AC Voltage Reference

I forgot to mention anything about power.

Basically, this gets hooked up to a transformer, rectified, filtered, voltage divided, and sampled by a microcontroller.

On a sunny day (Mon, 22 Dec 2008 13:53:12 -0800 (PST)) it happened Mio wrote in :

There are also transductors... A wild idea: What if you have say 200V AC, add a little 50 or 60 hz audio power amp (via one 1:X transformer in series, to make it max 250V AC. then you can feed that amp a variable 50 or 60 Hz (EU US). Measure the output voltage, and use it to control an input attenuator to the map, could be a simple junction FET. Note the phase... So basically: Supply only the bit of difference you expect to regulate, not the full range, and use standard components.....

So the power requirement is quite low?

How about:

signal generator -> power amp -> transformer run backwards.

I don't know if you can rely on a normal audio power amp to have gain constant to 1% - but I expect others here do. Otherwise you could make something, look for high power opamps etc.

--

John Devereux

I've used a power audio amplifier and a step-up transformer to generate stable 120 to 480 volts AC, for calibrating electric power meters. The signal source should be stable, like a function generator or a small benchtop ARB. I suppose you could do overall feedback, too, if you really need a stiff output.

One rig we have is a Peavey 400 watt RMS/channel stereo amp, which we modified to improve low-frequency response. One channel drives a step-up transformer to simulate line voltage, and one steps down to drive current transformers up to 200 amps. The signal source is a home-made thing that uses a uP and a couple of dacs to generate precise phase-controlled sine waves.

John

"Mio" snipped-for-privacy@gmail.com

I forgot to mention anything about power.

Basically, this gets hooked up to a transformer, rectified, filtered, voltage divided, and sampled by a microcontroller.

** So you are going for a world's record in pointless and futile circuit design ?

This loon would have to be the Google Groper f****it of the year.

..... Phil

Huh? What's the point in doing all that? What are you actually trying to do/achieve?

Dave.

We swap out components pretty often, and have to recalibrate based on the differences in transformers/resistors/diodes/etc. As our line voltage can vary by a lot, I was just trying to think of an easy way to do the recalibration.

"Mio" "Phil Allison"

We swap out components pretty often, and have to recalibrate based on the differences in transformers/resistors/diodes/etc.As our line voltage can vary by a lot, I was just trying to think of an easy way to do the recalibration.

** Makes absolutely no sense whatsoever.

Surely you can measure the mains voltage at any time and calibrate your device based on that reading.

Be aware the mains voltage is not a very good sine wave so the peak and rms values are not precisely related by the sq.rt of 2.

..... Phil

AC to DC and then DC to AC via an inverter, with a tight feedback loop based on a stable reference.

Variac ?

You can get a motorized type to have it controlled to maintain voltage. That's about as simple as it gets..

"

RMS? Peak? Limits on harmonics/THD? There's a lot more to defining precision for an AC voltage than one number.

The problem will be compounded by the type of load(s) you are driving. Anything rectified can present a highly nonlinear load to your supply, distorting its waveform. Unless this is taken into consideration in your inverter design (which is a non-trivial problem).

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
The only tools one needs in life:
WD-40 to make things go and duct tape to make them stop.

Hey, year's not over yet.

On a sunny day (Mon, 22 Dec 2008 15:23:24 -0800 (PST)) it happened Mio wrote in :

Here is an other transformer-less idea to make a stabilised AC voltage: ftp://panteltje.com/pub/AC_stabiliser_small.jpg So, got fed up with ASCII art, this works too.

The idea is that if no change is needed in the output, then the audio amp does nothing (zero volts out). If more is needed it adds some in phase voltage, and if less is needed, then it adds some 180 degrees opposite phase voltage. The only transformer you need is a very small one to supply the op amps. This will not work if there is DC flowing in the load. In that case you need a small low voltage 1:1 transformer in the output of the audio amp.

The J-FET works as variable resistor, when in the middle of range the sum of the voltages at the audio amp input is zero, when its impedance is lower, then the phase is -180 degrees, when higher +180 degrees at that input.

here's another:

on-line sine-wave UPS followed by a step-up transformer or better, a 230V sine-wave on-line UPS followed by a

230V to 224V autotransformer (can be made from a 230 to 6V transofmer)

just uase a variac, dial up 224V and then do your measurements. if you see frequent voltage fluctuations you may need to add an alarm to warn you when it needs readjustment.

Of course! Googles NEVER like the right answer - I think that's why they resort to googlegroups.

If you think buying a CVT off the shelf is expensive, just add up all of the costs of inventing any other of those schemes.

Good Luck! Rich