Boosting output from 120V inverter
Bevan Weiss
The transformer itself probably wont be rated that much above what the
entire inverter is rated at. So just replacing the MOSFET's wouldn't do the
trick, you'd have to replace the transformer as well, and the transformer is
almost the entire cost of an inverter, apart from design.
"DaveC" <an...@example.net> wrote in message
news:0001HW.B9DF7C5F...@news.covad.net...
> 200W inverter needs to run a bit more power. I know how cheap these things
> are, but I get my kicks from using and/or modifying what I have and
getting
> the experience of electronics theory/application.
>
> How should I go about matching up a higher-amperage MOSFET to replace the
> ones there? Same g-d and g-s voltages? Higher d current? Pin
configuration,
> of course. What are the critical specs I should focus on when looking for
an
> upgrade?
>
> The MOSFETS that switch the output are not heatsinked very well, and there
is
> room for a small cooling fan.
> --
> Note that my return address is corrupted in an attempt to reduce spam. If
you
> choose to e-mail me, please correct my address as described below.
>
> Thanks,
> Dave
> ---
> Dave Carpenter
> Sound Logic
> voic...@d8n8a8i.com
>
> Remove the numbers to reply via e-mail
>
Sofie
It is not wise to push these things near and past their design limits....
while you can fool around and maybe get a very small increase, you will
DECREASE the life of the components due to heat & stress and when (not if)
it does fail you will risk damaging any equipment you have hooked to it.
Spend your money more wisely on a higher wattage inverter...... as you
indicated they are quite cheap so why screw around with it and risk blowing
it up and your equipment that is powered by it.
--
Best Regards,
Daniel Sofie
Electronics Supply & Repair
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
"DaveC" <an...@example.net> wrote in message
news:0001HW.B9DFECA2...@news.covad.net...
> On Fri, 25 Oct 2002 23:03:30 -0700, Bevan Weiss wrote
> (in message <10356122...@news.orcon.co.nz>):
>
> > It's not that simple unfortuneately...
> > The transformer itself probably wont be rated that much above what the
> > entire inverter is rated at. So just replacing the MOSFET's wouldn't do
the
> > trick, you'd have to replace the transformer as well, and the
transformer is
> > almost the entire cost of an inverter, apart from design.
>
> Futz!
>
> Can I gain any increase by properly heatsinking the existing MOSFETs and
> adding a cooling fan? The 2SK1221 is rated at 250V 10A 80W. My
understanding
> is that it will handle most of this if kept cool, and less when hot.
>
> According to the data sheet, power handling capability of this device is
80W
> at 25oC and derated - pretty much linearly - down to zero W at 100oC. At
> about 50oC it can handle 50W, which seems like a reasonable temp to keep
it.
>
> But in any case, I think that cooling it as much as possible could only
help.
N. Thornton
I think it could work actually, with a fan. Transformers will handle
much more power when fan cooled. So I'd put as powerful a fan in there
as you can squeeze, uprate your fets, and watch the TF temp to see
what it'll manage continuous. Hopefully you'll get a significant
increase.
Since youre running off battery, you'll want to power the fan
intelligently.
If you're real determined, theres always the outer edge option of
putting the TF in a well heatsunk tin full of oil - no fan power then.
Regards, NT
Watson 'Atto Parsec' Name
about...
> DaveC:
> It is not wise to push these things near and past their design limits....
> while you can fool around and maybe get a very small increase, you will
> DECREASE the life of the components due to heat & stress and when (not if)
> it does fail you will risk damaging any equipment you have hooked to it.
> Spend your money more wisely on a higher wattage inverter...... as you
> indicated they are quite cheap so why screw around with it and risk blowing
> it up and your equipment that is powered by it.
> --
> Best Regards,
> Daniel Sofie
> Electronics Supply & Repair
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
I couldn't have said it better myself. It's foolish to risk damage. And
even if you don't let the somke out, you will probably blow a fuse that is
hidden somewhere inside. But then Dave, here, will probably wrap tinfoil
around it, huh. PFFFT!! :-p
--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
My email address is whitelisted. *All* email sent to it
goes directly to the trash unless you put NOSPAM in the
Subject: line. alondra101 <at> hotmail.com
Don't be ripped off by the big book dealers. Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html
@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@
daestrom
"AC/DCdude17" <Je...@prontoREMOVETHISmail.com> wrote in message
news:3DBB21A1...@prontoREMOVETHISmail.com...
> X-No-Archive: Yes
>
> Fan cooling will not much reduce internal hot spot. Hot spot is what
hurts the winding. Oil cooling will
> do. Power companies do it on their massive transformer. They've got oil
pump too.
>
And fans. And many large units have spacers between windings to allow the
oil to circulate between the windings. Helps lower the hot-spots by having
less distance for the heat to travel from the hot spot to the oil.
daestrom
Philip A. Marshall
as long as you cool the oil somehow. otherwise it will just heat up
slowly and then stay just very hot for a long time. as you say,
passive cooling may work, just so long as you watch the temperature of
the oil to make sure it stabalizes at a reasonable temperature.
Bevan Weiss
"N. Thornton" <big...@meeow.co.uk> wrote in message
news:a7076635.02102...@posting.google.com...
Transformers generally aren't the best items to heatsink...
That's not the only problem that they have...
You have to consider that each winding in the transformer is an inductor
(easily thought of)... Now each of these seperate inductors will be
designed such that they have a maximum saturation current equal to the
maximum current the device is rated at. Any current higher than the
saturation current will begin to cause a much faster rate of heating, you
will begin to lose serious wattage through the iron and copper losses of the
transformer. Couple this with the normally very tight windings of small (as
these kind of transformers are designed) transformers and you have a serious
heating problem. It's just not worth the extra effort...
Work out how they design inverters (the principles aren't hard at all), then
spend a little time thinking of better ways to go about certain aspects (ie
the dc->ac conversion, and then the transformation of the voltage), then
begin to breadboard a new inverter (designed by yourself). Start with a low
current design (ie low wattage), then check to see which parts are the
limiting factors, and expand from there.
I'm sure that it'll be worth your time if you really want to pursue the
extra wattage yourself as opposed to just buying a higher wattage inverter.
Bob Wilson
an...@example.net says...
>
>On Fri, 25 Oct 2002 23:03:30 -0700, Bevan Weiss wrote
>(in message <10356122...@news.orcon.co.nz>):
>
>> The transformer itself probably wont be rated that much above what the
>> entire inverter is rated at. So just replacing the MOSFET's wouldn't do
the
>> trick, you'd have to replace the transformer as well, and the transformer
is
>> almost the entire cost of an inverter, apart from design.
>
>
>Can I gain any increase by properly heatsinking the existing MOSFETs and
>adding a cooling fan? The 2SK1221 is rated at 250V 10A 80W. My
>understanding
>is that it will handle most of this if kept cool, and less when hot.
You're not listening to the man! It is totally pointless to increase the
current through the primary drive FETs, if the transformer they feed
saturates (and therefore blows up these same FETs!).
>According to the data sheet, power handling capability of this device is
>80W at 25oC and derated - pretty much linearly - down to zero W at 100oC.
>At about 50oC it can handle 50W, which seems like a reasonable temp to keep
>it. But in any case, I think that cooling it as much as possible could only
>help.
How do you think cooling is going to help? If all it does is to allow an
increase in primary (FET) current, then we're back to the same problem
again, namely that this increased power, when fed to the transformer(s) will
cause them to saturate and destroy the FETs.
Unless you understand the principles of high frequency transformer design,
and can redesign for higher power throuput, you should forget about the
entire thing. It ain't gonna fly.
Bob.
Bob Wilson
big...@meeow.co.uk says...
>
>"Bevan Weiss" <kaiz...@hotmailNOSPAM.com> wrote in message
>58...@news.orcon.co.nz>...
Baloney! The transformers used in the typical low cost inverter (and I can
speak from experience here) are designed to run close to their saturation
limits. Push them harder (even if you cool them) and they start going into
saturation.
Bob.
Kevin White
saturation!
The current in the secondary opposes the magentic field created by the
primary. (assuming a forward converter as I believe they usually are). If
there is feedback that increases the drive when the output drops then yes it
may put the transformer into saturation. But that is usually determined by
the input voltage.
This is assuming a system with a separate oscillator and output stage, the
traditional push-pull self-timed inverter always runs the transformer to
saturation anyway as part of its normal operation.
You may not be able to get much more out of the transformer because the
resistance and leakage inductance may be limiting the output power.
kevin
"Bob Wilson" <rfwi...@intergate.nospam.bc.ca> wrote in message
news:urmjsa7...@corp.supernews.com...
N. Thornton
First, loading up the TF will tend to take it out of saturation, not
into it.
Re the difference heatsinking makes, lets look at TF heat paths. Youve
got copper winding, which is an around 80% solid copper block, with
the remaining 20 or so % being mostly air. So thats a good heat
conductor. Then youve got the metal / air interface, which we know is
a poor conductor of heat. Youve also got the iron core which is
intermediate in conduction.
So the big heat loss block is the metal / air interface, and fan
cooling increases that heat flow point greatly. Thats why fan cooling
makes such a big difference.
Lets throw a number up. Say the breakdown point is 200C, and the TF
gets upto 100C on its outer winding surface - I'm picking easy numbers
here. Now if we reduce the outer winding surface to 50C by fanning,
the temp drop across the copper has gone from 100C to 150C. So the
hotspot temp wil be the same with 1.5x the power diss in the TF.
And with heatsinks and aggressive fanning one could hopefully do
better than that.
Actually no, I assumed there all the hewat was dissed at the hot spot,
whcih of course it isnt. So we should get better figs than that IRL.
Bottom line, I think its do-able, but of course not risk free.
>I couldn't have said it better myself. It's foolish to risk damage.
And
I'd say it depends what youre driving. If youre powering the fridge or
a light I wouldnt worry unduly. The control tronics is not going to
suffer from heat, and a brief soak test will check that runs properly
- eg driving some light bulbs. From then on I doubt the thing is
suddenly going to start putting out excess V, so you could run almot
anything off it without worry I think. Power failure would be the real
risk.
Obviously this is not a risk free project, and you live with the
results - which will most likely be positive I tihnk.
>>If you're real determined, theres always the outer edge option of
>>putting the TF in a well heatsunk tin full of oil - no fan power
then.
>as long as you cool the oil somehow. otherwise it will just heat up
>slowly and then stay just very hot for a long time. as you say,
>passive cooling may work, just so long as you watch the temperature
of
>the oil to make sure it stabalizes at a reasonable temperature.
Yup. If the TF is put in oil in a tin, I'm thinking one automatically
has larger contact surface with the air, plus considerebale thermal
capacity of oil too. I would choose thin oil - well, I wouldn't choose
oil at all really.
>You have to consider that each winding in the transformer is an
inductor
>(easily thought of)... Now each of these seperate inductors will be
>designed such that they have a maximum saturation current equal to
the
>maximum current the device is rated at. Any current higher than the
>saturation current will begin to cause a much faster rate of heating,
you
>will begin to lose serious wattage through the iron and copper losses
of the
>transformer. Couple this with the normally very tight windings of
small (as
>these kind of transformers are designed) transformers and you have a
serious
>heating problem. It's just not worth the extra effort...
and...
>Baloney! The transformers used in the typical low cost inverter (and
I can
>speak from experience here) are designed to run close to their
saturation
>limits. Push them harder (even if you cool them) and they start going
into
>saturation.
I've always understood that more load will reduce saturation effect.
Certainly it will send copper losses way up, but I'll address that. Am
I wrong?
>You may not be able to get much more out of the transformer because
the
>resistance and leakage inductance may be limiting the output power.
Lets see, invertors are normally specced at around 90% efficient: now
even if we attributed every last drop of that loss to TF copper
losses, we could still get far more power output. I dont think R is a
limiting factor myself.
Could you explain how leakage inductance would affect it? Thats
something I'm not so hot on myself.
>Unless you understand the principles of high frequency transformer
design,
>and can redesign for higher power throuput, you should forget about
the
>entire thing. It ain't gonna fly.
Well, I bet it will.
>Transformers generally aren't the best items to heatsink...
They heatsink reasonably well. I've done it before, every microwave
oven I've ever seen does it too. Try running an old zapper with no
fan, it wouldnt survive long. The fan makes a sgnificant difference.
I'd be interested to hear what happens if you try it, I am optimistic
myself.
Regards, NT
Peter Gottlieb
will not circulate freely.
Big UPSes are available cheaply at hamfests. Why go through all this to try
and boost a tiny one?
"N. Thornton" <big...@meeow.co.uk> wrote in message
news:a7076635.02102...@posting.google.com...
Kevin McMurtrie
DaveC <an...@example.net> wrote:
>200W inverter needs to run a bit more power. I know how cheap these things
>are, but I get my kicks from using and/or modifying what I have and getting
>the experience of electronics theory/application.
>
>How should I go about matching up a higher-amperage MOSFET to replace the
>ones there? Same g-d and g-s voltages? Higher d current? Pin configuration,
>of course. What are the critical specs I should focus on when looking for an
>upgrade?
>
>The MOSFETS that switch the output are not heatsinked very well, and there is
>room for a small cooling fan.
Put a scope on the MOSFET outputs. In the inverters I've seen, the
current output is limited by inductance leakage. The MOSFETs'
resistances loose very little power.
Tripp Lite's bulky 60-70 Hz transformers are the worst. There's a 50VAC
difference between the open circuit and the full-load voltages. Their
high current rating comes from MOSFETs that dissipate little.
Bob Wilson
kevin...@attbi.com says...
>
>Well actually, increasing the load will take the transformer out of
>saturation!
>
>The current in the secondary opposes the magentic field created by the
>primary. (assuming a forward converter as I believe they usually are). If
>there is feedback that increases the drive when the output drops then yes
it
>may put the transformer into saturation. But that is usually determined by
>the input voltage.
>
>This is assuming a system with a separate oscillator and output stage, the
>traditional push-pull self-timed inverter always runs the transformer to
>saturation anyway as part of its normal operation.
>
>You may not be able to get much more out of the transformer because the
>resistance and leakage inductance may be limiting the output power.
The standard el-cheapo inverter most certainly NEVER pushes the
transformer into saturation. You are referring to the Royer topology, which
is not used in any commercial products of any significance today.
The standard topology for cheap quasi-sine inverters uses a driven push-pull
that is driven with dead time between half cycles. The dead time reduces the
possibility of core walking (by allowing some time for residual field to
discharge into the load between each half cycle).
Push any inverter of this topology too hard, and the core can saturate.
Bob.
Dave Martindale
>Youve also got the iron core which is
>intermediate in conduction.
Inverters don't use iron cores, unless it's a big heavy inverter that
uses a 60 Hz transformer. Is that what you're talking about?
Otherwise, the transformer core is some kind of powdered iron or
ceramic, not a terribly good heat conductor, and the copper windings
are probably buried inside the core where a fan won't do any good.
>They heatsink reasonably well. I've done it before, every microwave
>oven I've ever seen does it too. Try running an old zapper with no
>fan, it wouldnt survive long. The fan makes a sgnificant difference.
It also doesn't have to have a decent operating life when it's used
continuously, because microwave ovens aren't used continuously.
Dave
Frithiof Andreas Jensen
> of course. What are the critical specs I should focus on when looking for
an
> upgrade?
Heat!
I would not bother replacing anything:
First off: Measure the temperature rise of the heatsink for the FET's, the
transformer (and any other component running hot) at rated load. This may
require several hours for the temperature to stabilise!
Next: Add a fan to blow air through the unit, heatsink & transformer should
be in the airflow. Measure again. Increase the load (very) gradually until
the temperature rise after several hours is the same as what you had without
the fan.
Measure the load -> this is what you can eek out of the unit with forced
cooling. It will not be possible to improve this without major surgery -
especially to the transformer.
There may be overcurrent trips, fuses and other stuff preventing you from
increasing the power. It is also dangerous - there is high voltage, heat
sinks may not be insulated etc. You know what you are doing rite?
Peter Gottlieb
battery run times and the transformer thermal inertia into consideration
meaning that the long-term output capability is somewhat less than the unit
rating.
I saw some nasty burns resulting from people wearing jewelry while working
on UPSes. It was not the voltage that got them, it was the extremely high
available currents from the batteries. In one case a guy had his wedding
band turn orange in an instant and almost lost his finger. So *please!*
Take *all* jewelry off while working on these things.
"Frithiof Andreas Jensen" <frithio...@removethis.ted.ericsson.dk> wrote
in message news:apjcp0$m51$1...@newstoo.ericsson.se...
Kevin White
these days - that's what I meant by using the word "traditional".
Without feedback though (e.g changing the dead time to perform regulation)
increasing the load current will take the transformer away from saturation
not into saturation.
kevin
"Bob Wilson" <rfwi...@intergate.nospam.bc.ca> wrote in message
news:urp2t6g...@corp.supernews.com...
N. Thornton
>A lot of those transformers are varnish or epoxy impregnated and the
oil
>will not circulate freely.
Well, I don't seriously suggest an oil slick, but if you do want to
try it, it will still work anyway, since the biggest heat flow barrier
is still the transformer to air interface.
>Big UPSes are available cheaply at hamfests. Why go through all this
to try
>and boost a tiny one?
Adding a fan... its not much, and it can uprate the OP's invertor.
Sounds OK to me. Also the OP probably _wants_ to do it, see what they
can achieve. And I bet fitting a fan would be quicker than going to a
hamsterfest. And far cheaper FWIW. I like a challenge anyway.
>>Youve also got the iron core which is
>>intermediate in conduction.
>Inverters don't use iron cores, unless it's a big heavy inverter that
>uses a 60 Hz transformer. Is that what you're talking about?
>Otherwise, the transformer core is some kind of powdered iron or
>ceramic, not a terribly good heat conductor, and the copper windings
>are probably buried inside the core where a fan won't do any good.
Yes, I was thinking of the low freq ones for some reason. A fan will
make just the same difference with a powder cored TF. Look at the heat
paths I mentioned: just the same principle applies with a powder cored
TF. Same result.
I think Frithiof has just the right idea.
Regards, NT
"Frithiof Andreas Jensen" <frithio...@removethis.ted.ericsson.dk> wrote in message news:<apjcp0$m51$1...@newstoo.ericsson.se>...
Peter Gottlieb
>
> >A lot of those transformers are varnish or epoxy impregnated and the
> oil
> >will not circulate freely.
>
> Well, I don't seriously suggest an oil slick, but if you do want to
> try it, it will still work anyway, since the biggest heat flow barrier
> is still the transformer to air interface.
True.
>
>
> >Big UPSes are available cheaply at hamfests. Why go through all this
> to try
> >and boost a tiny one?
>
> Adding a fan... its not much, and it can uprate the OP's invertor.
> Sounds OK to me. Also the OP probably _wants_ to do it, see what they
> can achieve. And I bet fitting a fan would be quicker than going to a
> hamsterfest. And far cheaper FWIW. I like a challenge anyway.
>
The larger of the small UPSes do have a fan and it definitely helps.
Licensed to Quill
any real knowledge of how to do this.
The units seems to turn on and lights up but as soon as I try to get any
power out of it, the available power drops to about 1 milliamp. (Fuse not
blown though?) Is this a MOSFET problem or a transformer one?
Licensed to quill
"DaveC" <an...@example.net> wrote in message
news:0001HW.B9DFECA2...@news.covad.net...
> On Fri, 25 Oct 2002 23:03:30 -0700, Bevan Weiss wrote
> (in message <10356122...@news.orcon.co.nz>):
>
> > It's not that simple unfortuneately...
> > The transformer itself probably wont be rated that much above what the
> > entire inverter is rated at. So just replacing the MOSFET's wouldn't do
the
> > trick, you'd have to replace the transformer as well, and the
transformer is
> > almost the entire cost of an inverter, apart from design.
>
> Futz!
>
> Can I gain any increase by properly heatsinking the existing MOSFETs and
> adding a cooling fan? The 2SK1221 is rated at 250V 10A 80W. My
understanding
> is that it will handle most of this if kept cool, and less when hot.
>
> According to the data sheet, power handling capability of this device is
80W
> at 25oC and derated - pretty much linearly - down to zero W at 100oC. At
> about 50oC it can handle 50W, which seems like a reasonable temp to keep
it.
>
> But in any case, I think that cooling it as much as possible could only
help.
Joakim
> I have one of these units which I am trying to fix but substantially without
> any real knowledge of how to do this.
>
> The units seems to turn on and lights up but as soon as I try to get any
> power out of it, the available power drops to about 1 milliamp. (Fuse not
> blown though?) Is this a MOSFET problem or a transformer one?
>
> Licensed to quill
Hi. I think it probably is something else; MOSFETS usually fail
catastrophically when they fail (shorted or sometimes open), and the
only failure mode for a transformer that I can think of is shorted
turns due to insulation melting. So it's probably something else that
is acting up...
Kevin McMurtrie
"Licensed to Quill" <vinta...@compuserve.nospam.com> wrote:
>I have one of these units which I am trying to fix but substantially without
>any real knowledge of how to do this.
>
>The units seems to turn on and lights up but as soon as I try to get any
>power out of it, the available power drops to about 1 milliamp. (Fuse not
>blown though?) Is this a MOSFET problem or a transformer one?
>
[snip]
The current sensing resistor is probably blown.
Colin
Sounds like dried out electrolytics to me.
Check them all in cct with an ESR meter.
Regards
Colin.
Kevin McMurtrie
Colin <cm...@usa.net> wrote:
Electrolytic capacitors don't suddenly dry out. That takes decades;
longer than portable power inverters have been around. They can leak,
but that makes a noticable mess.
Peter Gottlieb
or contamination or damage around the leads prevents sealing then the
volatile (frequently DMF) electrolyte solvent evaporates out.
Many times the capacitance loss is not noticed until there is almost none
left. I don't know if this particular problem is being caused by a dead
electrolytic or not but I have seen far too many relatively new
electrolytics dried up.
"Kevin McMurtrie" <mcmu...@sonic.net> wrote in message
news:mcmurtri-3B3595...@typhoon.sonic.net...
Spehro Pefhany
<pe...@NewYorkNERD.com> wrote:
>They most certainly can dry out quickly. If the case isn't sealed properly
>or contamination or damage around the leads prevents sealing then the
>volatile (frequently DMF) electrolyte solvent evaporates out.
>
>Many times the capacitance loss is not noticed until there is almost none
>left. I don't know if this particular problem is being caused by a dead
>electrolytic or not but I have seen far too many relatively new
>electrolytics dried up.
I've seen brand new capacitors straight from the manufacturer
missing the entire rubber bung! They dried out almost immediately.
Not very good QC (but you should see their environmental
management methods!).
Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
sp...@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.co