Measuring RMS current / voltage 2

[heaterguy]

We are using a solid state relay to control current / voltage to a resistive load. 480V single phase.

Using a Fluke meter, is there a way to determine the RMS current / voltage?

The meter bounces around. For instance, it will show 0.95 amps to 1.11 amps. Is the mid-point the rms current? The meter will also give a peak amps. If we divide the peak amps by 1.414, is it equal to the rms current?

TYIA,

Heaterguy

 

[VEBill]

RMS is usually a concept that is applied to the 60 Hz AC waveform itself. Most (all?) Fluke meters are sufficiently expensive that they would measure all AC functions (voltage and current) using a "True RMS" technique. The True RMS feature means that it will show you the true RMS value even if the waveform isn't a pure sin wave.

But the AC is still 60 Hz and it shouldn't bounce around.

If you're getting a variable reading on a simple resistive load, then something isn't right. Is the voltage steady?

The 1.414 ratio applies only to pure sin waves (like nice clean AC). There is no assurance that your variability is a sin wave (I'd guess it isn't).

Does the meter have an averaging capability? If so, then use it.

 

[heaterguy]

VE1BLL,

I should emphasize that we are using solid state relays to control the voltage / current. When we measured 0.95 to 1.11 amps, the relay was at 30% power. So we probably had 3 cycles on and 10 cycles off or some multiple of that ratio.

Regards,

Heaterguy

 

[itsmoked]

heaterguy; For starters if your control circuit is varying the power dramatically then you are not going to get an average that you can use with a typical meter.

Next if you are not using a zero crossing relay you cannot play the 1.414 game at all because that only works for a pure sine wave!

If your control signal is relatively steady then you can use a "TRUE RMS" type meter which most of the newer meters are now. It will say that on it somewhere!

But if your control swings a lot then you will need to use some sort of long average meter since typical hand-helds have a fixed sample rate of around a 1/3 of a second. The 1/3 of a second can beat with your control driving you nuts while you try to get the average.

 

[Barry1961]

The Fluke should be showing you the RMS voltage. I am not sure if the current reading is a true RMS but it should be very close.

If your peak current sampling time is less than 1.6 ms, .0016 seconds, you should be able to get an accurate enough peak amp reading to divide it by 1.414 to get the RMS current. If your sampling time is larger your accuracy will vary proportionally.

Barry1961

 

[Barry1961]

Are you trying to calculate the power you are using to heat water with a pulsed heater?

Barry1961

 

[heaterguy]

Yes

 

[VEBill]

Sorry - I didn't get that the relay was pulsing from your first post.

If the relay is pulsing, and of course the Fluke meter has it's own sampling time, then you may run into all sorts of subsampling issues if the two rates are close. In other words, the periods might be going in and out of phase leading to an apparent variation at a much slower rate than either of them. What a mess...

Can your Fluke average multiple readings?

 

[heaterguy]

Ve1Bll,

No averaging on this meter. It does have peak.

If we were to purchase a meter for this application what feature should we request (rms, averaging, ??)?

Regards,

Craig

 

[itsmoked]

heaterguy;

Two ways you can do this. One would be a high speed sampling system. Monitor the current and voltage simultaneously and equate the power averaging it for whatever period you want.

OR

Get a power watthour meter and let it run for say 30 minutes to a stopwatch then divide the watthours by time giving you precisely the average wattage.

OR

If you actually KNOW the percentage of ON time... Just turn the heater on full long enough to measure 100% current. Multiply that by the also measured voltage. This gives you the 100% power. Then it's simply the percentage of the 100% that you are running.

 

[VEBill]

Since you're dealing with a signal that varies over time (not just AC, but also On/Off and perhaps in-rush current surges), it might be time to put away the DVOM and start using an oscilloscope.

If you are forced to use a DVOM, then you'll need to average over a sufficiantly long time that the on/off cycles don't leave a significant residual depending on the relative start-stop timing.

If the in-rush current surge is not significant, then perhaps you could just measure the simple steady-state On current and multiply by the duty cycle (simple timing).

 

[heaterguy]

We do have an oscilloscope.

Thanks

 

[VEBill]

You could use the 'scope and a current probe to see the shape (envelope) of the current at start-up.

If the current in-rush isn't significant, then it should be easy to measure the On current (back to the DVOM?), perhaps with the relay 100% on for the duration of this measurement (if that is okay?). Then it becomes a simple duty cycle problem.

If there IS a significant in-rush current surge, then things get complicated, and you might want to find a DVOM with an averaging feature over a suficiently long period to average out many of the On/Off cycles.

 

[itsmoked]

VE1BLL you don't think the "current probe" isn't going to cost a bundle? It's not going to have its own contribution to the scope waveform?

If this is a "resistance" load and not a "tungsten" load then the fact that the element is at fairly steady state temperature means any inrush is unlikely.

heater guy; You need to look closely at my last response there in lies your answer.

 

[heaterguy]

itsmoked,

We don't have a high speed sampling system, nor a power watthour meter, nor do we know the percentage of ON time. If the oscillosope method doesn't work, we can invest in either a high speed sampling system or a watthour meter.

The load is resistive and there is little if any in rush current.

We have the oscilloscope hooked up and we are trying to count the on / off cycles. This is similar to your high speed sampling system? Will the oscillosope method work?

Regards,

Heaterguy

 

[VEBill]

"...its own contribution to the scope waveform?"

At 60 Hz? I don't think so.

"...cost a bundle..."

I was half-assuming that he might jury-rig the Fluke probe into the 'scope just to check the shape of the current in-rush. There are probably other options, but I can't see his gear through the Internet.

"...inrush is unlikely."

I agree, but worth checking if the in-rush time is a significant fraction of the on-time. It wouldn't take much for this to be a significant source of error. If the heater is a typical enclosed-in-a-tube, then its temperature can vary from ambient to red-hot even if it is immersed in a constant temperature bath. I agree that significant in-rush is not likely, but it could be THE largest source of error if you assume it wrong. I'd check because I just like to be careful...

I think that we both agree that it can be moved to the time domain (duty-cycle) once you know the 100% load (assuming it is flat-topped).

 

[itsmoked]

Hi heaterguy,

I am trying to be diplomatic to V1BLL. I have to make the measurements you are talking about on a regular basis since I design heater controls and PIDs for a living. I cannot see how a scope will help you in the least. Unless you don't care if your measurements are off 10-20%. Which I gathered you did care about, as your original measurements are probably 'that' close.

You haven't described anything else about your control system or what kind of heater or what the heater is in, and why you need the measurement so we are all kinda shooting in the dark er... cold here.

But you have stated(up top)that the "relay was at 30%". If you know the percentage and lots of time you can know this, because you can set controllers to fixed power settings, then my third option above works very well! Using a scope will drive you nuts for this kind of measurement and I wouldn't stake a dime on the results.

Another problem, again because we have no details, is the control signal to power translation. Zero crossing SSRs can have unexpected differences. Looking at the control side of an SSR DOES NOT give an accurate picture of what is happening on the power side. There are so many variables here it would take me an hour to describe.. It boils down to: If you want to know the power, measure the power on the power side. If the power control is varying constantly and you want to know the average process power, then take a long baseline average(minutes)(watthour scheme).

 

[heaterguy]

Here's the results from the oscilloscope...

We can count the number of cycles on/off; however, the waveforms are not identical. Sometimes the first wave on is a normal size and sometimes it much smaller (about 10-40%). Therefore, itsmoked is correct. The proper way to measure is with a power meter.

We do not own a power meter. Does anyone have a suggestion of one to purchase?

 

[fangas]

Please describe the following, there might be a cheap & dirty way.

Your controlling device
The signal from the controller to the SSR
Your SSR

Ed

 

[itsmoked]

I use several of these.. They are a lot of bang for the buck. Look at the full brochure. It seems to have the important info.

http://www.powermeterstore.com/plug/wattsup.php

I use the standard one. For what you want to do, you may want to drop the additional 40 bucks for software and communications stuff on the "pro".

Check its specs etc. You need to be able to run your heater power wiring through it.

I have used them to audit everything in my house and many work projects and computers. They quickly answer the "how much does it cost" to leave my PC on all the time questions.

If you are running something other than 120Vac we will need to come up with something else.