can someone recommend a really TRUE RMS meter?

Maybe what You need is a Scope ?

--
"I'd rather have a bottle in front of me than a frontal lobotomy"

"Daily Thought:

  SOME PEOPLE ARE LIKE SLINKIES. NOT REALLY GOOD FOR ANYTHING BUT
  THEY BRING A SMILE TO YOUR FACE WHEN PUSHED DOWN THE STAIRS.
http://webpages.charter.net/jamie_5"

Honest-to-Gawd true RMS meters are relatively uncommon, but they do exist. Look for something like this:

which actually measures the "heating power" of the signal in question, and therefore is getting you as close as you're likely to get to what "RMS" is supposed to be without capturing the waveform and doing the math (which generally would require a fairly sophisticated scope).

Bob M.

If it doesn't say "RMS" then it probably measures the peak & assumes a sinusoid.

If it does say "True RMS", then it's like the Fluke meter and measures the RMS of the AC portion of the signal -- but it's pretty easy to measure twice and calculate sqrt(a^2 + b^2), so what's the problem?

Lots of spikes, and lots of frequency content above 60Hz, may cause problems -- you're describing a signal with a high "crest factor", which will limit the meter's abilities, and I wouldn't just assume that a meter has good performance above a kHz or so, either.

Careful reading of the specs may tell the truth -- look for crest factor and bandwidth, and see what they say.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

A digital scope with a true-RMS calculation feature could give you the numbers, I suppose... or, just capture the data with a fast digital scope, upload it to your PC and do the math there.

Or, look around to see if you can find somebody who has an olde style HP 3400 true-RMS voltmeter (I got mine for a song at a local hamfest a couple of years ago). To measure RMS current you'd either need to have a low-impedance sense resistor installed in the current path, or use a current probe (HP 456A, or some sort of home-brew current transformer with resistive termination, and then some work to calibrate it).

The 3400 uses the two-thermocouples-and-a-feedback-loop method of RMS detection.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

A properly designed true-RMS meter can easily be more accurate than a scope. Scopes are no better than 1-2% on a good day. Analog Devices used to make a true-RMS sensor that used a heater and (iirc) a diffused silicon resistor bridge.

Cheers,

Phil Hobbs

There are a number of ICs and meters that measure true RMS by using the thermal method. Some have a 100Khz bandwidth, and others are better.

SHAMELESS PLUG: My Daqarta sound-card software has a true RMS Voltmeter. It actually does the root-mean-square computation. If you can trigger on the waveform (Daqarta has lots of trigger controls, so you probably can unless it is really noisy), then the computation will take place over an integer number of cycles. Otherwise, on an untriggered source the computation will be done over 1024 samples.

Daqarta also includes a signal generator, which (depending upon your particular situation) you may be able to use to initiate your pulses. In that case, Daqarta can use Gen Sync to trigger off its own generator and insure it knows the exact cycle length.

Or, if you have an external trigger source available (that is driving the signal), you can feed that into the other input channel and just use that as the trigger.

In any case where you can get a stable trigger, you can also use Daqarta's synchronous waveform averaging to reduce noise.

IMPORTANT: Note that although the RMS computation handles DC correctly, the sound card is AC coupled.

Also note that you'll need to calibrate your system first. Daqarta has an auto-calibration option that determines the relative step sizes of the sound card mixer using a loopback connection, but you'll also need a full-scale range measurement. If you already have a meter that reads AC Volts, you can use that as your reference, with a sine wave input (that can be provided by Daqarta, if you wish). You just take a

reading and do a simple calculation to get the value to enter into Daqarta's Full Scale Range dialog. All described in great detail in the Help system.

Since your signal can run larger than sound cards can handle, you should also provide an input attenuator to keep things in the +/-3V range or so. (The card probably won't be damaged by +/-12, but the input ranges will clip.)

Daqarta is US$29 to purchase (Personal/Hobby license), but it sounds like this may be only a short-term need. If so, you may be able to do your whole job within the 30-day/30-session trial period. Enjoy!

Best regards,

Bob Masta DAQARTA v3.50 Data AcQuisition And Real-Time Analysis

Scope, Spectrum, Spectrogram, FREE Signal Generator Science with your sound card!

While what you say is true those instruments tend to be expensive. For most any purpose any "TrueRMS" class handheld DMM or small benchtop TrueRMS DMM will do OP's task quite nicely. OP may have to read the datasheet to verify that it includes DC in the "TrueRMS" calculation. Well designed RMS to DC converter ICs have been around for over 20 years.

The most popular hand-held meter that gave true rms voltage used to be the Fluke 87, which could cover the audio bandwidth, using an analog rms converter (100KHz limit). It produced rms current measurements bt measuring a shunt resistor voltage, which is the most common current measurement method.Other meters are available that use the same kind of hardware.

The old Fluke 8922a bench powered meter used to be usefull with higher voltage/power circuits. It used a thermal converter, followed by conventional voltage metering and is highly sensitive to calibration. This has a 1MHz or 11MHz bandwidth, depending on input voltage range. There are other similar meters.

More recently, given the availability of digital scopes with good bandwidth and math capability, mathematical derivation is more common.

You can't just multiply an RMS current and RMS voltage measurement to get power. Their phase relationship is important. As an extreme example, an amplifier driving an inductor with 1Vrms will dissipate internal losses equivalent to driving a short circuit, while the load power is minimal.

A digital scope math calculation uses time-coherent multiplication to produce a fairly reliable power indication for related waveforms in its display/memory.

DC-offset will also affect measurements, the ability of the meter to produce meaningful readings and the operator's ability to understand what the reading means.

RL

Precisely.

Chris

"You can't just multiply an RMS current and RMS voltage measurement to get power." - RL

So.... I've read that Vrms*Irms = Pavg..... but..... you're saying I need to include the phase shift?! Isn't the phase shift implied in the waveform? I have two meters hooked up, and I'm reading a Vrms and an Irms, if I multiply these together don't I get the Apparent Power? Aren't RMS values scalars? No phaeses involved.... and isn't Vrms*Irms = P apparent?

You really do have to worry about the phase shift. When the voltage and current are in phase, the power is going in the same direction as the current. When they're out of phase, the power is going in the opposite direction. From the wall socket's perspective, that's the difference between a motor and a generator.

If you imagine a load that drew some enormous current right when the voltage was near 0, and much less current near the peaks, you could be off by a factor of 100 or more if you just multiplied RMS voltage times RMS current. The same is true of a purely reactive load, in which the voltage is 1/4 cycle out of phase with the current. For instance, this is what it would look like with a capacitor across the line.

I___ ___ / X \ V / / \ \ /---/---\---\---/---------------------- / \ \ / / \ X

-- --- -- | | | | | | | | | | 1 2 3 4

In quarter-cycle 1, I >0 and V < 0, so the VI < 0 : power is going from the capacitor into the wall socket. In QC 2, I>0 and V>0, so VI > 0: the power is going from the wall socket into the capacitor. In QC 3, I0, so VI < 0 again, and In QC 4, I

Hmmmmm..... ok, well my voltages will definitely be out of phase since there's some reactance involved.... and I agree with everything you are saying here, but this is looking at just one cycle of each.... I'm assuming a meter is constantly sampling the signal and so the RMS value is some sort of average that is constantly being updated, and the average is maybe taken over a hundred cycles or somethign like that not just one cycle, so do you think that since the meter is taking an rms average of many cycles that somehow the phase shift would then be included in there when I multiply the values together? I'm fine with the fact that I'm getting the reacitve power (the negative power going back to the source) in my readings, I'm fine with the fact that I'm getting apparent power instead of true/active power... but I only have meters and am trying to avoid the phase issue.... I'm using the two-wattmeter method to do this (thanks Phantom)

J.

No, since every cycle is just like every other, if the total is zero over one cycle, it's zero, period. Sorry about that.

Cheers,

Phil Hobbs

yes, completely agree, that a purely reactive circuit has no real/true power component to it. I don't mind that, I know that I am measuring Apparent Power which is not True Power.... but my question is, Do I Have To Worry About The Phase Shift For Making Apparent Power Measurements By Taking RMS Voltage and Current Readings From A DMM... at this point, I do not think that I need to know the phase shift to make a measurement of the apparent power... Vrms*Irms=apparent power, RMS values are scalars and contain no phase information...

Apparent power is what you'd get by doing that. On the other hand, you might just as well throw darts at a piece of graph paper. "Apparent power" is not a term of praise.

If you know the sign of the reactance, you could correct the power factor and measure it afterwards.

Cheers,

Phil Hobbs

What you're doing isn't the two-wattmeter method. You are using two separate meters and two separate meters don't make one wattmeter:

The essence of an electrodynamometer wattmeter is that the movement has two coils, a current coil and a voltage coil. The torque that deflects the needle is proportional to the instantaneous product of the current and voltage. The mechanical inertia of the movement smooths out the pulsations that might occur due to the fact that the instantaneous power varies from 0 to some magnitude rapidly.

The reactive power is dealt with by the fact that when the power is returning to the source, the needle would ordinarily be deflected backward, so the average of forward and reverse power is what needle indicates.

When you use two separate RMS responding meters, you are in fact measureing apparent power. The real power will always be less than or equal to the apparent power. If the load is purely resistive, they will be equal. If the load is more or less reactive, the real power will be less than the apparent power, and sometimes the real power may be MUCH less than the apparent power.

If you apply, for example, a 24 VAC stimulus (using a transformer for safety) to a 10 microfarad capacitor (not an electrolytic), the apparent power would be

2.17 volt-amperes. I have such a capacitor lying around in my junk box, and it has an ESR of 50 milliohms. The real power drawn in this case would only be 4.5 milliwatts. This is almost a one thousand to one ratio of apparent power to real power.

Furthermore, you have what amounts to a three-phase load. If you let one lead of the motor circuit be reference, and measure the voltage and current in the other two with respect to that reference with separate RMS responding meters, you will not be carrying out the two-wattmeter method. You have to use actual wattmeters, or their equivalent, a two channel scope with trace math.

The three phase nature of your load additionally complicates things. I think that since your load is probably inductance + resistance, the two separate measurements probably add up to the apparent power, but I'd have to think about a little to be sure.