliminates the arc.
>At lower voltages I would consider dc to be safer than a corresponding
>ac voltage. The body responds particularly badly to frequencies in
>the 50-60Hz range. Funny that this has become the world standard.
>At the 300 or so volts that he will need, both are pretty lethal.
But as I've just pointed out, you only need about 160VDC, which is still a
voltage level that demands a degree of respect, but is certainly safer than
300VDC.
>The devices are being "marketed" here as FET-bridges. Each bridge
>will require a timing and control circuit. The outputs will beat
>against each other, producing voltages between two outlets which may
>be unexpected and may cause trouble with some older audio equipment,
>which connects signal and chassis ground to one side of the line.
There is the potential for that but it can be solved in several ways:
1. drive all the outlets in each room from a single FET bridge so that they
are all synchronised. That means that unless someone is silly enough to
run an extension lead from an adjoining room just to power their tuner, CD
or whatever while the rest of the gear is plugged into a different bridge
then there should be no problems.
2. use a central clock circuit which ensures that all the FET bridges are
synchronised. I believe that this would be unnecessary overkill and erode
the benefits of multiple redundancy that the FET bridge concept offers.
>An earlier poster mentioned that the devices should have a safe
>failure mode. Whenever I have had a similar idea I have been
>reminded by the people I work with that "H-bridges go BOOM in the night".
Indeed they can although failure mode is more frequently upon the
application or removal of load than when in a quiescent state. The most
probably cause of failure is driving a highly inductive load that generates
large voltage spikes -- but the output filtering (between the bridge and
the outlet) provides significant protection against such spikes.
Open-circuit FET failure can be easilly detected and used to shut-down the
drive circuitry (thus avoiding a DC or half-voltage output condition). A
double-short is best handled by an over-current circuit (which can be as
simple as a fuse since the FETs are already destroyed at that point).
>A central inverter, on the other hand, can provide transformer
>isolation between the dc and ac components. Failures should
>damage only the inverter, rather than suddenly applying dc to
>your various induction motors around the house. Any audible whine
>from the inverter can be locked in the garage, rather than being
>allowed to come from a box mounted behind each outlet. RF noise
>can be filtered at the inverter and never distributed through the
>house. (wishfull thinking)
As I pointed out, it's very easy to detect DC or half-voltage conditions
resulting from a failure in the bridge and automatically shut down all
output (thus protecting valuable appliances).
Regarding RFI, the FET bridge should produce even less RFI than an inverter
for the following reasons:
1. They are physically small units (no transformers and a small component
count using much lower-current semiconductors) which allows them to be
mounted in cases that provide a high level of RF shielding.
2. Simple filtering of the >5Khz switching curents (and harmonics) can
tends to reduce the RF radiated by the DC and AC wiring to very low levels.
>Be careful.
Ain't that the truth :-)
BTW: are their any enterprising manufacturers out there who might be
interested in turning a working prototype into a commercial item?
Hugh Piggott
>dun...@shack.punk.net wrote:
>
>>To get 230v(rms) from these units he'd need more than 240vdc input,
>>assuming he is aiming for a sine wave output.
>
>No, you only need 180VDC to get 230V (RMS) AC through the FET bridge.
>
>"Gasp" .. I hear you say, "how can that be?"
>
>Simple.. the peak to peak voltage of 230VAC is around 325. It varies from
>a maximum positive amplitude of +162V to a maximum negative amplitude of
>-162V.
I can't let this pass unremarked. A 230V ac supply has an rms voltage of 230 volts, from the line to the neutral. The peak rises t=
o 325V positive, and falls to 325 V negative. to achieve a full sine wave youwould need to have access to 325 volts. Not peak to p=
eak, but line to neutral. This voltage needs to be reversed every half cycle.
The FET bridge works by effectively reversing the polarity of the
>batteries across the output for each half cycle so the voltage goes from
>+162V down to zero (then the battery is effectively reversed) and up to
>-162V. The result is a peak to peak voltage of 325VAC and since it's a
>sine-wave, the RMS voltage is around 230V.
Sorry, but I just can't accept this. surely there must be others out there who agree with me on this!
>
>For 110VAC (RMS), you only need a DC voltage of around 85VDC so seven 12V
>batteries in series will do the job.
>
>One could argue therefore that since the DC voltage is less than 70% of the
>peak AC voltage that this system is safer than a mains-only setup.
>
>(before I get flamed, please note that I don't have a calculator handy so
>these figures are "rough" but certainly in the ballpark.)
I can't help flaming a bit. your calculations are OK, but the theory is all wrong. Peak voltages are certainly not calculted peak-=
to-peak.
>
>>I wouldn't worry about switching -- just avoid capacitors on the
>>input to the small units, or the contact arc will wipe out the
>>switches.
>
>You can perform "zero current" switching through the FET bridges anyway
>which totally eliminates the arc.
>
There is no 'zero current' when you are switching off a DC supply. You just have to break in and stop the current.
Hugh Piggott
>dun...@shack.punk.net wrote:
>
>>To get 230v(rms) from these units he'd need more than 240vdc input,
>>assuming he is aiming for a sine wave output.
>
>No, you only need 180VDC to get 230V (RMS) AC through the FET bridge.
>
>"Gasp" .. I hear you say, "how can that be?"
>
>Simple.. the peak to peak voltage of 230VAC is around 325. It varies from
>a maximum positive amplitude of +162V to a maximum negative amplitude of
>-162V.
I can't let this pass unremarked. A 230V ac supply has an rms voltage of 230 volts, from the line to the neutral. The peak rises t=
o 325V positive, and falls to 325 V negative. to achieve a full sine wave youwould need to have access to 325 volts. Not peak to p=
eak, but line to neutral. This voltage needs to be reversed every half cycle.
The FET bridge works by effectively reversing the polarity of the
>batteries across the output for each half cycle so the voltage goes from
>+162V down to zero (then the battery is effectively reversed) and up to
>-162V. The result is a peak to peak voltage of 325VAC and since it's a
>sine-wave, the RMS voltage is around 230V.
Sorry, but I just can't accept this. surely there must be others out there who agree with me on this!
>
>For 110VAC (RMS), you only need a DC voltage of around 85VDC so seven 12V
>batteries in series will do the job.
>
>One could argue therefore that since the DC voltage is less than 70% of the
>peak AC voltage that this system is safer than a mains-only setup.
>
>(before I get flamed, please note that I don't have a calculator handy so
>these figures are "rough" but certainly in the ballpark.)
I can't help flaming a bit. your calculations are OK, but the theory is all wrong. Peak voltages are certainly not calculted peak-=
to-peak.
>
>>I wouldn't worry about switching -- just avoid capacitors on the
>>input to the small units, or the contact arc will wipe out the
>>switches.
>
>You can perform "zero current" switching through the FET bridges anyway
>which totally eliminates the arc.
>
There is no 'zero current' when you are switching off a DC supply. You just have to break in and stop the current.
Hugh Piggott
>dun...@shack.punk.net wrote:
>
>>To get 230v(rms) from these units he'd need more than 240vdc input,
>>assuming he is aiming for a sine wave output.
>
>No, you only need 180VDC to get 230V (RMS) AC through the FET bridge.
>
>"Gasp" .. I hear you say, "how can that be?"
>
>Simple.. the peak to peak voltage of 230VAC is around 325. It varies from
>a maximum positive amplitude of +162V to a maximum negative amplitude of
>-162V.
I can't let this pass unremarked. A 230V ac supply has an rms voltage of 230 volts, from the line to the neutral. The peak rises t=
o 325V positive, and falls to 325 V negative. to achieve a full sine wave youwould need to have access to 325 volts. Not peak to p=
eak, but line to neutral. This voltage needs to be reversed every half cycle.
The FET bridge works by effectively reversing the polarity of the
>batteries across the output for each half cycle so the voltage goes from
>+162V down to zero (then the battery is effectively reversed) and up to
>-162V. The result is a peak to peak voltage of 325VAC and since it's a
>sine-wave, the RMS voltage is around 230V.
Sorry, but I just can't accept this. surely there must be others out there who agree with me on this!
>
>For 110VAC (RMS), you only need a DC voltage of around 85VDC so seven 12V
>batteries in series will do the job.
>
>One could argue therefore that since the DC voltage is less than 70% of the
>peak AC voltage that this system is safer than a mains-only setup.
>
>(before I get flamed, please note that I don't have a calculator handy so
>these figures are "rough" but certainly in the ballpark.)
I can't help flaming a bit. your calculations are OK, but the theory is all wrong. Peak voltages are certainly not calculted peak-=
to-peak.
>
>>I wouldn't worry about switching -- just avoid capacitors on the
>>input to the small units, or the contact arc will wipe out the
>>switches.
>
>You can perform "zero current" switching through the FET bridges anyway
>which totally eliminates the arc.
>
There is no 'zero current' when you are switching off a DC supply. You just have to break in and stop the current.
Hugh Piggott
>Hugh Piggott <hugh.p...@enterprise.net> wrote:
>
>>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>>Gasp! Nobody using solar, wind or other low-output electrical generation
>>>should be using a water heater unless it's a heat-pump. Heating water with
>>>valuable electricity is such a waste!
>
>> Waste is all relative in this context. As a windpower user, I dump a lot of
>> power as heat, but it is not every day. If I built a heatpump, this would not
>> be worth while unless it was in frequent, if not constant use. I would do
>> better building another windmill than a heatpump. Then I would have more
>> power during periods of light winds. Or maybe some photovoltaics. But
>> when there is loads of wind, I just enjoy basking in the dump load heat,
>> and soaking up nature's abundance of power...
>
>But why consider "dumping" energy it when you can store it and use it later
>or at least offset your other energy demands by pre-heating your hot water
>supply.
Obviously I do heat my water with it if necessary, or I heat a room if it is cold. a heat pump or a bigger storage system would be =
better ideally, but it all comes down to economics. Beyond a certain point it is beter value to capture more energy from the free s=
ource than to obsessively conserve every morsel of what you caught before.
>
>>Sounds a bit ambitious, but that's not really a criticism.
>>240 volts Dc is said to be pretty lethatl though..
>
>50VDC can be lethal if you don't treat it with respect. Here in New
>Zealand we have a mains voltage of 230VAC which is 320V peak-to-peak yet
>very few people are electrocuted each year. In fact I suspect that if
>you're energy self-sufficient, you're in more danger of dying in a
>house-fire caused by high-resistance terminations overheating at the high
>currents demanded at 12V/24VDC than you are from electrocution.
>
It isn't peak to peak, its line to neatral peak, whaich is quite different.
DC is very heavy on the switch contacts. It is also harder to break a Dc arc than an Ac one, so the shock hazard is higher. You ca=
n design around all of these, but they are problems to be faced.
A 325 volt battery makes sense to me, but I would cntralise the inverter.
John A. Stanley
br...@faxmail.co.nz (Bruce Simpson) wrote:
>Hugh Piggott <hugh.p...@enterprise.net> wrote:
>
>>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>>Gasp! Nobody using solar, wind or other low-output electrical generation
>>>should be using a water heater unless it's a heat-pump. Heating water with
>>>valuable electricity is such a waste!
>
>> Waste is all relative in this context. As a windpower user, I dump a lot of
>> power as heat, but it is not every day. If I built a heatpump, this would not
>> be worth while unless it was in frequent, if not constant use. I would do
>> better building another windmill than a heatpump. Then I would have more
>> power during periods of light winds. Or maybe some photovoltaics. But
>> when there is loads of wind, I just enjoy basking in the dump load heat,
>> and soaking up nature's abundance of power...
>
>But why consider "dumping" energy it when you can store it and use it later
My experience with wind energy is that on average I get one
third to one half of the maximum out of it, as it is not every
day that there is a 25MPH wind blowing. Alternative energy
systems will always see days where there is far more energy
coming in than can be stored unless one goes to the expense of
buying an obscenely expensive energy storage system.
>or at least offset your other energy demands by pre-heating your hot water
>supply.
And space heating isn't one of the other energy demands? Heck,
if it's cold, windy, sunny, and the battery's full I switch on a
1500 watt electric heater!
John A. Stanley
>RF noise can be filtered at the inverter and never distributed
>through the house. (wishfull thinking)
Which inverter are you using? I run an AM radio powered by a
wall-wart running on juice from a Trace SW 4024 and there's no
inverter RF that I can detect. For me, your wishfull thinking is
solid reality.
Duncan Campbell
>>>I wouldn't worry about switching -- just avoid capacitors on the
>>>input to the small units, or the contact arc will wipe out the
>>>switches.
>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>You can perform "zero current" switching through the FET bridges anyway
>>which totally eliminates the arc.
Hugh Piggott <hugh.p...@enterprise.net> wrote:
>There is no 'zero current' when you are switching off a DC supply. You
>just have to break in and stop the current.
I believe that where I was thinking of a small inverter powered
from a 325 volt DC source switched through a standard light
switch -- much like standard house wiring -- Bruce was thinking
of the output from the switch as a digital input to the control
circuitry of that small inverter, instructing the bridge to shut
down after the present half-cycle is complete or instructing it
to start the load by coming out of the "off" state and initializing
a new output cycle.
Switched outlets for flourescent lighting, garbage disposals, etc...
The second method requires an input to the inverter that is always
hot, and may require one to pull new wires. Because it allows one
to quit worrying about damage to the switch when it is called upon
to charge a set of filter capacitors this method will allow one
to address the problem of RF radiation from the DC input wires with
standard filtration techniques.
Yet another thought comes to mind with this scheme. Some people
really like the output of incandescent / halogen light bulbs. The
voltages available in his house would be 325vdc and 240vrms. It
seems silly to run an incandescent load off an inverter, but the
available DC is much too strong for the bulbs available readily.
He could tap off his battery bank at 240vDC, but he will then risk
overcharging the batteries excluded from that tap. He could change
all his fixtures to take _three_ US-style 110 volt bulbs in series.
He could live with the irony of wasting a bit of power at the inverter
in order to feed a pretty sine wave to a load that won't appreciate it.
I'm glad to see people are thinking along the same lines I follow...
and outside the context of work, too. Look, ma, I'm helping, I'm helping.
:-)
Duncan.
Bruce Simpson
>In article <4abfiq$9...@news.express.co.nz>,
>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>A "good" wind generator will have a variable pitch propeller, the pitch of
>>which can be either mechanically controlled by a centrifugal mechanism or
>>(in the case of a more sophisticated unit) by a microprocessor and small
>>servomechanisms.
>I'm not aware of any small wind generator with those features.
>All the wind generators for individual AE systems I've seen use
>fixed blades with a mechanism to tilt the unit back in high wind.
Actually the first thing that struck me when I initially looked at
alternative energy solutions was how lacking the whole area was in the use
of modern technology. Inverters, fixed-pitch blades, low primary voltages,
no use of intelligent control systems (when microprocessors are under a
dollar each in quantity).
Perhaps it's just that those in the industry are "back to basics" types or
maybe it's just that the market will buy whatever's offered - I don't know.
It strikes me that since it's really quite difficult to reliably and
economically extract reasonable amounts of power from the environment (ie:
levels that don't require you to alter your whole life-style to accomodate
them), building systems which make optimal use of such resources is a
pretty good idea.
My ultimate (but as yet unrealisable) goal is to create an alternative
energy solution which allows an "average" house to be disconnected from the
grid and switched to environment-only energy without the need to buy new
appliances, add a gas-fired hot-water cylinder, worry inceasantly about the
dangers of accidentally leaving a light on overnight etc.
We already have the technology to achieve much of this. Well designed wind
and PV generation systems can provide reasonable levels of energy in a wide
range of conditions. Energy storage is a bit of a problem at present with
conventional choices such as lead-acid batteries only having a relatively
low life and limited capacities while newer technologies such as ammonia
dissociation/recombination and fuel cells show promise but have yet to
become a viable option.
Until we get *major* breakthroughs in energy generation and/or storage, the
best solution is to work at improving efficiencies. Saving 15W per hour by
using a 200V wind-generator instead of a 12V one is one example.
PIR-controlled room-lighting can also trim a lot of watt/hours off of
consumption as can variable-pitch wind-generators, heat-pumps and the like.
My own calculations indicate that by taking advantage of these efficiencies
I'll be able to disconnect from the grid with the only concessions being
the addition of solar water-heating pannels and a gas-oven instead of
electrical (we already use the microwave more than most) and a standby
generator "just in case".
I've never examined the size of the alternative-energy market but it must
be very small or I'm sure that someone would have already produced better
solutions than those which most people seem to be using at present.
Ted Dunning
> The FET bridge works by effectively reversing the polarity of the
>batteries across the output for each half cycle so the voltage goes from
>+162V down to zero (then the battery is effectively reversed) and up to
>-162V. The result is a peak to peak voltage of 325VAC and since it's a
>sine-wave, the RMS voltage is around 230V.
Sorry, but I just can't accept this. surely there must be others
out there who agree with me on this!
you are dead on, hugh. 120 V service is 120Vrms line to neutral which
is 170V peak to neutral and 340 V p-p.
another handy number to have is that three 120V phases are 208V line
to line.
Eric Riehl
[ stuff deleted ]
My ultimate (but as yet unrealisable) goal is to create an alternative
energy solution which allows an "average" house to be disconnected from the
grid and switched to environment-only energy without the need to buy new
appliances, add a gas-fired hot-water cylinder, worry inceasantly about the
dangers of accidentally leaving a light on overnight etc.
We already have the technology to achieve much of this. Well designed wind
and PV generation systems can provide reasonable levels of energy in a wide
range of conditions. Energy storage is a bit of a problem at present with
conventional choices such as lead-acid batteries only having a relatively
low life and limited capacities while newer technologies such as ammonia
dissociation/recombination and fuel cells show promise but have yet to
become a viable option.
Until we get *major* breakthroughs in energy generation and/or storage, the
best solution is to work at improving efficiencies. Saving 15W per hour by
using a 200V wind-generator instead of a 12V one is one example.
PIR-controlled room-lighting can also trim a lot of watt/hours off of
consumption as can variable-pitch wind-generators, heat-pumps and the like.
[ more stuff deleted ]
A year ago when I started reading about the current state of solar power
technology, I was supprised by both how much more expensive it is to run
a solor powered home, and also by how much more efficient some of the
appliances in a solar powered home are in comparison to conventional
appliances. However, things such as a sunfrost, compact florescent
lights aren't showing up in people's homes, who done't read this
newsgroup or similar material.
IMHO, there's a lot that can be done now to change the infrastructure
in the United States and probably other countries so that the total
power requirements can be met by environmental friendly methods.
Also, It's inconceivable to me that in my lifetime people who live in
high density, low wage housing will be using anything by grid
electricity in my lifetime.
For example, the "politically correct" company I was working for part
time had an extra old laser printer, so they used it as an extra
printer with paper which was already printed on one side. But when I
pointed out just how much electricty it took to keep the printer on,
they turned if off, and started trying to use duplex mode more often
on the main laser printer.
A lot of people still use incadescent lights because that just don't
realize how good compact florescent lights have become. I holiday
gift of a compact florescent light to people like this, with the
suggestion to place it in a fixture that they tend to keep on for long
periods of time would help change the climate towards florescent home
lighting.
When I was out in-line skating at a middle school track the other day,
I thought how perfect a school would be for a solar powered hot water
system, since the hot water would be used after gym classes, and
after-school sports, the times that would minimize the need for
storage. A well written proposal to a school board might get a system
installed for the school, and at least would get people who read the
proposal to realize how short the pay-back period is on modern
systems. They might install one for their own home.
This said, I think it's great some people are going completed
self-sufficient witheat some people are going completed
self-sufficient with their electrical needs despite the extra cost,
and hope to do that same thing myself some day after I finish graduate
school.
Eric Riehl
eri...@or.unc.edu
Harry H Conover
: you are dead on, hugh. 120 V service is 120Vrms line to neutral which
: is 170V peak to neutral and 340 V p-p.
You're 100% correct. Let me put it into slightly more verbose
wording:
A conventional 240V household service consists of two 120V lines
that are 180-degrees out of phase, and a neutral. RMS line-to-line
voltage is 240, and each line to neutral is 120. Peak-to-peak
voltages are 170 and 340, respectively.
: another handy number to have is that three 120V phases are 208V line
: to line.
But only if you have a delta wiring configuration. In a 3-phase 'Y'
configuration, line to line voltage is 240, and line to common is
120.
Harry C.
Hugh Piggott
a degree of pitch control which allows them to run free without
overspeed. But it is usually more useful to keep them loaded anyway.
Apart from being quieter, the power can be used for water heating or
somethng, even if the main storage battery is full.
br...@faxmail.co.nz (Bruce Simpson) wrote:
>
>Actually the first thing that struck me when I initially looked at
>alternative energy solutions was how lacking the whole area was in the use
>of modern technology. Inverters, fixed-pitch blades, low primary voltages,
>no use of intelligent control systems (when microprocessors are under a
>dollar each in quantity).
>
There is a lovely quote which Bergey Windpower use, and I don't know if I remember it right, but it goes something like as follows:
"Perfection is achieved, not when nothing more can be added, but when nothing more can be taken away."
Those who know wind turbines, will appreciate that anything which can go wrong, will go wrong. There is something about wind machin=
es that makes reliability a big issue.. Maybe it's the long running hours, the variable conditions, the attempt to keep costs down, =
the fact that they are poorly understood, but for whatever reason, they need to be kept simple if possible.
>Perhaps it's just that those in the industry are "back to basics" types or
>maybe it's just that the market will buy whatever's offered - I don't know.
>It strikes me that since it's really quite difficult to reliably and
>economically extract reasonable amounts of power from the environment (ie:
>levels that don't require you to alter your whole life-style to accomodate
>them), building systems which make optimal use of such resources is a
>pretty good idea.
>
It's a balance between whether it's worth squeezing out an extra 5% of efficiency, or just building a slightly larger machine. The =
wind is free.
>
>Until we get *major* breakthroughs in energy generation and/or storage, the
>best solution is to work at improving efficiencies. Saving 15W per hour
Excuse me but there is no such thing as a "watt per hour".
>
>I've never examined the size of the alternative-energy market but it must
>be very small or I'm sure that someone would have already produced better
>solutions than those which most people seem to be using at present.
It's true that what is available now is disappointing, but I suggest you try to do better before you rubbish everyone else's efforts=
Bruce Simpson
>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>dun...@shack.punk.net wrote:
>>
>>>To get 230v(rms) from these units he'd need more than 240vdc input,
>>>assuming he is aiming for a sine wave output.
>>
>>No, you only need 180VDC to get 230V (RMS) AC through the FET bridge.
>>
>>"Gasp" .. I hear you say, "how can that be?"
>>
>>Simple.. the peak to peak voltage of 230VAC is around 325. It varies from
>>a maximum positive amplitude of +162V to a maximum negative amplitude of
>>-162V.
>I can't let this pass unremarked. A 230V ac supply has an rms voltage
>of 230 volts, from the line to the neutral. The peak rises to 325V positive,
>and falls to 325 V negative. to achieve a full sine wave youwould need to have
>access to 325 volts. Not peak to peak, but line to neutral. This voltage
>needs to be reversed every half cycle.
You are of course quite correct -- I don't know where my brain was at..
that will serve me right for posting after 20hrs without sleep :-(
> The FET bridge works by effectively reversing the polarity of the
>>batteries across the output for each half cycle so the voltage goes from
>>+162V down to zero (then the battery is effectively reversed) and up to
>>-162V. The result is a peak to peak voltage of 325VAC and since it's a
>>sine-wave, the RMS voltage is around 230V.
>Sorry, but I just can't accept this. surely there must be others out there
>who agree with me on this!
Yes, me for one (having had the prerequisite 8hrs of "zzz".
>I can't help flaming a bit. your calculations are OK, but the theory is all wrong.
>Peak voltages are certainly not calculted peak-to-peak.
Flames accepted.
>>>I wouldn't worry about switching -- just avoid capacitors on the
>>>input to the small units, or the contact arc will wipe out the
>>>switches.
>>
>>You can perform "zero current" switching through the FET bridges anyway
>>which totally eliminates the arc.
>There is no 'zero current' when you are switching off a DC supply.
>You just have to break in and stop the current.
By zero current I mean that you can electronically turn off all the
bridges, effectively reducing the current draw on the DC side to zero
before any mechanical switching is performed.
I've just hauled out my schematics and earlier workings (which I haven't
looked at for about 6 months) and original calculations are correct (ie: I
was posting in my sleep last night :-)
I could say "I was just testing to see if everyone was paying attention"
but I'll come clean and admit that I screwed up in that posting.
Bruce Simpson
>br...@faxmail.co.nz (Bruce Simpson) wrote:
>>Until we get *major* breakthroughs in energy generation and/or storage, the
>>best solution is to work at improving efficiencies. Saving 15W per hour
>Excuse me but there is no such thing as a "watt per hour".
Tell the utility company that presently charges me by the Kilowatt/hour :-(
>>I've never examined the size of the alternative-energy market but it must
>>be very small or I'm sure that someone would have already produced better
>>solutions than those which most people seem to be using at present.
>It's true that what is available now is disappointing, but I suggest you try to
>do better before you rubbish everyone else's efforts
I was not intending to cast aspersions on the efforts of others, just point
out that it's an area which seems slow to adopt the benefits of modern
technology. I'm sure that there are lots of people using existing
technology to great effect it just seems a shame that for the sake of an
extra few $ worth of manufacturing and amortized development costs, many
aspects of environmental energy generation and utilization could be
improved by a worthwhile amount.
Bruce Simpson
>I'd venture that it's just a simple case of economics and KISS.
>Why create a more elaborate and more expensive system when a
>simpler and less expensive one can accomplish the same task?
>Personally, I'm perfectly happy that there isn't pitch control
>on my windmill; it's one less thing to break.
Fair enough, that's a valid perspective. Mind you, I suggest that there
are a lot of people who would prefer something a little efficient albeit at
the cost of some complexity. I draw the comparison between the Model-T
ford or a LADA (KISS) and a Honda Prelude VTEC (very complex but remarkably
efficient). Horses for courses I guess.
I can see that there would be quite a degree of comfort to knowing that the
only way your windmill will stop is if the vanes are physically wrenched
from the hub. Even a simple centrifugal pitch control has an extra handful
of parts and where there are moving parts there's always a need for regular
mantenance (which can be awkward if your generator's up a 40' pole).
Ross Alexander
Well, that would depend strongly on how demanding one was about
eliminating RF noise. Listening to commercial AM broadcasts isn't
exactly in the same league as trying to dig out Europeans from the
natural noise on the 160m ham band.
I run a rather old (but utterly reliable) Trace 2024, which is nowhere
as clean as a 4024. I've been forced to take extensive measures to
get the output (and input - a 2024 puts noise back down the input
leads) power cleaned up enough to not wipe out 160m. As for getting
things clean enough to allow me to DX the aircraft beacons between 200
and 450 KHz, I've given up on that and instead shut down to direct
battery power.
On the other hand, since the inverter is centralized, I've only had
to build one set of filters and one RF-tight enclosure. I'd hate
to have to build one per room.
regards,
Ross ve6pdq
--
Ross Alexander, ve6pdq -- (403) 675 6311 -- r...@cs.athabascau.ca
Ross Alexander
>[...] Those who know wind turbines, will appreciate that anything
>which can go wrong, will go wrong. There is something about wind
>machines that makes reliability a big issue.. Maybe it's the long
>running hours, the variable conditions, the attempt to keep costs
>down, the fact that they are poorly understood, but for whatever
>reason, they need to be kept simple if possible.
Perhaps it's the fact that they're installed at the tops of towers in
windy places which tends to discourage routine maintenance ;). I
myself have no fear of heights, which makes me very popular among my
height-aversive ham friends - I'm always the guest of honour at
antenna installation parties. But it can certainly get cold at the
top of a tower in a subzero wind. Tools get dropped, fingers get
stiff, one becomes eager to finish and get down.
BTW, I've been thinking about erecting a 40m tower and mounting a
windcharger on it. My power requirements are modest; 5 to 8 kWh a
week. Two questions:
(A) do any of the popular small windchargers present their output as
single or three-phase AC (the run back to the house must needs be
long)?
(B) since I plan to run 500 watts of 160m RF into the tower, are
there any rigs which are particularly RF tolerant? (There's no
requirement to run the charger and transmit simultaneously; I'm quite
prepared to walk out the base of the tower and manually disconnect.)
For example, I'd be quite sceptical about the long-term reliability of
a microprocessor controlled unit under this regime.
Hugh Piggott
Sorry for my rather hot remarks.
There is a difference between kwhours and kw/hour
Kilowatt hours is a measure of energy= number of kilowatts times the
number of hours.
Kilowatts per hour measures the rate of increase in power consumption.
A frightening concept.
Ted Dunning
In article <4afj0q$o...@sundog.tiac.net> con...@max.tiac.net (Harry H Conover) writes:
: another handy number to have is that three 120V phases are 208V line
: to line.
But only if you have a delta wiring configuration. In a 3-phase 'Y'
configuration, line to line voltage is 240, and line to common is
120.
this is incorrect.
the only difference between Y and delta wiring is whether or not you
actually use the neutral line.
in a three phase system with line to neutral voltage of 120, then the
line to line voltage will be 208. if it were 240, then the phase
difference would have to be 180 degrees, and it just isn't far enough
around the circle to get three phases 180 degrees from each other.
you can also have things wired so that line to line is 240 and the
line to neutral is about 138. this is essentially never done in
residential settings.
Hugh Piggott
>Hugh Piggott <hugh.p...@enterprise.net> writes:
>
>>[...] Those who know wind turbines, will appreciate that anything
>>which can go wrong, will go wrong. There is something about wind
>>machines that makes reliability a big issue.. Maybe it's the long
>>running hours, the variable conditions, the attempt to keep costs
>>down, the fact that they are poorly understood, but for whatever
>>reason, they need to be kept simple if possible.
>
>Perhaps it's the fact that they're installed at the tops of towers in
>windy places which tends to discourage routine maintenance ;). I
>myself have no fear of heights, which makes me very popular among my
>height-aversive ham friends - I'm always the guest of honour at
>antenna installation parties. But it can certainly get cold at the
>top of a tower in a subzero wind. Tools get dropped, fingers get
>stiff, one becomes eager to finish and get down.
>
>BTW, I've been thinking about erecting a 40m tower and mounting a
>windcharger on it. My power requirements are modest; 5 to 8 kWh a
>week. Two questions:
>
>(A) do any of the popular small windchargers present their output as
>single or three-phase AC (the run back to the house must needs be
>long)?
They nearly all produce AC from alternators, then rectify it.
Contrary to popular belief, DC is better than AC for keeping
transmission losses down. Unless you tranform the voltage up of course.
The Marlec FM1800 (a British machine with 250 watt output) has high
voltage transmission as standard, and end use outputs of 12,24, 36 or 48
volts.
>
>(B) since I plan to run 500 watts of 160m RF into the tower, are
>there any rigs which are particularly RF tolerant? (There's no
>requirement to run the charger and transmit simultaneously; I'm quite
>prepared to walk out the base of the tower and manually disconnect.)
You would usually need to keep a load on the windmill, or it will run
wild. Switching to a resistive dump load or even shorting it out would
often do OK.
Russell B. Huss
>From: br...@faxmail.co.nz (Bruce Simpson)
>Subject: Re: TRACE SW 4024 - True Sine Wave Inverter
>Date: Sat, 09 Dec 1995 08:55:57 GMT
>ro...@hal.COM (Robert Kleinschmidt) wrote:
(snip)
>>2) To my mind, high voltage DC seems to have some safety
>> issues. (Arcing, need for special switches, etc.). I
>> have also heard it asserted that AC "lets go of you"
>> more readily than DC, should you complete a circuit
>> with your hands.
>There is nowhere that you need to switch the DC voltages (using a switch or
>relay) regularly. The Fet bridge uses (passes) no current when there's no
>load and so the switch can be on the AC side. In this situation it's
>exactly the same as switching normal 110VAC. As regards to the safety
>issues of DC vs AC, I've read nothing to substantiate the "belief" that AC
>"lets go of you". If you examine the situation: at 50Hz there are two
>zero-voltage points and two periods when the voltage is "safe", say 50V or
>less. Assuming that these two "safe" periods account for 50% of the cycle
>we then find that the length of such a "safe" period is around 20mS. I
>don't think that your muscles could relax quickly enough to let go of
>anything "live" that you might have grabbed ahold of.
(snip)
When Thomas Edison set out to electrify the world, he chose to go with DC.
Nikola Tesla developed the necessary "stuff" for the use of AC, put into use
with the help of George Westinghouse. The fight was a long, dirty one over
which would be used, but Tesla's system won, since it was the better system.
Many of the roumors (lies) that Edison used are still with us.
Ref: Tesla, Man Out of Time by Margaret Cheney, 1981 (a good book).
russ
Robert Kleinschmidt
Bruce Simpson <br...@faxmail.co.nz> wrote:
>
>>The devices are being "marketed" here as FET-bridges. Each bridge
>>will require a timing and control circuit. The outputs will beat
>>against each other, producing voltages between two outlets which may
>>be unexpected and may cause trouble with some older audio equipment,
>>which connects signal and chassis ground to one side of the line.
>
>There is the potential for that but it can be solved in several ways:
>
>1. drive all the outlets in each room from a single FET bridge so that they
>are all synchronised. That means that unless someone is silly enough to
>run an extension lead from an adjoining room just to power their tuner, CD
>or whatever while the rest of the gear is plugged into a different bridge
>then there should be no problems.
>
>2. use a central clock circuit which ensures that all the FET bridges are
>synchronised. I believe that this would be unnecessary overkill and erode
>the benefits of multiple redundancy that the FET bridge concept offers.
I seem to recall a reasonably good electronics engineer getting himself
into trouble by connecting things to multiple circuits. For consumers,
synchronized A.C. sounds extremely desirable. It is also entirely possible
to have multiple circuits in a single room. Seems as though a central
clock (with spare) would be prudent.
A couple of further questions/comments (Hope you are not completely
sick of the thread by now).
1) What would you estimate the cost per rated watt to be ?
(Trace I think runs about $.60 U.S. per rated watt)
In our house, we have ~8 separate circuits, ~50 wall
outlets plus ~15 switched light fixtures. To provide
the kind of distributed layout you speak of would
seem to require a very low parts cost.
2) How much variation in DC voltage would an FET device
tolerate and still give reliable A.C. voltage ? For
a normal lead-acid battery bank, I think you would
have to figure ~20% variation between peak charge
and max-discharge voltages. Could this be handled
by the FET devices without other regulation ?
3) It seems as if you are trading redundancy and flexibility
at one part of the system for inflexibility and single
points of failure in others. I had asked about P.V.
layouts, and you have reasonably pointed out that
P.V. current could be stepped up to higher DC voltages,
but this still leaves a single point of failure in the
charger, plus the battery bank itself. (More difficult
to configure a high voltage battery bank or to reconfigure
if one or more batteries go bad plus size/weight issues
for boats, motorhomes etc.).
4) Why a per-outlet or even per-circuit solution ? In ~7 years,
our Trace 2024 unit went into one graceful shutdown due to
overload, but I have yet to hear of a real failure. Can
anyone on the net cite a failure story ? Seems to me as if
even a pair of devices would provide more than adequate
redundancy. (I grant you that the ability to build a system
incrementally and possible cost reductions are highly desirable.)
5) Seems as though you also loose a "pass-through" mode for grid
or genset power. Certainly not a show stopper, but forces
you to size the FET devices and battery charger for worst
case motor loads. (See also DC failure described below).
For whatever it's worth, we did in fact also experience one low
voltage wiring failure. (Original installer used an alloy connector
on 24 V. battery line.). Damage was contained by metal switch box
and limited to the wire and connector, although I went to a new
and larger switch box as well. I was able to effect temporary and
later permanent repairs with the battery disconnected, running the
house from the genset. (Good argument for "pass-through" mode ?)
Regards, and thanks again for some interesting ideas.
Rob Kleinschmidt
Ross Alexander
>4) Why a per-outlet or even per-circuit solution ? In ~7 years,
> our Trace 2024 unit went into one graceful shutdown due to
> overload, but I have yet to hear of a real failure. Can
> anyone on the net cite a failure story ? [...]
A tomcat who lives with me once decided that my 2024 would make a good
territorial marker. Ie, he urinated on it repeatedly. Eventually it
shut down (quite gracefully) and it required a full teardown and
distilled-water wash job to get it running again. Since just about
everything was either potted or made from stainless steel, there was
no permanent damage.
regards,
Ross
Bruce Simpson
>A couple of further questions/comments (Hope you are not completely
>sick of the thread by now).
>1) What would you estimate the cost per rated watt to be ?
> (Trace I think runs about $.60 U.S. per rated watt)
> In our house, we have ~8 separate circuits, ~50 wall
> outlets plus ~15 switched light fixtures. To provide
> the kind of distributed layout you speak of would
> seem to require a very low parts cost.
I haven't priced the components recently .. I'll do that and post a figure.
>2) How much variation in DC voltage would an FET device
> tolerate and still give reliable A.C. voltage ? For
> a normal lead-acid battery bank, I think you would
> have to figure ~20% variation between peak charge
> and max-discharge voltages. Could this be handled
> by the FET devices without other regulation ?
The FET bridge also acts as a regulator to ensure that the output is held
to the nominal voltage through as wide a range of input voltages as
possible. The degree of regulation is limited by the excess of DC voltage
over the required p-p AC voltage. It is wise to have enough headroom in
your DC supply to allow for the nominal p-p voltage plus switching losses
when the batteries are at their lowest allowable discharge level. The fact
that this may see a headroom (excess DC voltage) of 50 volts or more when
the system is fully charged is irrelevant since the bridge is a
high-efficiency switching regulator that will reduce the voltage to the
nominal AC voltage in a very efficient manner.
>3) It seems as if you are trading redundancy and flexibility
> at one part of the system for inflexibility and single
> points of failure in others. I had asked about P.V.
> layouts, and you have reasonably pointed out that
> P.V. current could be stepped up to higher DC voltages,
> but this still leaves a single point of failure in the
> charger, plus the battery bank itself. (More difficult
> to configure a high voltage battery bank or to reconfigure
> if one or more batteries go bad plus size/weight issues
> for boats, motorhomes etc.).
Your battery bank is an energy store. You get the same amount of energy
out of a 10Ah 240V battery as you do from a 100AH 24V battery and there's
often little difference in size or weight. The larger number of batteries
required to store 240V can be seen as a bonus insomuch as each battery is
cheaper and can be replaced independently of the others (avoiding the need
to throw out an expensive battery just because a single cell has failed).
This is along the lines of the more expensive stackable cells sometimes
used for deep-discharge storage solutions -- just cheaper :-)
>4) Why a per-outlet or even per-circuit solution ? In ~7 years,
> our Trace 2024 unit went into one graceful shutdown due to
> overload, but I have yet to hear of a real failure. Can
> anyone on the net cite a failure story ? Seems to me as if
> even a pair of devices would provide more than adequate
> redundancy. (I grant you that the ability to build a system
> incrementally and possible cost reductions are highly desirable.)
There's no reason why you couldn't use just one or two bridges to service
then entire house, it's only my preference to have one per room or outlet.
It lets me hang heavy loads off multiple outlets without having to worry
about the total loading and its effect on a central unit.
>5) Seems as though you also loose a "pass-through" mode for grid
> or genset power. Certainly not a show stopper, but forces
> you to size the FET devices and battery charger for worst
> case motor loads. (See also DC failure described below).
There's no reason why a pass-through mode couldn't be built into the bridge
unit(s), a simple relay to switch the bridge out of circuit and route the
AC straight to the outlet would suffice. Proabably just another $20 worth
of components.
>Regards, and thanks again for some interesting ideas.
If ideas were $ I'd be a millionaire :-)
Duncan Campbell
>ro...@hal.COM (Robert Kleinschmidt) wrote:
>
>> but this still leaves a single point of failure in the
>> charger, plus the battery bank itself. (More difficult
>> to configure a high voltage battery bank or to reconfigure
>> if one or more batteries go bad plus size/weight issues
>> for boats, motorhomes etc.).
>
>Your battery bank is an energy store. You get the same amount of energy
>out of a 10Ah 240V battery as you do from a 100AH 24V battery and there's
>often little difference in size or weight. The larger number of batteries
>required to store 240V can be seen as a bonus insomuch as each battery is
>cheaper and can be replaced independently of the others (avoiding the need
>to throw out an expensive battery just because a single cell has failed).
I have to disagree with this response. Robert asserts that
it is more difficult to set up or reconfigure a high voltage
battery bank than a low voltage one. There should be no doubt
that this is the case.
If your building block is the six volt Trojan golf cart battery
then the minimum 24 volt setup will require four batteries. The
minimum 240 volt setup will require fourty batteries, which would
cause trouble if loaded on a motorhome or boat. The fact that
the larger battery bank holds ten times the charge is sometimes of
little use.
Assuming you set up your banks to hold the same amount of stored
energy then a 24 volt and 240 volt battery bank might both
consist of fourty 6v batteries interconnected.
If one of the batteries in the 24 volt setup goes bad -- drains,
fails the specific gravity test, etc, one can just disconnect it,
drive to your local battery exchange, and hook the new one up when
you return. With a 10% drop in capacity your load sees no change.
If one of the batteries in your 240 volt setup goes bad you must
shut off ALL dc loads before considering breaking the circuit
with your bare hands. If you miss any loads then there will be,
between the battery post and the connector you have just removed,
the full potential of the battery bank.
This risk can be reduced if the bank capacity is doubled again, to
eighty batteries, as this would allow a parallel branch to maintain
the voltage across the load and reduce the voltage across an opening
in the other branch. The voltage across the gap may still be
significant, depending on how well balanced the two branches of
the battery bank are. Now, while you are making the purchase at
the battery exchange, your load is running on 50% of your prior
storage capacity.
Another consideration is the possibility for some part of the DC
circuit to become grounded, either deliberately or accidentally. This
would put some of the battery connections at a dangerous potential
with respect to the concrete floor below you, the water pipe behind
you, etc.
[Bruce, repeated]
>often little difference in size or weight. The larger number of batteries
>required to store 240V can be seen as a bonus insomuch as each battery is
>cheaper and can be replaced independently of the others (avoiding the need
>to throw out an expensive battery just because a single cell has failed).
The cost of the individual batteries in the above example is the
same whether they are wired up in a low or high voltage configuration.
There is no difference in size or weight between two banks configured
to store the same amount of energy (watt-hours). If a single cell
fails in either arrangement you still have to replace the battery
containing that cell. As you make that replacement you will have
to shut down your 240 volt equipment, due to the single series path
that would be broken. The bad battery in the 48/24/12 volt bank
could be replaced without shutting down the loads. Unexpected
voltages may spring up between a connector and battery post when
a single series path is broken. Unexpected voltages may cause trouble
if the operator is inadvertantly grounded while configuring or
maintaining his high voltage battery bank.
(sorry for rambling -- this was just bugging me on the way home from
work this morning. )
Duncan.
Thomas C. J. Sefranek
Can there possibly be two Russell B. Huss(s0 in the world?
The Russ I knew was at Ft. Devens and one of the finest!
Perhaps your him?
Tom
WA1RHP
Bruce Simpson
>If your building block is the six volt Trojan golf cart battery
>then the minimum 24 volt setup will require four batteries. The
>minimum 240 volt setup will require fourty batteries, which would
>cause trouble if loaded on a motorhome or boat. The fact that
>the larger battery bank holds ten times the charge is sometimes of
>little use.
You would only need 20 x 10AH batteries.
>Assuming you set up your banks to hold the same amount of stored
>energy then a 24 volt and 240 volt battery bank might both
>consist of fourty 6v batteries interconnected.
But why use 6V batteries?
>If one of the batteries in the 24 volt setup goes bad -- drains,
>fails the specific gravity test, etc, one can just disconnect it,
>drive to your local battery exchange, and hook the new one up when
>you return. With a 10% drop in capacity your load sees no change.
In this case you make sure that your battery pack has enough voltage
"headroom" to allow the bypassing of one or more of your batteries. This
way you can remove a battery from the circuit without the need to
disconnect the DC load and without affecting the AC output. This is done
by connecting a suitably rated diode across the output of each battery such
that if that battery is removed, the diode then services to bypass the
missing cell. If you've got the recommended 40V of headroom then you can
easily remove one or two of the 12V batteries without compromising the AC
output or reducing capacity by more than the rated capacity of the
individual batteries ... in this case 10AH. The inclusion of the diode
also ensures that even when there is an AC (and resulting DC) load on the
system when the battery is removed, no more than 0.7volts will appear
across the removed battery leads, thus reducing the potential for
electrical shock.
>If one of the batteries in your 240 volt setup goes bad you must
>shut off ALL dc loads before considering breaking the circuit
>with your bare hands. If you miss any loads then there will be,
>between the battery post and the connector you have just removed,
>the full potential of the battery bank.
Nope... as I just explained, if you have a diode permanently connected
across the leads then the maximum voltage will be 0.7V.
>This risk can be reduced if the bank capacity is doubled again, to
>eighty batteries, as this would allow a parallel branch to maintain
>the voltage across the load and reduce the voltage across an opening
>in the other branch. The voltage across the gap may still be
>significant, depending on how well balanced the two branches of
>the battery bank are. Now, while you are making the purchase at
>the battery exchange, your load is running on 50% of your prior
>storage capacity.
Nope, by allowing enough headroom and using the very simple diode technique
you loose no more than the capacity of the battery which has been removed.
>Another consideration is the possibility for some part of the DC
>circuit to become grounded, either deliberately or accidentally. This
>would put some of the battery connections at a dangerous potential
>with respect to the concrete floor below you, the water pipe behind
>you, etc.
This is of course a possibility but a simple safety device (such as those
sold now for protecting from mains shock based on earth-path leakage
currents) could be used to virtually eliminate this problem. It's also a
case of good design -- I've seen a guy nearly lose a hand when the wrench
he was using fell across a very high capacity battery and shorted the
terminals -- electricity is dangerous, you've got to design for safety.
>The cost of the individual batteries in the above example is the
>same whether they are wired up in a low or high voltage configuration.
>There is no difference in size or weight between two banks configured
>to store the same amount of energy (watt-hours). If a single cell
>fails in either arrangement you still have to replace the battery
>containing that cell. As you make that replacement you will have
>to shut down your 240 volt equipment, due to the single series path
>that would be broken.
Diodes, diodes, diodes :-)
>The bad battery in the 48/24/12 volt bank
>could be replaced without shutting down the loads. Unexpected
>voltages may spring up between a connector and battery post when
>a single series path is broken. Unexpected voltages may cause trouble
>if the operator is inadvertantly grounded while configuring or
>maintaining his high voltage battery bank.
I think I have satisfied those points n'est pas?
>(sorry for rambling -- this was just bugging me on the way home from
> work this morning. )
That's okay, you make some good points ... points which I'd already
considered and feel that I've solved to my own level of satisfaction.
Robert Kleinschmidt
Ross Alexander <r...@cs.athabascau.ca> wrote:
A tomcat who lives with me once decided that my 2024 would make a good
territorial marker. Ie, he urinated on it repeatedly. Eventually it
shut down (quite gracefully) and it required a full teardown and
distilled-water wash job to get it running again. Since just about
everything was either potted or made from stainless steel, there was
no permanent damage.
Good story. Sad though that safe design has to prevent the inverter from
vaporizing the cat.
Rob Kleinschmidt
>Ross
Robert Kleinschmidt
Bruce Simpson <br...@faxmail.co.nz> wrote:
>ro...@hal.COM (Robert Kleinschmidt) wrote:
>>3) [Possible difficulties of configuring a high voltage battery bank]
>
>Your battery bank is an energy store. You get the same amount of energy
>out of a 10Ah 240V battery as you do from a 100AH 24V battery and there's
>often little difference in size or weight. The larger number of batteries
>required to store 240V can be seen as a bonus insomuch as each battery is
>cheaper and can be replaced independently of the others (avoiding the need
>to throw out an expensive battery just because a single cell has failed).
>This is along the lines of the more expensive stackable cells sometimes
>used for deep-discharge storage solutions -- just cheaper :-)
Not necessarily true. Depends on commercial availability of appropriate
batteries. As in the case of another poster, my current :-) battery bank
is a 16 x 6v x 350 AH. array, which could not be reconfigured to the
higher voltage. Not sure what might be available in smaller units or
stackable 2v cells, but suspect a higher cost-per-watt for smaller
capacity batteries. Any thoughts on commercially available battery choices ?
>>4) Why a per-outlet or even per-circuit solution ? In ~7 years,
>> our Trace 2024 unit went into one graceful shutdown due to
>> overload, but I have yet to hear of a real failure. Can
>> anyone on the net cite a failure story ? Seems to me as if
>> even a pair of devices would provide more than adequate
>> redundancy. (I grant you that the ability to build a system
>> incrementally and possible cost reductions are highly desirable.)
>
>There's no reason why you couldn't use just one or two bridges to service
>then entire house, it's only my preference to have one per room or outlet.
>It lets me hang heavy loads off multiple outlets without having to worry
>about the total loading and its effect on a central unit.
Clearly there are some limitations imposed just by wire sizes. To my
mind, a device-per-outlet configuration would either allow circuit
overloads or would limit the loads on given outlets. I had at one
time considered separate A.C. feeds for our TVs, VCRs and Satellite
receiver, but decided that it would be unwise to give outlets individual
personality quirks unless I went to a unique plug/outlet type to protect
against non-knowledgeable users.
In addition to wiring concerns, it seems vaguely possible that too
much parallelism might also allow undesirably high rates of battery
discharge, which might lead you back to a central protection device
anyway.
>>5) [a "pass-through" mode for grid or genset power. ]
>There's no reason why a pass-through mode couldn't be built into the bridge
>unit(s), a simple relay to switch the bridge out of circuit and route the
>AC straight to the outlet would suffice. Proabably just another $20 worth
>of components.
Basically, all of the arguments I have raised are ones that lead me towards
a more conventional lower voltage centralized setup, with some minimal
redundancy and (hopefully) the ability to build capacity in increments.
As in the case of the $20 relay component, it would seem that too much
parallelism might lead to a multiplication of costs. (perhaps a single
relay and a shutdown of the clocking would accomplish this ?)
Additionally, for the U.S., there is the "split" wiring setup, where
a house will contain both 110-120 and 220-240 circuits. Not sure
how well this works when you try to match battery voltages to final
delivered voltage. (I suppose this could be accomplished by multiple
taps to the battery bank and/or more electronics).
>>Regards, and thanks again for some interesting ideas.
>
>If ideas were $ I'd be a millionaire :-)
>
Stranger things have happened to people I guess.
Duncan Campbell
Bruce Simpson <br...@faxmail.co.nz> wrote:
>dcam...@goblin.punk.net (Duncan Campbell) wrote:
> vvvvvvvvvv
>>If your building block is the six volt Trojan golf cart battery
>>then the minimum 24 volt setup will require four batteries. The
>>minimum 240 volt setup will require fourty batteries, which would
>
>You would only need 20 x 10AH batteries.
yes. 20 batteries at 12 volts each. 40 batteries at 6 volts each.
>
>But why use 6V batteries?
>
Given the suppliers available to me and the needs of my 24 volt
auction surplus inverter, I found that the 6 volt Trojans pack a
lot of power in a reasonable space within the budget I allotted.
The batteries I happened upon are rated at 205 amp-hours, which
will drive a single computer load, plus a few peripherals, for
approximately a day after the PG&E fails.
I preferred the all-series connection to balance the charging
and discharging currents through the individual batteries.
>
>In this case you make sure that your battery pack has enough voltage
>"headroom" to allow the bypassing of one or more of your batteries. This
>way you can remove a battery from the circuit without the need to
>disconnect the DC load and without affecting the AC output. This is done
>by connecting a suitably rated diode across the output of each battery such
>that if that battery is removed, the diode then services to bypass the
>missing cell.
This is an excellent idea, though rather expensive to implement.
I congratulate you for coming up with it, and slap myself for
missing it.
"suitably rated" would mean it can comfortably handle the entire
load current while you are off dealing with the new battery. These
would be hefty rectifiers mounted on isolated heat sinks with thick
cable, and would only be used when one of the batteries has been
removed from the system.
The cost of the diodes would definately swing ones choice of battery
back to the higher voltage units -- you would need 20 where I had
planned on 40.
With one battery removed the battery bank cannot be charged. No
current will flow _into_ the battery bank, though it may be used
to fill in the moments when the charger is not giving output.
Any attempt to charge the battery bank will place a voltage of
(applied voltage) minus (current bank voltage) across the diode
and across the terminals you will connect to the missing battery.
That voltage may not be significant, but may confuse your charge
regulator. The batteries may call for charge but as soon as it
connects the charging source it will see a much higher battery
voltage and shut itself off. Depending on how your charge reg
handles this you may have an overspeed condition on a windmill,
if you use one (memory fails me).
Best to disable your charger anyway if you are messing with the
battery bank to keep the charge even across your batteries.
>
>easily remove one or two of the 12V batteries without compromising the AC
>output or reducing capacity by more than the rated capacity of the
>individual batteries ... in this case 10AH. The inclusion of the diode
Technical detail: when you remove one battery from a series
string of 10AH batteries the remaining bank will still be
rated at 10AH (unless you removed the last one.) It will deliver
its 10AH at a lower voltage, deliver fewer watt-hours, store
less energy, etc, but will retain the 10AH rating.
>also ensures that even when there is an AC (and resulting DC) load on the
>system when the battery is removed, no more than 0.7volts will appear
>across the removed battery leads, thus reducing the potential for
>electrical shock.
>
Some high current silicon rectifiers can exceed two volts when
forward biased. Then there's the reverse voltage, should something
try to charge the bank... (repeat, repeat, repeat)
>>[ I mentioned an open series circuit ]
>[ You mentioned diodes and 0.7 volts ]
>>[ I mentioned parallel branches to maintain voltages ]
>[ You mentioned headroom and diodes ]
>>[ I mentioned ground faults ]
>
>This is of course a possibility but a simple safety device (such as those
>sold now for protecting from mains shock based on earth-path leakage
>currents) could be used to virtually eliminate this problem. It's also a
These things (GFCIs) do not all work as well as the package would
lead you to believe. A friend of mine installed one in his bathroom
in order to feel safe. I stuck my key in the outlet and started
batting at the water faucet with my other hand. Plenty of shocks.
The safety switch never tripped. (125 volts -- you can get away with
being stupid like that.)
But I do see your point. Be careful, own a voltmeter, know there
is high voltage present, and you'll be ok.
>>[ same cost, same weight, difficulty of maintainance ]
>Diodes, diodes, diodes :-)
>>[ series path, operator safety, load interruption ]
>I think I have satisfied those points n'est pas?
>
>>(sorry for rambling -- this was just bugging me on the way home from
>> work this morning. )
>
>That's okay, you make some good points ... points which I'd already
>considered and feel that I've solved to my own level of satisfaction.
>
Its been fun bouncing ideas back and fourth with you. I did
mention the H-bridge distributed inverter setup to a local
electronics company, along with your email address and a few
of our postings back and fourth. In the mean time I would
suggest ripping apart a small StatPower or TrippLite inverter
and feeding it an external high voltage input after its own
DC-to-DC converter. It will probably still need 12 volts for
its logic/control circuitry, but at a very low current.
I have had better luck with IGBTs than with FETs in power
control. A nifty isolated IGBT(*) driver board made by Unitrode
is available through Allied, if you can ever get in touch with
Allied. They are not particularly responsive. Put it this way...
Allied's catalog suggests that one visit their web site. The
web page directs one to phone an 800 number for their BBS. The
BBS reports that it needs vt220 or a higher vt model number on
your end, and gives you an ftp address where you can get one.
Back to the 'net, retrieve the suggested term package, call the
BBS back, and it says press F6 for current availability. The
function keys in the suggested term package don't have the
desired effect -- F6 sends "8" -- and Allied ignores email asking
politely how to access this information.
Ain't storage fun? :-)
* IGBT - insulated gate bipolar transistor. Three terminals,
labled collector, emitter, and gate.
Duncan.
Hugh Piggott
Very interesting correspondence between
br...@faxmail.co.nz (Bruce Simpson) and
dcam...@goblin.punk.net (Duncan Campbell)
I should like to add the following remarks:
>dcam...@goblin.punk.net (Duncan Campbell) wrote:
>>the larger battery bank holds ten times the charge is sometimes of
>>little use.
>
>You would only need 20 x 10AH batteries.
Question is where do you get these small capacity deep cycle cells?
>
>>Assuming you set up your banks to hold the same amount of stored
>>energy then a 24 volt and 240 volt battery bank might both
>>consist of fourty 6v batteries interconnected.
>
>But why use 6V batteries?
Question really is, can you get small capacity, deep cycle 2V cells
as units for the right price?
>
>>If one of the batteries in the 24 volt setup goes bad -- drains,
>>fails the specific gravity test, etc, one can just disconnect it,
>>drive to your local battery exchange, and hook the new one up when
>>you return. With a 10% drop in capacity your load sees no change.
>
>In this case you make sure that your battery pack has enough voltage
>"headroom" to allow the bypassing of one or more of your batteries. This
>way you can remove a battery from the circuit without the need to
>disconnect the DC load and without affecting the AC output. This is done
>by connecting a suitably rated diode across the output of each battery such
>that if that battery is removed, the diode then services to bypass the
>missing cell. If you've got the recommended 40V of headroom then you can
>easily remove one or two of the 12V batteries without compromising the AC
>output or reducing capacity by more than the rated capacity of the
>individual batteries ... in this case 10AH. The inclusion of the diode
>also ensures that even when there is an AC (and resulting DC) load on the
>system when the battery is removed, no more than 0.7volts will appear
>across the removed battery leads, thus reducing the potential for
>electrical shock.
>
Very ingenious idea! :-) but..
How do you charge the batteries?
How do you regulate charge voltage?
Ho do yuo pass reverse current?
Do you charge seperate 24V chunks, at all different 24volt tappings on the system?
Panels will then be live ('hot'?) re:earth
>>Another consideration is the possibility for some part of the DC
>>circuit to become grounded, either deliberately or accidentally. This
>>would put some of the battery connections at a dangerous potential
>>with respect to the concrete floor below you, the water pipe behind
>>you, etc.
>
>This is of course a possibility but a simple safety device (such as those
>sold now for protecting from mains shock based on earth-path leakage
>currents) could be used to virtually eliminate this problem. It's also a
>case of good design -- I've seen a guy nearly lose a hand when the wrench
>he was using fell across a very high capacity battery and shorted the
>terminals -- electricity is dangerous, you've got to design for safety.
>
All in favour of safety. I have a neighbour here who has used 110V dc off a string of telephone exchange batteries for years withou=
t any shock-hazard protection, and it gives me the horrors.
Where do yu get a DC earth leakage trip? Mains ones, in the UK, rely on the transformer coupling of the leakage current.
Let's all keep chucking in the ideas :-)
Bruce Simpson
>Bruce, we both seem quite good at repeating ourselves. :-)
Shall we call it "clarification"? :-)
>"suitably rated" would mean it can comfortably handle the entire
>load current while you are off dealing with the new battery. These
>would be hefty rectifiers mounted on isolated heat sinks with thick
>cable, and would only be used when one of the batteries has been
>removed from the system.
Correct but remember that since we're dealing with higher voltages the
currents are correspondingly lower. ie: if you're drawing 110W at 110VAC
then the current is only 1A. Compare this to a 24V system where the DC
curernt will be between 4A and 5A depending upon inverter efficiencies.
Not so important at lower loadings but when you get up to (say) 1,100 watts
you're looking at 10A for my system and 40A-50A for a 24V system.
If you allow for a maximum power draw of 5.5KW (pretty reasonable for a
wind/solar system) then the currents are 50A and 200A-250A respectively. A
good 80A continuous diode (allowing a 60% over-load rating) is not that
expensive. Even 20 of them won't break the bank.
>The cost of the diodes would definately swing ones choice of battery
>back to the higher voltage units -- you would need 20 where I had
>planned on 40.
>With one battery removed the battery bank cannot be charged. No
>current will flow _into_ the battery bank, though it may be used
>to fill in the moments when the charger is not giving output.
Even this is not insurmountable. A small shunt circuit could be attached
which automatically switched on when the current flow changed from
discharge to charge.
Unfortunately this all adds to complexity and I think that in keeping with
KISS, the *best* option may be to have a fuse-rack with a fuse that can be
inserted across each battery. When a battery is removed, the diode takes
over. A fuse is then inserted (rated at 1.5 times the max current draw).
This then shunts the diode and allows for forward and reverse current flow
through the batteries. The fuse should be removed before the battery is
replaced in position but even if you forgot, all that would happen is that
the fuse would blow. You'd need a fuse carrier for each battery but only
one or two fuses (for those batteries which were being removed from
circuit). How's that?
>Any attempt to charge the battery bank will place a voltage of
>(applied voltage) minus (current bank voltage) across the diode
>and across the terminals you will connect to the missing battery.
>That voltage may not be significant, but may confuse your charge
>regulator. The batteries may call for charge but as soon as it
>connects the charging source it will see a much higher battery
>voltage and shut itself off. Depending on how your charge reg
>handles this you may have an overspeed condition on a windmill,
>if you use one (memory fails me).
I think my previous paragraph offers a simple enough solution.
>Best to disable your charger anyway if you are messing with the
>battery bank to keep the charge even across your batteries.
>>
>>easily remove one or two of the 12V batteries without compromising the AC
>>output or reducing capacity by more than the rated capacity of the
>>individual batteries ... in this case 10AH. The inclusion of the diode
>Technical detail: when you remove one battery from a series
>string of 10AH batteries the remaining bank will still be
>rated at 10AH (unless you removed the last one.) It will deliver
>its 10AH at a lower voltage, deliver fewer watt-hours, store
>less energy, etc, but will retain the 10AH rating.
Yes, sorry, my mistake I should have said watt/hours.
>>also ensures that even when there is an AC (and resulting DC) load on the
>>system when the battery is removed, no more than 0.7volts will appear
>>across the removed battery leads, thus reducing the potential for
>>electrical shock.
>>
>Some high current silicon rectifiers can exceed two volts when
>forward biased. Then there's the reverse voltage, should something
>try to charge the bank... (repeat, repeat, repeat)
The currents are not as high as you might think (due to the higher
voltages) and the fuse concept solves the charging problem (repeat, repeat
repeat :-)
>These things (GFCIs) do not all work as well as the package would
>lead you to believe. A friend of mine installed one in his bathroom
>in order to feel safe. I stuck my key in the outlet and started
>batting at the water faucet with my other hand. Plenty of shocks.
>The safety switch never tripped. (125 volts -- you can get away with
>being stupid like that.)
I don't think they're supposed to stop you getting a shock per-se, their
role is to stop you from getting a *fatal* shock. I suspect that the trip
current is deliberately set higher than the "tingle threshold" to avoid
false triggering and to let people know (buzz buzz) if their equipment is
faulty.
Duncan Campbell
Makes it sound like that chemical process that takes skin
oil, clumps it together, and makes it stick to your hot
tub filter.
>dcam...@goblin.punk.net (Duncan Campbell) wrote:
>>"suitably rated" would mean it can comfortably handle the entire
>>load current while you are off dealing with the new battery. These
>
>If you allow for a maximum power draw of 5.5KW (pretty reasonable for a
>wind/solar system) then the currents are 50A and 200A-250A respectively. A
>
50 amps for a few hours is significant. For a few seconds it
wouldn't matter much. Save on heat sinks. I like the fuse idea.
In response to recent email I came up with an even cheaper solution
than the fuses -- just apply jumper cables across the terminal
wires once the battery is removed. Take the cables off before
attaching the new battery.
If you leave your charger operational you might have to adjust
it to the new effective battery bank voltage so that it doesn't
overcharge the remaining batteries. Not a big deal if you are
missing one of twenty 12v batteries (extra 0.63 volts per battery),
but could cause trouble if you replace a larger fraction of
your bank.
>
>>Any attempt to charge the battery bank will place a voltage of
>>(applied voltage) minus (current bank voltage) across the diode
>>and across the terminals you will connect to the missing battery.
>
solved with shunts applied across the diodes in place of the
batteries. Perhaps a battery-shaped shut, with terminals,
made from a toasted battery. :-)
>>batting at the water faucet with my other hand. Plenty of shocks.
>>The safety switch never tripped. (125 volts -- you can get away with
>
>I don't think they're supposed to stop you getting a shock per-se, their
>role is to stop you from getting a *fatal* shock. I suspect that the trip
So you're saying I should test it again to see if it will pass
a fatal shock? More on GFCIs from my mailbox (sender withheld)
mail| (the National Electric Code to be specific). To be acceptable
mail| I know the code requires the interrupt to trip when return current is less
mail| that hot current by .005 amps (5 milliamps to be clear). What I don't
mail| recall is what it has to say about how fast it must trip. I'm not
mail| surprised your GFCI didn't trip if you were just slapping the faucet.
mail| I'd be surprised if it didn't trip if you were to grab ahold of the
mail| faucet. Obviously, I don't recommend this experiment ;-)
I also came up with a simple design for a DC ground fault detector.
Dropped it in the mail and can't find my copy of it anywhere.
Anyway, such faults can be detected.
I will be out of town for a week or so, which is longer than my
server holds onto articles in this group. If this thread still
has yet to die, please Cc: your articles to me.
It's been fun.
Duncan.