Voltage vs. current in an incandescant..
Existential Angst
Inyone have data (or a link), or feel like doing the 'speriment?
25, 50, 75, 100, 120, 125, 150 V? AC and DC?
60 W, 100 W buhbs?
Can't find a VOM with an ammeter, lent out my variac.... goodgawd....
--
EA
Jordan
It's a curve, showing variable resistance.
Jordan
>
> It's a curve, showing variable resistance.
Or should I say, varying resistance, as the incandescant lamp is a
non-linear resistor.
pent...@yahoo.com
Stormin Mormon
--
Christopher A. Young
Learn more about Jesus
www.lds.org
.
"Existential Angst" <fit...@optonline.net> wrote in message
news:4d7b3c7e$0$31285$607e...@cv.net...
Denis G.
Jim Wilkins
Guess that means me, most everyone else I know would rather argue than
measure.
http://en.wikipedia.org/wiki/Aristotle
60W bulb, analog clamp-on ammeter with 10X range extender, HF
voltmeter calibrated at 12.000VDC. I set the current to the major tick
marks to squeeze maximum accuracy from the analog scale.
3....@0.1A, 20...@0.2A, 46...@0.3A, 82...@0.4A, 127...@0.5A.
The filament takes about half a second to stabilize at each new
temperature.
jsw
Jim Wilkins
> 3...@0.1A, 20...@0.2A, 46...@0.3A, 82...@0.4A, 127...@0.5A.
I see bogus Email address masking, so:
3.1V at 0.1A, 20.5V/0.2A, 46.4V/0.3A, 82.3V/0.4A, 127.2V/0.5A.
jsw
Ned Simmons
<fit...@optonline.net> wrote:
Per Fink's "Standard Handbook for Electrical Engineers":
(amps/AMPS) = (volts/VOLTS)^t
where t=.541 for gas-filled lamps and t=.580 for evacuated lamps.
Figure 26-7 in the 12th edition also shows this and other
relationships in graphical form. Fink references "LD-1 Lamp Bulletin,"
GE, 1946, by CE Weitz.
--
Ned Simmons
Bob Engelhardt
> ...
> http://imageupload.org/?di=712999347565
> ...
That chart shows voltage vs. current. I.e., current on the x
(horizontal) axis and voltage on the y (vertical). Through all my
engineering schooling and my own graphing, the x axis is the independent
variable and the y is the dependent one.
So, is the grapher in the linked chart really thinking about how the
voltage varies as the current changes? Or is he/she just an
ass-backwards grapher? Either way it seems ass-backwards to me.
Whenever I see this (switched independent and dependent variables) I
find it confusing and (hence) irritating. Yeah, I know, I'm
obsessive-compulsive.
Rant off,
Bob
BTW - while I'm ranting, let me get this off my chest: why does
Thunderbird have such a shitty spell checker? When it gives you a list
of suggested replacements, it often avoids the obvious; it doesn't
automatically accept the plural of singular words that it knows; and its
model for capitalization is screwy. Is this their little joke against
the obsessive-compulsives amongst us?
Jim Wilkins
Energy in equals energy out.
Black-body radiant heat loss:
http://hyperphysics.phy-astr.gsu.edu/hbase/thermo/stefan.html
I don't know how much the Argon transfers by conduction and
convection, or how closely they control the gas pressure. That "60W"
bulb which I use for a battery+inverter test load shows 55W on a KAW
at 120V.
jsw
Josepi
take two parameters to specify that..
I believe you would mean a non-static resistance as it varies..
------------------------
"Jordan" wrote in message
news:ZAHep.12766$gM3....@viwinnwfe01.internal.bigpond.com...
Josepi
----------------------
"Bob Engelhardt" wrote in message news:ilg16...@news5.newsguy.com...
Rant off,
Bob
pent...@yahoo.com wrote:
Josepi
us use them for precision power measurements as the waveform adopts many
harmonics in some units.
This could drive your accuracy of power testing around the bend is you want
more than one digit of accuracy. Multiply a distorted current waveform
measurement times a distorted voltage measurement (from the same voltage
source) and you can get power than is 10-20% in error very quickly.
---------------------
"Existential Angst" wrote in message
news:4d7b3c7e$0$31285$607e...@cv.net...
Awl --
whit3rd
> Inyone have data (or a link), or feel like doing the 'speriment?
Oddly, this cannot usefully be done as a V-versus-I plot.
The resistance of a tungsten filament, measured cold, is very
different from the resistance when incandescent and hot.
And, that difference means that a hot filament, lowered
to 60V, may have different current than a cold filament
raised to 60V. Try rotating a dimmer clockwise until
the light comes on, then turn counterclockwise to turn
it off- the effect is clear.
So-called 'ballast tubes' are just incandescent lamps designed
so the current is near constant over a range of voltages.
Bill Noble
all this information is available, in detail, for a large variety of
bulbs in the GE lamp handbook, among other places. The lamp is a highly
non-linear resistance element - in the incadescent region it
approximates a constant current device well enough to be used in some
ballast applications
Rich Grise
> Careful. Variacs are known waveform distorters. The government would not
> let us use them for precision power measurements as the waveform adopts
> many harmonics in some units.
>
Where do you get this information? A variac is nothing but an adjustable
autotransformer - what part introduces the distortion?
Thanks,
Rich
Jim Wilkins
A highly capacitive load, such as a light bulb filament.
Bob AZ
> A highly capacitive load, such as a light bulb filament.
Jim
Hunh?
Bob AZ
Lewis Hartswick
Variac is a registered trade mark. :-)
...lew...
Michael A. Terrell
Josepi wrote:
>
> Funny my DNS server shows this as a dead link.
Then you need a better DNS server.
--
You can't fix stupid. You can't even put a Band-Aid™ on it, because it's
Teflon coated.
Michael A. Terrell
They aren't tungsten, like a lightbulb.
Jim Wilkins
> On Saturday, March 12, 2011 1:27:19 AM UTC-8, Existential Angst wrote:
> > Inyone have data (or a link), or feel like doing the 'speriment?
>
> Oddly, this cannot usefully be done as a V-versus-I plot.
>
> The resistance of a tungsten filament, measured cold, is very
> different from the resistance when incandescent and hot.
> And, that difference means that a hot filament, lowered
> to 60V, may have different current than a cold filament
> raised to 60V. Try rotating a dimmer clockwise until
> the light comes on, then turn counterclockwise to turn
> it off- the effect is clear.
Yes, but it's from the dimmer, not the filament. I saw a thermal
settling time when I changed the Variac voltage and had to chase the
current setting I wanted.
The rule of thumb I've heard and confirmed was a 12X resistance change
from cold to hot. The current roughly doubles from 25V to 125V.
jsw
DoN. Nichols
> pent...@yahoo.com wrote:
>
> http://imageupload.org/?di=712999347565
> Funny my DNS server shows this as a dead link.
I find that I have it blocked -- because of spam and attacks
against the sshd port. It is in Bahrain, FWIW. Why choose that as an
image server? I'm not going to unblock it just to see these images.
> ----------------------
> "Bob Engelhardt" wrote in message news:ilg16...@news5.newsguy.com...
> That chart shows voltage vs. current. I.e., current on the x
> (horizontal) axis and voltage on the y (vertical). Through all my
> engineering schooling and my own graphing, the x axis is the independent
> variable and the y is the dependent one.
>
> So, is the grapher in the linked chart really thinking about how the
> voltage varies as the current changes? Or is he/she just an
> ass-backwards grapher? Either way it seems ass-backwards to me.
Well -- diode junctions are typically graphed as voltage across
vs current applied -- because the curve gets *very* steep if you are
varying the voltage applied instead.
This may be so done with the lamp filament because of similar
constraints.
Enjoy,
DoN.
--
Remove oil spill source from e-mail
Email: <BPdnic...@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
Jim Wilkins
> > "Bob Engelhardt" wrote in messagenews:ilg16...@news5.newsguy.com...
> > That chart shows voltage vs. current. I.e., current on the x
> > (horizontal) axis and voltage on the y (vertical). Through all my
> > engineering schooling and my own graphing, the x axis is the independent
> > variable and the y is the dependent one.
>
> > So, is the grapher in the linked chart really thinking about how the
> > voltage varies as the current changes? Or is he/she just an
> > ass-backwards grapher? Either way it seems ass-backwards to me.
>
> Well -- diode junctions are typically graphed as voltage across
> vs current applied -- because the curve gets *very* steep if you are
> varying the voltage applied instead.
>
> This may be so done with the lamp filament because of similar
> constraints.
>
> Enjoy,
> DoN.
When the test equipment can control either voltage or current it isn't
so obvious which should be the dependent and independent variable.
jsw
DoN. Nichols
> us use them for precision power measurements as the waveform adopts many
> harmonics in some units.
Are you *sure* about this? I know that constant voltage
transformers (e.g. the Sola units) do introduce significant waveform
distortion, but a Variac (adjustable autotransformer) is simply a
toroid transformer which should not introduce distortion as long as the
input voltage is not significantly in excess of the design voltage, and
the frequency is near the design frequency. (E.g. I would not use a
400 Hz Variac which I have for 60 Hz operation.)
Now -- I guess that intermittent contact at the wiper might
cause some tiny distortion, but nothing like what a Sola CV transformer
(or similar) produces.
> This could drive your accuracy of power testing around the bend is you want
> more than one digit of accuracy. Multiply a distorted current waveform
> measurement times a distorted voltage measurement (from the same voltage
> source) and you can get power than is 10-20% in error very quickly.
So -- if you don't trust a Variac -- how would you trust a
multi-tapped transformer and a switch for producing the varied voltage?
Of course -- the RMS voltage applied should produce the same
result as DC -- assuming that the filament's thermal inertia is
sufficient to keep it from following the AC waveform. If it is changing
temperature that quickly, you have a more serious source of distortion
in the varying load from the filament than I would expect from a Variac
or other variable autotransformer.
DoN. Nichols
> On Mar 12, 5:23 pm, Rich Grise <ri...@example.net.invalid> wrote:
>> Josepi wrote:
>> > Careful. Variacs are known waveform distorters. The government would not
>> > let us use them for precision power measurements as the waveform adopts
>> > many harmonics in some units.
>>
>> Where do you get this information? A variac is nothing but an adjustable
>> autotransformer - what part introduces the distortion?
[ ... ]
> A highly capacitive load, such as a light bulb filament.
Is this a troll?
Jim Wilkins
> On 2011-03-12, Jim Wilkins <kb1...@gmail.com> wrote:
>
> > On Mar 12, 5:23 pm, Rich Grise <ri...@example.net.invalid> wrote:
> >> Josepi wrote:
> >> > Careful. Variacs are known waveform distorters. The government would not
> >> > let us use them for precision power measurements as the waveform adopts
> >> > many harmonics in some units.
>
> >> Where do you get this information? A variac is nothing but an adjustable
> >> autotransformer - what part introduces the distortion?
>
> [ ... ]
>
> > A highly capacitive load, such as a light bulb filament.
>
> Is this a troll?
Bait for one.
dca...@krl.org
> Careful. Variacs are known waveform distorters. The government would not let
> us use them for precision power measurements as the waveform adopts many
> harmonics in some units.
>
Which brings up a couple of points. The temperature of the filament
in vacuum tubes does vary if you have AC applied to it. So you will
get some distortion from that. But most of the better DVM's are true
RMS meters, at least on Voltage. I would suspect they would also
measure true RMS current. But does anyone know for sure?
Dan
Jim Wilkins
This says they do:
http://www.newportelect.com/manuals/ACGUIDE.htm
I think the RMS converter is after the current shunt that turns
current into voltage.
jsw
Josepi
disapproving devices for energy metering inspection.
Variacs can have distortion in the laminations...usually slots for some
reason (heat expansion?) that cause the output waveform to have some ringing
in it. I haven't done any scoping or analysis personally and am paraphrasing
NBS information.. Electrical utility meter testboards are frequently
rejected for usage by Measurement Canada for use of Variacs in their
potential and current supplies. I believe some types may pass depending on
the core shape and configuration.
Any transformer with irregular shaped laminations can cause this. They
usually appear as slots for expansion between the E and I laminations or
mounting holes.
-----------------------
"Rich Grise" wrote in message
news:ilgrr9$s3v$1...@news.eternal-september.org...
Where do you get this information? A variac is nothing but an adjustable
autotransformer - what part introduces the distortion?
Thanks,
Rich
------------------
Josepi
Not sure but I think Variacs are not usually complete toroids but have a
break in the iron.
----------------------------
"DoN. Nichols" wrote in message
news:slrnino6hh.n8...@Katana.d-and-d.com...
Josepi
distortion or phase shift on it's own but RMS amperes x RMS voltage is not
always accurate for power measurements with distorted waveforms.
Yes, most good DVMs, these days, measure a reasonably accurate form of RMS
with various methods.
As Jim's posts says the current is usually measured via a shunt resistance
IR drop.
---------
wrote in message
news:8300c654-282f-4aa6...@e8g2000vbz.googlegroups.com...
Josepi
filament and proceeded to scope current inrush through a 100W bulb for days,
taking pictures of the old storage scope (phosphor memory screen and 35mm
camera OMG!).
We found the best peak was about 10 times the quiescent full-on current. I
forget how long it took to fade to normal current. IIRC It was about 2-10
cycles depending on the mass of the filament and how you hit the start in
the cycle zero crossing.
-----------------
"Jim Wilkins" wrote in message
news:e916ad2c-f8e0-4265...@e9g2000vbk.googlegroups.com...
Yes, but it's from the dimmer, not the filament. I saw a thermal
settling time when I changed the Variac voltage and had to chase the
current setting I wanted.
The rule of thumb I've heard and confirmed was a 12X resistance change
from cold to hot. The current roughly doubles from 25V to 125V.
jsw
---------------
The resistance of a tungsten filament, measured cold, is very
different from the resistance when incandescent and hot.
And, that difference means that a hot filament, lowered
to 60V, may have different current than a cold filament
raised to 60V. Try rotating a dimmer clockwise until
the light comes on, then turn counterclockwise to turn
Jim Wilkins
> Years ago my boss became obsessed with the inrush current on an incandescent
> filament and proceeded to scope current inrush through a 100W bulb for days,
> taking pictures of the old storage scope (phosphor memory screen and 35mm
> camera OMG!).
Another tech did that for fuses. I didn't realize what he was doing
until he asked me to order another case of film. He was really
downfallen and discouraged after I showed him the I^2*T curves in the
Littelfuse catalog.
> "Jim Wilkins" wrote in message
>
> news:e916ad2c-f8e0-4265...@e9g2000vbk.googlegroups.com...
> Yes, but it's from the dimmer, not the filament. I saw a thermal
> settling time when I changed the Variac voltage and had to chase the
> current setting I wanted.
>
> The rule of thumb I've heard and confirmed was a 12X resistance change
> from cold to hot. The current roughly doubles from 25V to 125V.
>
> jsw
>
> ---------------
>
> The resistance of a tungsten filament, measured cold, is very
> different from the resistance when incandescent and hot.
> And, that difference means that a hot filament, lowered
> to 60V, may have different current than a cold filament
> raised to 60V. Try rotating a dimmer clockwise until
> the light comes on, then turn counterclockwise to turn
> t off- the effect is clear.
You quoted what I saw but did you read and understand it? The voltage
corresponding to the setpoint current was the SAME whichever way I
approached it, after the temperature stabilized, which took perhaps
half a second. The lamp dimmer had considerable hysteresis but the
Variac didn't. I used the same equipment to slow down the laminate
trimmer earlier, and they behaved the same way.
I once salvaged a bunch of damaged Powerstats (The Superior Electric
equivalent) to make a few complete ones. The cores are rolls of steel
tape, like duct tape. There are no notches or air gaps. I tried to
make speaker crossovers from them but their response fell off to
useless around 400Hz.
We had built the Powerstats into a large machine for GE. The break
whistle sounded while it was on the forklift there and the union
driver didn't wait to lower it. When they returned it had tipped the
forklift forward and smashed in the front panels on the floor.
jsw
Martin Eastburn
you know they draw 10x on warmup and then at incandescent
temp it comes to the rated wattage.
I find it hard to believe that a VOM doesn't have a current
selection. Older ones didn't have A-C.
So what is the deal - why - and what is the argument....
Martin
On 3/12/2011 3:27 AM, Existential Angst wrote:
> Awl --
>
> Inyone have data (or a link), or feel like doing the 'speriment?
Bob Engelhardt
> ...
> Well -- diode junctions are typically graphed as voltage across
> vs current applied -- because the curve gets *very* steep if you are
> varying the voltage applied instead.
So it's a matter of the aspect ratio of the graph: they'd rather have a
wide & short graph than a narrow & tall one? But that doesn't make
sense - you can have any aspect ratio you want by changing the scales.
And it isn't so much the setup used to determine the graph (e.g.,
varying the voltage & measuring current), as the *function* represented
by the graph. I.e., how one thinks of the device represented. You can
think of a resistor as a device that has a voltage across it as a
function of the current through it. Or as a device that "allows" a
current as a function of the voltage across it. Both are legitimate,
but for a non-linear resistor like a light bulb, I can understand it
more easily if the graph shows current as a function of voltage.
I suspect that for the diode-junction function, current IS the
independent variable. I.e., the current is determined by other factors
and one finds the junction voltage as a function of that current.
Bob
Martin Eastburn
It was a panel meter 0-5 amps. Normal DC shunts could do it for DC,
but for AC ? and RF ?
There was a four terminal 'tube'. There was a heater and a
thermocouple. The meter was measuring the voltage generated by the
thermocouple!
Using this method, any waveform or complex AC riding on or modulated
within an RF signal can be measured.
Martin
Rich Grise
> On Mar 12, 5:23 pm, Rich Grise <ri...@example.net.invalid> wrote:
>> Josepi wrote:
>> > Careful. Variacs are known waveform distorters. The government would
>> > not let us use them for precision power measurements as the waveform
>> > adopts many harmonics in some units.
>>
>> Where do you get this information? A variac is nothing but an adjustable
>> autotransformer - what part introduces the distortion?
>
Huh? Where's the capacitance in a filament?
You need to go to the library and read a basic electricity/electronics book,
or read some basic electricity/electronics tutorials on the internet or
something.
By the way, a triac or SCR lamp dimmer is NOT a variac:
http://www.google.com/images?hl=en&q=define%3Avariac
Good Luck!
Rich
Rich Grise
> On 2011-03-12, Jim Wilkins <kb1...@gmail.com> wrote:
>> On Mar 12, 5:23 pm, Rich Grise <ri...@example.net.invalid> wrote:
>>> Josepi wrote:
>>> > Careful. Variacs are known waveform distorters. The government would
>>> > not let us use them for precision power measurements as the waveform
>>> > adopts many harmonics in some units.
>>>
>>> Where do you get this information? A variac is nothing but an adjustable
>>> autotransformer - what part introduces the distortion?
> [ ... ]
>> A highly capacitive load, such as a light bulb filament.
>
> Is this a troll?
>
Cheers!
Rich
Rich Grise
...
> Careful. Variacs are known waveform distorters. The government would not
> let us use them for precision power measurements as the waveform adopts
> many harmonics in some units.
Oh, yes. And as we all know, governments are infallible fonts of wisdom.
</sarcasm>
Thanks,
Rich
Rich Grise
> Rich Grise wrote:
>> Josepi wrote:
>>
>>> Careful. Variacs are known waveform distorters. The government would not
>>> let us use them for precision power measurements as the waveform adopts
>>> many harmonics in some units.
>>>
>> Where do you get this information? A variac is nothing but an adjustable
>> autotransformer - what part introduces the distortion?
>>
> Variac is a registered trade mark. :-)
Yabbut, so are Kleenex, Windex, and Drano.
Dunno about Raisin Bran, however. ;-D
Cheers!
Rich
Rich Grise
> On 2011-03-12, Josepi <J.R.M> wrote:
>> pent...@yahoo.com wrote:
>>
>> http://imageupload.org/?di=712999347565
>
>> Funny my DNS server shows this as a dead link.
>
> I find that I have it blocked -- because of spam and attacks
> against the sshd port. It is in Bahrain, FWIW. Why choose that as an
> image server? I'm not going to unblock it just to see these images.
>
usually the folks who hang out there are pretty tolerant of off-topic
images, especially if they're kewl. ;-)
Plus, every proper ISP I've had has offered some kind of "personal web
space," where you can post a personal web page, and usually have room to
put up some images, e.g.:
http://mysite.verizon.net/richgrise/images/Me+Gals1.jpg
Cheers!
Rich
Rich Grise
> DoN. Nichols wrote:
>> ...
>> Well -- diode junctions are typically graphed as voltage across
>> vs current applied -- because the curve gets *very* steep if you are
>> varying the voltage applied instead.
>
> wide & short graph than a narrow & tall one? But that doesn't make
> sense - you can have any aspect ratio you want by changing the scales.
No, it's because it's hard to control the voltage with enough precision
to get good resolution on the V/I curve.
Hope This Helps!
Rich
clarkm...@gmail.com
ideal resistor in the steady state, but has a big inrush current.
When we worry about Voltage vs current is in a florescent. That curve
looks like a mountain range. In order to make steps in brightness, I
drive with an AC current source and a slow feed back loop with an opto
diode.
pent...@yahoo.com
Variacs and any transformer can introduce a small amount of
waveform distortion because of the effect of the electrical non
linearity of the iron circuit on the finite resistance of the
primary. It is normally negligible unless the iron is operated at
high flux density - approaching saturation.
The instantaneous magnetising current (no load current) depends
on the permeability of the iron circuit at that current. For
moderate flux densities, permeablity is roughly constant so the
magnetising current is sinusoidal.
However at high flux densities, as the magnetising current
aproaches its peak, the permeablity decreases. This results in
a spiky non sinusoidal magnetising current (3rd harmonic
distortion).
The turns ratio of the Variac or transformer is unaltered so, if
the primary primary resistance is zero, the output waveform will
remain a true sinusoid.
In a practical transformer the primary resistance is not zero so
the non linear magnetising current will cause a small nonlinear
voltage drop in the effective primary voltage. The effect of
this nonlinear voltage drop will be reflected in the output
waveform which will now have a small amount of 3rd harmonic
distortion.
This is a second order effect that can normally be ignored but
may need to be considered in high precision applications.
Jim
Jim
Jim Wilkins
> ....>
> I'm hoping he's merely uninformed.
>
> Cheers!
> Rich
Your sarcasm filter needs cleaning.
jsw
Jim Wilkins
> ...>
> However at high flux densities, as the magnetising current
> aproaches its peak, the permeablity decreases. This results in
> a spiky non sinusoidal magnetising current (3rd harmonic
> distortion).
>
> The turns ratio of the Variac or transformer is unaltered so, if
> the primary primary resistance is zero, the output waveform will
> remain a true sinusoid.
>
I don't have the equipment to measure distortion. This is the
magnetizing current and power factor as measured by a KillAWatt for a
20A Variac fed from the end (120V) and voltage boost (140V) taps.
120V: 0.10A, PF=0.58
140V: 0.13A, PF=0.51
I've seen the waveform distort under load when the output is turned
above 120V in the boost configuration.
jsw
pent...@yahoo.com
<mike.t...@earthlink.net> wrote:
>
>Josepi wrote:
>>
>> Funny my DNS server shows this as a dead link.
>
>
> Then you need a better DNS server.
This link seems to have problems.
Alternates are:-
http://img843.imageshack.us/i/6w15w2.jpg/
http://img534.imageshack.us/i/tungstenlamp.jpg/
.
Recommendations for reliable image hosters would be welcome.
Jim
Jim Wilkins
> ...
> I don't have the equipment to measure distortion. This is the
> magnetizing current and power factor as measured by a KillAWatt for a
> 20A Variac fed from the end (120V) and voltage boost (140V) taps.
> jsw
I suppose that could be confusing if you haven't wired them.
http://www.nmr.mgh.harvard.edu/~reese/VariacPage/
The 120V tap (4) applies input AC across the full winding, the boost
tap (2) is part way in so if connected there the line voltage passes
through fewer turns and (4) delivers about 140V.
jsw
Jim Wilkins
> ...>
> Recommendations for reliable image hosters would be welcome.
> Jim
You could try Picasa, which works without the downloaded software.
Photobucket clashes with my high security settings, as do Enco and
MSC.
jsw
Jim Wilkins
> ...
> </sarcasm>
And you complained about my blatant, self-evident absurdity.
jsw
Larry Jaques
Wow, Seventies Week!
--
Whomsoever controls the volume of money in any country is
absolute master of all industry and commerce and when you
realize that the entire system is very easily controlled,
one way or another, by a few powerful men at the top, you
will not have to be told how periods of inflation and
depression originate. --James Garfield
Bob Engelhardt
[nice explanation]
> This is a second order effect that can normally be ignored but
> may need to be considered in high precision applications.
Thanks.
Another good example of the nature of the 2nd Information Revolution
(the 1st being the printing press-based one). Prior to the Internet,
the way such information was communicated was through books or a
professional environment. It's unlikely that any of us would have gone
to the considerable trouble to find this information in a library, or
have colleagues who would know. Yet, here we are, well informed on a
sophisticated subject.
In awe of it all,
Bob
Lewis Hartswick
> On Mar 12, 5:23 pm, Rich Grise <ri...@example.net.invalid> wrote:
>>> Careful. Variacs are known waveform distorters. The government would not
>>> let us use them for precision power measurements as the waveform adopts
>>> many harmonics in some units.
>> autotransformer - what part introduces the distortion?
>>
>> Rich
>
> A highly capacitive load, such as a light bulb filament.
>
get on the internet.
"capacitive load such as a light bulb filament"
Even if you take into account the inductance of the
coil of the filament it would be inductive but for
any thing short of the most severe scientific details
a filament is a pure resistive (although very non-
linear) load. Sheesh!
...lew...
Lewis Hartswick
> On 2011-03-12, Jim Wilkins <kb1...@gmail.com> wrote:
>> On Mar 12, 5:23 pm, Rich Grise <ri...@example.net.invalid> wrote:
>>> Josepi wrote:
>>>> Careful. Variacs are known waveform distorters. The government would not
>>>> let us use them for precision power measurements as the waveform adopts
>>>> many harmonics in some units.
>>> Where do you get this information? A variac is nothing but an adjustable
>>> autotransformer - what part introduces the distortion?
>
>
>> A highly capacitive load, such as a light bulb filament.
>
>
I think it's just "Blather" by some idiot.
...lew...
Josepi
Either is "Moby Dick"
-------------------------------
"Rich Grise" wrote in message
news:ilhpb5$kkl$9...@news.eternal-september.org...
Lewis Hartswick
>
> You need to go to the library and read a basic electricity/electronics book,
> or read some basic electricity/electronics tutorials on the internet or
> something.
> Rich
>
Rich the internet is a good source of just exactly the kind
of information he is spouting. :-)
...lew...
Josepi
Trouble with those units, from a power metering standpoint is the scale.
Thermal is run by power and the scale become logarithmic. Constant
complaints of reading not accurate below the first line of indication
welcomed the digital era.
Thermal is the "real good" method of getting true RMS values. There has been
too many cheap method so indicating these values, otherwise. Big item is to
understand your instruments.
-----------
"Martin Eastburn" wrote in message
news:lzXep.475812$ZM.4...@en-nntp-02.dc1.easynews.com...
Michael A. Terrell
pent...@yahoo.com wrote:
>
> On Sat, 12 Mar 2011 19:39:02 -0500, "Michael A. Terrell"
> <mike.t...@earthlink.net> wrote:
>
> >
> >Josepi wrote:
> >>
> >> Funny my DNS server shows this as a dead link.
> >
> >
> > Then you need a better DNS server.
>
> This link seems to have problems.
I had no problem with it. I use Open DNS, becasue the Earthlink DNS
server dies from time to time and I get their error page.
> Alternates are:-
>
> http://img843.imageshack.us/i/6w15w2.jpg/
>
> http://img534.imageshack.us/i/tungstenlamp.jpg/
>
> .
>
> Recommendations for reliable image hosters would be welcome.
>
> Jim
--
You can't fix stupid. You can't even put a Band-Aid™ on it, because it's
Teflon coated.
Wild_Bill
accessed the first image at the imageshack site.
For anything that's worthwhile and/or relevant to be archived, the dropbox
would be a good place to save the info.. and that way it can be linked to in
the future
http://www.metalworking.com/dropbox/__how_to_use_dropbox.html
--
WB
.........
<pent...@yahoo.com> wrote in message
news:b0hpn6llbsu2i8puf...@4ax.com...
Jim Wilkins
For the humor-impaired it was a facetious guess of why a Variac would
distort a sine wave.
http://dictionary.reference.com/browse/facetious
jsw
DoN. Nichols
> See my other post.
Not hit that one yet. Perhaps later in the thread since my
newsreader follows threads in order of posting within a branch.
> Not sure but I think Variacs are not usually complete toroids but have a
> break in the iron.
All Variac cores that I have seen are multi--turn tape-wound
toroids. No gap. Actually a much more efficient magnetic structure
than the typical transformer made of 'E' and 'I' laminations. (And the
better of those alternate sides for the 'E's and 'I's to minimize gap in
the construction.) I've taken enough of those apart when a Kid.
The winding of the copper wire is typically on something like
270 degrees of the core -- leaving room for the wiper pickoff ring, and
providing reasonable separation between the full CCW and full CW ends of
the coil (where the maximum voltage differential is).
Enjoy,
DoN.
--
Remove oil spill source from e-mail
Email: <BPdnic...@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
DoN. Nichols
> On Mar 12, 2:02 pm, "Josepi" <J.R...@greasynews.calm> wrote:
>> Careful. Variacs are known waveform distorters. The government would not let
>> us use them for precision power measurements as the waveform adopts many
>> harmonics in some units.
>>
>
> in vacuum tubes does vary if you have AC applied to it.
Perhaps on the lightweight filaments once common in
battery-powered radios and such, but most normal vacuum tubes had the
filament stuffed inside a hollow metal tube which was heated by the
filament and which served as the cathode (emitter of electrons). The
mass of this assembly was such that the variation in temperature was
damped out.
For that matter -- things like the 80 rectifier tubes, the
filament was a ribbon of metal perhaps 1/8" wide, so even without
separate cathodes, the thermal mass was rather high considering a 60 Hz
AC current.
You might get more distortion from the voltage difference
between the ends of the filament and the nearby grid on tubes without a
separate cathode.
> So you will
> get some distortion from that. But most of the better DVM's are true
> RMS meters, at least on Voltage. I would suspect they would also
> measure true RMS current. But does anyone know for sure?
Since the DVM measures the AC voltage developed across a low
resistance (perhaps amplified) to measure current, the same circuit
which measures RMS voltage in the multimeter is measuring the RMS
current.
But not all DVMs are capable of measuring true RMS -- only the
more expensive ones. Some fake it by integrating the signal over a full
cycle of the AC -- which is fine as long as you are using the frequency
for which it is designed. Change the frequency (e.g. to 50 Hz or 400 Hz
for a 60 Hz designed one) and you are integrating over something other
than an integer number of cycles, thus introducing errors.
Josepi
We were told that variac were bad due to slots in the laminations or some
kind of irregularity in the iron that caused some harmoinic distortion.
Unfortunately I cannot provide a document from Measurement Canada with
specifics. I only remember that revenue meter calibration testboards would
not pass their qualifications due to variacs used in the circuits. (so we
were told then??) This may be too minor to consider here. I am not sure of
the magnitude of the alleged distortion. They were looking for 0.1% accuracy
and nobody here would ever look that closely (especially using a lightbulb
wattmeter....LOL).
The distortion affected the outcome of certain metering devices. Now there
was some stupid meter units out there for a while too using rectified
voltages and currents (especially) to meter kVA (recti-thermal meters). When
you put 40 elements, each with their own full wave bridge rectifier, in
series, with a low voltage test supply...of course you get distortion! You
get a huge notch of no conduction up to 80 silicon diode junction drops. But
I digress. Just a funny thing to mention since you seem to be an electrical
(And share a similar name with an old inspector)
--------------------
"DoN. Nichols" wrote in message
news:slrninqs6g.k1...@Katana.d-and-d.com...
Josepi
it would get you so the expression became popular.
"Know your equipment."
The guys used to laugh when I would get out the old AVO analogue needle
meter, with the mirrored scale, from time to time but the digital ones (with
all their so-called accuracy) would get confused in certain applications.
"DoN. Nichols" wrote in message
news:slrninqt77.k1...@Katana.d-and-d.com...
Winston
Hey Jim-
Did you mean 'thermally capacitive' in that the resistance
change of the filament lags the voltage change across it?
--Winston
Jim Wilkins
> ...>
> Did you mean 'thermally capacitive' in that the resistance
> change of the filament lags the voltage change across it?
>
No, I meant it in the way that black is white, war is peace and I love
Big Brother.
jsw
Winston
You got me. I thought you were the real Jim. D'oH!
--Winston
Jim Wilkins
Nope, I'm his bitterly sarcastic alter ego who has been arguing with a
couple of unrepentent neo-Nazis in another forum.
Larry Jaques
wrote:
I think he got into the kid's windowpane again.
Sunworshiper, OTOH, sounds like he's on blotter lately.
--
Small opportunities are often the beginning of great enterprises.
-- Demosthenes
Jim Wilkins
wrote:
> ...>
> I think he got into the kid's windowpane again.
> Sunworshiper, OTOH, sounds like he's on blotter lately.
"If not lasers, dope or not, anyone caught crossing illegally should
be
given the 50 round test. Anyone surviving 50 rounds from a gat or
squad machine gun (as well as collectible sized parts of the rest)
should be deported. "
And you call ME high???
DoN. Nichols
> Without looking further
>
> We were told that variac were bad due to slots in the laminations or some
> kind of irregularity in the iron that caused some harmoinic distortion.
O.K. There could be a an air gap between layers of the wound
magnetic material -- and it could even vibrate under varying magnetic
forces, so a tiny amount of distortion is possible under those
circumstances. Frequency dependent, because of possible mechanical
resonances in the structure formed by the tape-wound core.
> Unfortunately I cannot provide a document from Measurement Canada with
> specifics. I only remember that revenue meter calibration testboards would
> not pass their qualifications due to variacs used in the circuits. (so we
> were told then??) This may be too minor to consider here. I am not sure of
> the magnitude of the alleged distortion. They were looking for 0.1% accuracy
> and nobody here would ever look that closely (especially using a lightbulb
> wattmeter....LOL).
Yes -- I could see 0.1% distortion being possible under the
circumstances. And if that is important, yes avoid the Variacs --
unless the cores were wound with a goop of epoxy between layers, and
that is cured without air bubbles being entrained. Once the magnetic
tape is mechanically stabilized, I don't see much likelihood of
significant distortion from that. The *good* transformers in the old
days were in metal cans which were then filled with melted tar -- and
likely vacuum cycled to remove trapped air -- so the chance of changing
magnetic structure was rather minimized.
And, of course, instrumentation for measuring distortion in high
quality audio devices would be highly unlikely to include Variacs or the
like in the measurement loop -- since they tend to be optimized for
particular frequencies. Most that I have are for 60 Hz (and will
tolerate 50 Hz with a little less headroom), but I do have at least one
400 Hz one somewhere. (Finding it might be tricky these days. :-)
As for using a lightbulb as a Wattmeter -- even measuring the
illumination in an enclosed space using a good photographic light meter
would not get you to that kind of sensitivity.
And about equal sensitivity with two blocks of paraffin (US use
-- white wax, not kerosene) separated by an opaque sheet (metal) with
the two light sources on either side and equally spaced from the
paraffin. (This is the way that light outputs were once compared.)
> The distortion affected the outcome of certain metering devices. Now there
> was some stupid meter units out there for a while too using rectified
> voltages and currents (especially) to meter kVA (recti-thermal meters). When
> you put 40 elements, each with their own full wave bridge rectifier, in
> series, with a low voltage test supply...of course you get distortion!
Of course. Couple the diode effects with the possible reactance
seen through the bridge and you have even more likelihood for
unpredictable distortions.
If you want to minimize that -- tap off the AC and do the
rectification using an op-amp rectifier, where the diode curve is
removed by the feedback loop and careful design.-
> You
> get a huge notch of no conduction up to 80 silicon diode junction drops. But
> I digress. Just a funny thing to mention since you seem to be an electrical
> (And share a similar name with an old inspector)
Intersting. I'm old, but not an inspector. :-)
Many years as an electronics technician, (lots of design work at
home as well) followed by some years as a unix network administrator
prior to retiring.
Winston
> On Mon, 14 Mar 2011 14:21:05 -0700, Winston<Win...@BigBrother.net>
> wrote:
>
>> Jim Wilkins wrote:
>>> On Mar 14, 4:46 pm, Winston<Wins...@BigBrother.net> wrote:
>>>> ...>
>>>> Did you mean 'thermally capacitive' in that the resistance
>>>> change of the filament lags the voltage change across it?
>>>>
>>>> --Winston-
>>>
>>> No, I meant it in the way that black is white, war is peace and I love
>>> Big Brother.
>>
>> You got me. I thought you were the real Jim. D'oH!
>
> I think he got into the kid's windowpane again.
> Sunworshiper, OTOH, sounds like he's on blotter lately.
That is the downside of USENET 'instant communication'.
I find it too easy to press the 'send' key without allowing
any time for reflection and correction.
This sometimes occasions inadvertent obfuscation.
D'oH!
--Winston
Existential Angst
news:ilhpp7$kkl$1...@news.eternal-september.org...
> DoN. Nichols wrote:
>> On 2011-03-12, Josepi <J.R.M> wrote:
>>> pent...@yahoo.com wrote:
>>>
>>> http://imageupload.org/?di=712999347565
>>
>>
>> against the sshd port. It is in Bahrain, FWIW. Why choose that as an
>> image server? I'm not going to unblock it just to see these images.
>>
> What about that "dropbox" thingie? I use alt.binaries.schematics.eletronic
> ;
> usually the folks who hang out there are pretty tolerant of off-topic
> images, especially if they're kewl. ;-)
>
> Plus, every proper ISP I've had has offered some kind of "personal web
> space," where you can post a personal web page, and usually have room to
> put up some images, e.g.:
> http://mysite.verizon.net/richgrise/images/Me+Gals1.jpg
Dewd, izzat YOU?? Can I be you????
What year was that taken?
--
EA
>
> Cheers!
> Rich
>
Rich Grise
> "Rich Grise" <ri...@example.net.invalid> wrote in message
>>> On 2011-03-12, Josepi <J.R.M> wrote:
>>>> pent...@yahoo.com wrote:
>>>>
>>>> http://imageupload.org/?di=712999347565
>>>
>>>> Funny my DNS server shows this as a dead link.
>>>
>>> I find that I have it blocked -- because of spam and attacks
>>> against the sshd port. It is in Bahrain, FWIW. Why choose that as an
>>> image server? I'm not going to unblock it just to see these images.
>>>
>> What about that "dropbox" thingie? I use
>> alt.binaries.schematics.eletronic ;
>> usually the folks who hang out there are pretty tolerant of off-topic
>> images, especially if they're kewl. ;-)
>>
>> Plus, every proper ISP I've had has offered some kind of "personal web
>> space," where you can post a personal web page, and usually have room to
>> put up some images, e.g.:
>> http://mysite.verizon.net/richgrise/images/Me+Gals1.jpg
>
> Dewd, izzat YOU?? Can I be you????
> What year was that taken?
I think it was in the 1980's to early 1990's. I was going to a "church"
called the "Intuitive Arts Fellowship" in Cardiff-by-the-Sea, CA, which
was a very touchy-feely, new-agey, "science of mindey" thing - after
the "sermon," they had a couple of psychics who would give readings. The
hot babes were all from Hollywood - A couple of them were actresses, the
hot babe on my immediate right whose boobs I'm staring at was a Ventura or
Venice street performer who did a fire-breathing act until she got
seriously burned and had to retire. the pic was at a "retreat" in
Idyllwild, it cost about a hundred bucks a person to go to this sort of
lodge thingie up in the wooded foothills.
Since then, my beard has gone white, I've lost some weight, and I lost the
great gig I had at the time; these days probably none of those babes would
give me a second look. The point of the pic was, I said, "Hey, how about a
pic of me with you gals, to show around to my friends and make them
envious?" and they said, "OK." The one to my immediate left was an actress
who had read for some part on Star Trek: TNG; she had met Johnathan Frakes,
and when I asked her about him, she got stars in her eyes and said, "Oh,
he's just a love." The two on the ends were producers and stuff. The only
one whose name I remember was Gigi, the fire-breather. I heard about her
serious burns at "church" one morning; about 3 or 5 years later, she was
featured on the teevee show "How'd they do that?"
http://www.imdb.com/title/tt0158661/
showing how they'd surgically restored her face which had been practically
burned off in her fire-breathing accident.
Can you be me? Sure! Take my identity, please! You would then owe the IRS
about $30,000 dollars! In exchange, can I have your job?
Cheers!
Rich
Josepi
- The offending "variacs" may have been cheaper brands. Everything I have
researched tells absolutely no distortion and complete toroid construction.
I have to withdraw that comment, apparently...LOL I never experienced it or
witnessed it personally. Our systems always passed other than the usual
minor winding ratio errors. (you can't get a half a turn)
- The current circuit would be a little hard to op amp out diode drops
(precision rectifier), as you mentioned. The currents were up to 200 amperes
for testing. We couldn't lift the heat sink and refrigeration unit attached
to that 741...LOL
Have a great one!
---------------
"DoN. Nichols" wrote in message
news:slrnintn1k.ho...@Katana.d-and-d.com...
DoN. Nichols
> Two items
> - The offending "variacs" may have been cheaper brands. Everything I have
> researched tells absolutely no distortion and complete toroid construction.
> I have to withdraw that comment, apparently...LOL I never experienced it or
> witnessed it personally. Our systems always passed other than the usual
> minor winding ratio errors. (you can't get a half a turn)
It still could have been that the higher-ups *believed* it,
which would have been sufficient to cause them to ban such transformers.
The ones by Superior Electric (Powerstat name IIRC) were likely as good
as the General Radio Variacs. And back the, they probably would have
had few other choices in variable autotransformers.
> - The current circuit would be a little hard to op amp out diode drops
> (precision rectifier), as you mentioned. The currents were up to 200 amperes
> for testing. We couldn't lift the heat sink and refrigeration unit attached
> to that 741...LOL
Build in a shunt to give 50 mV (common shut output voltage to
meter movements), amplify that with one op-amp to get enough voltage to
make the amplified rectifier work with minimal distortion.
Jim Wilkins
> ...
> > - The current circuit would be a little hard to op amp out diode drops
> > (precision rectifier), as you mentioned. The currents were up to 200 amperes
> > for testing. We couldn't lift the heat sink and refrigeration unit attached
> > to that 741...LOL
>
> Build in a shunt to give 50 mV (common shut output voltage to
> meter movements), amplify that with one op-amp to get enough voltage to
> make the amplified rectifier work with minimal distortion.
>
> Enjoy,
> DoN.
AC line currents are usually sensed with current transformers:
http://www.crmagnetics.com/
The power wire through the center is the primary, the secondary
current will be reduced by the turns ratio, ie with 1000 turns 1Amp
through the middle gives 1mA out. The output leads have to be
connected across a current meter or low value sensing resistor etc or
else the open-circuit voltage will jump very high. Commonly they are
wound to use a 5A ammeter graduated for whatever the full line current
would be.
http://www.crmagnetics.com/Support/TechnicalReference/tabid/62/ctl/Details/mid/372/ItemId/0/Default.aspx
There are small ones in Ground Fault outlets:
http://www.o-digital.com/uploads/2179/2225-2/Current_Transformer_for_GFCI_522_ZG522_656.jpg
They are connected such that the hot and neutral currents cancel and
only a difference between them such as leakage to ground appear on the
output.
jsw
Lewis Hartswick
>
> The guys used to laugh when I would get out the old AVO analogue needle
> meter, with the mirrored scale, from time to time but the digital ones
> (with all their so-called accuracy) would get confused in certain
> applications.
>
should be on the analog meter side of that discussion.
Those GD blinking digits, you cant tell if the reading
is going up or down.
...lew...
Jim Wilkins
I know, the analog meter feels easier, but you can read it to perhaps
1 part in 100 while the digital ones resolve 1 part in 1999, the
better ones 1 in 19,999, and if they let you play around inside the
+/- indicator can be zeroed to half a count or less.
jsw
Pete Snell
> I know, the analog meter feels easier, but you can read it to perhaps
> 1 part in 100 while the digital ones resolve 1 part in 1999, the
> better ones 1 in 19,999, and if they let you play around inside the
> +/- indicator can be zeroed to half a count or less.
Hey Jim, here's a question. I presume you have 'played around
inside', do you have a feel for how much drift a good DMM has over its
range? i.e. if you 'zero' it, is it still within a half count at the
full scale? Past experience has shown me that things start to fall apart
a bit near end of the scale, but are pretty good till the last 10% or so.
Pete
--
Pete Snell
Department of Physics
Royal Military College
Kingston, Ontario,
Canada
-----------------------------------------------------------------------
"Never tell people how to do things.
Tell them what to do and they will surprise
you with their ingenuity."
General George S. Patton (1885-1945)
Jim Wilkins
> On 3/16/2011 2:50 PM, Jim Wilkins wrote:
>
> Hey Jim, here's a question. I presume you have 'played around
> inside', do you have a feel for how much drift a good DMM has over its
> range? i.e. if you 'zero' it, is it still within a half count at the
> full scale? Past experience has shown me that things start to fall apart
> a bit near end of the scale, but are pretty good till the last 10% or so.
You can't really assume any better accuracy than the manufacturer's
specs and then only if it's within calibration - and the cal was done
carefully. DMMs use a dual slope integrator that relies on the
stability and lack of dielectric absorption of a good stable
capacitor. AFAIK the auto-zero circuit does perform well. Published
results shouldn't assume more accuracy than the calibration
guarantees, no matter how clever the lab tech may be.
I haven't been inside one since the 80's when I worked for a genius
Ph.D formerly from Keithley Instruments and with his advice built a
4-1/2 digit meter using one of these:
http://www.intersil.com/products/deviceinfo.asp?pn=ICL7135
The application notes are useful.
Since then I've built a more precise instrument using an Analog
Devices module
http://www.analog.com/en/analog-to-digital-converters/ad-converters/products/index.html
and an extremely fast one with Tektronix digital scope A/D converters,
but they were self-contained building blocks where the accuracy didn't
depend much on my actions.
jsw
DoN. Nichols
Yes. Some of the digital multimeter, e.g. the Fluke ones which
I have, have a bargraph approximation of a needle. Not as much
resolution, but a fairly good guide until you get near the very fine
detail part of the job.
If you want to really be driven nuts by a digital, look at the
really old NLS (Non Linear Systems) DVMs. (From the late 1950s.) A
Kelvin-Varley divider, switched by a collection of (four) stepper
switches. One bank of each stepper illuminates a single lamp (of the
327 style) which illuminates the edge of a sheet of Lucite with a digit
engraved in it. This results in a single digit visible with a bunch of
invisible ones which are not illuminated. The stack of sheets is thick
enough so the different digits have a very visible depth offset.
When The voltage being measured is creeping up, there will be a
tick ... tick ... tick as it tracks. However, when the voltage is
creeping *down*, since the steppers can only move in one direction,
there is a brapp brapp brapp brapp as all four steppers reset to zero
position and then step up -- at about twenty steps per second, so there
is a several second delay before the divider again matches the input
voltage -- until it creeps down yet another step leading to another
series of brapps.
For tuning for a peak or a null, it is a nightmare. The case of
the thing is loaded with sound deadening foam rubber -- but when you
have one opened up to work on it, it is *very* noisy. :-)
John
If you are tuning to a peak or null an analog meter is the easiest to
use. If you are tuning to a set value the digital meter works best.
Some digitals have the bar under the digits which makes it somewhat
similar to the analog meter.
John
Josepi
calibration and are injected with up to a 200 ampere current. 40 diode
junction drops distorts the crap out of the current (I) waveform. You would
need a precision rectifier circuit with a 200 ampere output and it usually
consumes about 3 kVA...LOL
----------------
"DoN. Nichols" wrote in message
news:slrnio0cuv.5p...@Katana.d-and-d.com...
Josepi
digits you want but when they are out by 20 or 50% of the reading sometimes
the analogue meters did a better job.
This is usually a problem with digital updates not syncronizing or actually
in synch with a bouncing signal. The averaging is missing on the digital
instrument. Use a strobe on a rotating part and when it is in synch the mark
on the engine is always at the top. In reality it is turning and all the way
around but the sampling give a false impression that the mark is always at
the top. This is analogous to some of the simpler errors on a digital meter.
"Know your instruments"
----------------
"Jim Wilkins" wrote in message
news:76af5860-0100-48ab...@l2g2000prg.googlegroups.com...
Josepi
They can be 50% out at times.
----------------------
"Lewis Hartswick" wrote in message
news:SIudnT2bk_91Zh3Q...@earthlink.com...
Jim Wilkins
> The point was in certain applications the digital meter can have all the
> digits you want but when they are out by 20 or 50% of the reading sometimes
> the analogue meters did a better job.
>
> This is usually a problem with digital updates not syncronizing or actually
> in synch with a bouncing signal. The averaging is missing on the digital
> instrument. Use a strobe on a rotating part and when it is in synch the mark
> on the engine is always at the top. In reality it is turning and all the way
> around but the sampling give a false impression that the mark is always at
> the top. This is analogous to some of the simpler errors on a digital meter.
>
> "Know your instruments"
Analog meter movements have inertia and mechanical resonance issues
too. Neither type is reliably accurate for rapidly varying signals,
they just fail differently. The right method is to capture the signal
on a digital storage scope and read the signal level at the point you
want with the cursor, or calculate RMS or whatever with the math
functions.
I've got a nice 1 Gig HP, but a PC sound card can be hacked into an
audio frequency digital storage scope if you need one and have more
time than money:
http://www.zeitnitz.de/Christian/scope_en
If you add a level shifter you can inject the signal after the input
blocking cap and make it read down to DC.
jsw
Rich Grise
> The point was in certain applications the digital meter can have all the
> digits you want but when they are out by 20 or 50% of the reading
> sometimes the analogue meters did a better job.
>
> This is usually a problem with digital updates not syncronizing or
> actually in synch with a bouncing signal. The averaging is missing on the
> digital instrument. Use a strobe on a rotating part and when it is in
> synch the mark on the engine is always at the top. In reality it is
> turning and all the way around but the sampling give a false impression
> that the mark is always at the top. This is analogous to some of the
> simpler errors on a digital meter.
>
Or the wagon wheels turning backwards in the old westerns. ;-)
Cheers!
Rich
Michael A. Terrell
I've never seen a 4 1/2 digit analog equivalent, or an analog meter
that displays a .01 dB signal change.
--
You can't fix stupid. You can't even put a Band-Aid� on it, because it's
Teflon coated.
Michael A. Terrell
Not always true.
Michael A. Terrell
HP made a DVM plugin for their 5245L counter that used 'end view'
Nixie tubes.
Michael A. Terrell
Josepi wrote:
>
> Digital meters are not good for indicating fast changing signals.
>
> They can be 50% out at times.
What kind of crap do you use?
Jim Wilkins
wrote:
> Josepi wrote:
>
> > Digital meters are not good for indicating fast changing signals.
>
> > They can be 50% out at times.
>
> What kind of crap do you use?
>
51" and an analog multimeter set on 15VAC. Down to 20Hz the needle is
stable at 10VAC. At 10Hz it wobbles about a volt, ~2V at 5Hz, and
below 3Hz it goes crazy, swinging nearly full scale. Below 1.5Hz it
makes a noise when the needle bangs into the stop.
On DC, swinging from 0 to 4V at 0.1Hz, the response falls off fast
above 1Hz and 10Hz looks like 2VDC with a slight wiggle.
It's so old it uses 709 op amps, and it doesn't have a front panel DC
offset control, just a small trim in back.
jsw
John
I once worked with an analog meter with mirror scale that was huge. I
don't remember exactly how large it was but it sure was easy to read.
It had to be at least 10 inches across the scale face.
John
DoN. Nichols
> "DoN. Nichols" wrote in message
> news:slrnio0cuv.5p...@Katana.d-and-d.com...
> Build in a shunt to give 50 mV (common shut output voltage to
> meter movements), amplify that with one op-amp to get enough voltage to
> make the amplified rectifier work with minimal distortion.
> I don't think you understand. The diodes are inside the meters under
> calibration and are injected with up to a 200 ampere current. 40 diode
> junction drops distorts the crap out of the current (I) waveform. You would
> need a precision rectifier circuit with a 200 ampere output and it usually
> consumes about 3 kVA...LOL
I was thinking in terms of a shunt in *place* of the meter --
200 A and 50 mV equals 250 mOhms -- and a maximum dissipation of 10 kVA.
Make the shunt a permanent part of the circuit.
Take the signal from that shunt, amplify it with one op-amp, and
then use a precision rectifier circuit to convert that to a DC drive for
whatever kind of meter you want -- analog or digital.
Your 40 diode drops suggests that you are measuring the voltage,
not the current -- or you are measuring both in a wattmeter. If that is
the case, a few more op amps -- to scale the voltage and to multiply the
processed voltage reading by the processed current reading. Or -- if
you are after instantaneous power, don't convert to DC at all -- but
your analog meters aren't going to give you that either.
If you aren't familiar with the precision rectifier circuit(s),
here is a nice discussion of them:
<http://sound.westhost.com/appnotes/an001.htm>
Granted, it's (the web site's) focus is audio applications, but
that should not matter here. Anything good enough for audio bandwidths
is more than sufficient for whatever distortion is likely in power
applications (unless you are running wired wireless on top of the power
line. :-)
DoN. Nichols
>
> Josepi wrote:
>>
>> Digital meters are not good for indicating fast changing signals.
>>
>> They can be 50% out at times.
>
>
> What kind of crap do you use?
Note that he said "fast changing".
It is more a question of how fast is the signal changing.
Remember that most digital meters measure AC by sampling at a specific
point in time (often timed to match the power line frequency), or by
integrating over a full cycle to get an approximation of RMS. If the
signal is changing more rapidly than that (including significant
distortion levels), your readings can be quite far out. Digital is not
a magic cure-all. It gives more resolution more quickly than analog
meters -- or things like the really old Fluke differential voltmeters,
where you dial in decade voltage steps to balance out the signal under
test (for DC signals only -- lots of precision, but *very* slow.) And
actually -- early digital voltmeters (e.g. the NLS brand ones which I
described in another post in this thread) were an attempt to automate
this nulling process.
Ecnerwal
John <amd...@intergrafix.net> wrote:
> I once worked with an analog meter with mirror scale that was huge. I
> don't remember exactly how large it was but it sure was easy to read.
> It had to be at least 10 inches across the scale face.
>
> John
I used to have a lovely ammeter with a low-mass needle - there was a
mirror on the sensing coil, and a light beam was projected onto it and
reflected to be the "needle". The scale was a good 8 inches across. By
the time I was using it it was a historical oddity, but I suspect it was
a very significant purchase when it was new.
--
Cats, coffee, chocolate...vices to live by
Michael A. Terrell
"DoN. Nichols" wrote:
>
> On 2011-03-17, Michael A. Terrell <mike.t...@earthlink.net> wrote:
> >
> > Josepi wrote:
> >>
> >> Digital meters are not good for indicating fast changing signals.
> >>
> >> They can be 50% out at times.
> >
> >
> > What kind of crap do you use?
>
> Note that he said "fast changing".
Then he needs a true RMS voltmeter that integrates the input before
reading the level. Something like the Fluke 8920A or a Boonton 9200
digital RF milivoltmeter, instead of a $4 meter from Harbor Freight.
The Fluke is good to 20 MHz and the boonton is good from 10 MHz into
the GHz range.
> It is more a question of how fast is the signal changing.
> Remember that most digital meters measure AC by sampling at a specific
> point in time (often timed to match the power line frequency), or by
> integrating over a full cycle to get an approximation of RMS. If the
> signal is changing more rapidly than that (including significant
> distortion levels), your readings can be quite far out. Digital is not
> a magic cure-all. It gives more resolution more quickly than analog
> meters -- or things like the really old Fluke differential voltmeters,
> where you dial in decade voltage steps to balance out the signal under
> test (for DC signals only -- lots of precision, but *very* slow.) And
> actually -- early digital voltmeters (e.g. the NLS brand ones which I
> described in another post in this thread) were an attempt to automate
> this nulling process.
I have a rack mount Fluke differential meter I picked up for parts.
Days that it is useful on a production bench are long past.