### Author Topic: wattmeter needed  (Read 1293 times)

0 Members and 1 Guest are viewing this topic.

#### eevcandies

• Regular Contributor
• Posts: 92
• Country:
##### wattmeter needed
« on: September 25, 2020, 08:26:13 pm »
I need a wattmeter for DC project....up to 36 volts, and 10 or so amps max.  Would like 0.1 watt resolutions (ex: 132.6 W) & 0.5% accuracy or better.

Do any multimeters have watts capability???  It needs to readout continuous, as I make adjustments & need to see the wattage.

Obviously it needs more than 2 terminals---3 or 4 depending on the hookup.

##### Re: wattmeter needed
« Reply #4 on: September 25, 2020, 10:05:07 pm »
like measuring a switching power supply output

#### xxx2fan

• Contributor
• Posts: 26
• Country:
##### Re: wattmeter needed
« Reply #5 on: September 25, 2020, 11:13:12 pm »
200A DC Digital Monitor LCD Volt Amp Watt Meter RC Battery Solar Power Analyser
you can get them down to 10A or less.
Less than 20.00

just an idea.

#### bdunham7

• Super Contributor
• Posts: 1170
• Country:
##### Re: wattmeter needed
« Reply #6 on: September 25, 2020, 11:28:47 pm »
Athough you can technically, mathematically, 'RMS' anything, I don't think you want 'RMS watts'.

To measure average power on a system with varying voltage and a purely resistive load, you just measure RMS voltage and calculate the power.  To measure instantaneous power you measure instantaneous voltage and calculate the power.

To measure average power on a system with varying voltage and a reactive or changing load, you measure instantaneous power by measuring instantaneous voltage and current at the same exact time and then average the power readings over some period.  To measure instantaneous power.....well you can figure that out.  In this case, you can't just measure the RMS voltage and RMS current and multiply to determine the power.

edit:  To continue, power meters don't work by using RMS, rather they use the instantaneous measurements to calculate active and reactive power, power factor and whatever else they happen to do.  In your case, you appear to want the 'real' or active power and assuming it is steady over the period you want to observe, the most correct term I can think of for it would be 'continuous average power'.

To specify a meter to do a reasonable job of this, we would have to know the the voltage, current and bandwidth involved, as well as the nature of both the source (what 'waveforms' do you refer to) and the load.
« Last Edit: September 26, 2020, 01:52:16 am by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people.

#### joeqsmith

• Super Contributor
• Posts: 6754
• Country:
##### Re: wattmeter needed
« Reply #7 on: September 26, 2020, 12:06:16 am »
I need a wattmeter for DC project....up to 36 volts, and 10 or so amps max.  Would like 0.1 watt resolutions (ex: 132.6 W) & 0.5% accuracy or better.

Do any multimeters have watts capability???  It needs to readout continuous, as I make adjustments & need to see the wattage.

Obviously it needs more than 2 terminals---3 or 4 depending on the hookup.

I see a bunch around $20 for RC cars but I'm not sure if they measure RMS. My readings must be RMS due to the various waveforms involved. I thought a trueRMS Fluke model might offer watts readings, but no dice. I don't care whether this meter cost$20 or $200, just need something. I have 10 other meters, none with watts. Even a "watts module" that outputs a voltage usable with a multimeter would be fine. Again, all readings must be RMS. Many handhelds will not support 10A continuous. The Gossen Energy can use an external DC clamp. https://www.gossenmetrawattusa.com/media/78441/metrahit-energy-ba_gb.pdf How electrically robust is your meter?? https://www.youtube.com/channel/UCsK99WXk9VhcghnAauTBsbg #### eevcandies • Regular Contributor • Posts: 92 • Country: ##### Re: wattmeter needed « Reply #8 on: September 26, 2020, 03:11:07 am » What multimeters can measure power? Any uni, mastek, fluke , etc? What method do the RC wattmeters use? I cant find ANY info on their method or crest factor https://www.ebay.com/itm/Digital-LCD-Monitor-Watt-Meter-60V-100A-DC-Ammeter-RC-Battery-Power-Amp-Analyzer/132432985864?hash=item1ed59eff08:g:k8MAAOSwyXNaL6xi If this uses rms measurments, it would be great. How are they calculating watts? -RMS power is: K* square root (realtime volts*amps squared & integrated--summed, for a small time) or it is the RMS volts * RMS amps * powerfactor A multimeter that uses average will likely not give an accurate value in the first place, even if you are just measuring the voltage of the shaped wave, let alone power. You need rms when you have wave shapes, not sure why this is even a question...that is why you have RMS meters. « Last Edit: September 26, 2020, 03:18:12 am by eevcandies » #### bdunham7 • Super Contributor • Posts: 1170 • Country: ##### Re: wattmeter needed « Reply #9 on: September 26, 2020, 03:54:40 am » You need rms when you have wave shapes, not sure why this is even a question...that is why you have RMS meters. RMS power is as valid a concept as RMS stock prices. Both can be calculated, neither makes the least bit of sense. (edit: what you have written after "RMS power is" is not correct) Instantaneous (or steady state DC) power = V*I and and since V=I/R, power also equals I²R or V²/R. When you have a non-reactive, purely resistive load, an AC voltage or current measured in RMS terms produces power in that load equal to the same value DC current. However, it is not correct to call the resulting value 'RMS power', not just as a matter of semantics, it is just plain wrong. If the load is reactive or changes for other reasons, then none of these power calculations or methods are correct. If the load is not purely resistive, you cannot measure the power using either RMS voltage or current and having both is not really helpful in most cases. In the special case where the load is reactive, but has only either a capacitance or an inductance value, the current will be out of phase with the voltage by a certain angle and the actual power is Vrms * Irms *cos(phase angle). That last term is the power factor. In real life, a wattmeter can't know that any of those special conditions will hold and the power factor is not easy to determine. The usual technique is to sample both the voltage and current at the same time at a rate higher than the anticipated bandwidth of the system and calculate the power, power factor and all of those things using appropriate algorithms. The continuous average power is just the average of n [V*I(t)] samples over a period of t(0) to t(n). The other calculations are a bit more complex. VA (Volt-amperes) or 'apparent power' is what you get when you multiply Vrms and Irms, for example. The reason you don't get this in many meters is that it is difficult to implement in the general case. Power meters are usually designed to work on specific systems--like mains voltage at 50 or 60Hz, DC, etc. A wide-range wide-bandwidth power meter is not going to come as a feature on a$20 or $200 multimeter. Quote What method do the RC wattmeters use? I'm sure they use an MCU with an ADC, a small current shunt and the appropriate firmware. I'm sure they don't use RMS. The most pertinent question in your case would be their sample rate and your system bandwidth. If you want a reference, I just looked at the Wikipedia articles and they aren't bad. Hopefully you can now see why this is a question. https://en.wikipedia.org/wiki/AC_power https://en.wikipedia.org/wiki/Wattmeter « Last Edit: September 26, 2020, 04:21:45 am by bdunham7 » A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people. #### eevcandies • Regular Contributor • Posts: 92 • Country: ##### Re: wattmeter needed « Reply #10 on: September 26, 2020, 07:53:27 am » I'll give you partial credit, you should not do the square and square root, like a normal rms calc, however, you must use integration. If you can't/don't want integration & just use two signal readings, they must be RMS measurements, not avg measurement. So a meter that is going to simply multiply 2 values (Volt & amps), must use RMS to be accurate. Of course the best way is to instantly read and mult I & V to form watts in real time & get an overall reading. I suspect chapo meters are not doing this...they might just take the avg of each & multiply those, giving poor results. This method provides a true power measurement and true RMS measurements for any waveform, sine or distorted, including harmonic content up to the bandwidth of the instrument. (see image) This is what must be avoided here: If the DC-voltage as well as the DC-current aren't ripple and noiseless than its impossible to measure the power from individual instrument average readings. The mean voltage in this example is 10 V and the mean current is 2 A. A simple multiplication would lead to a power of 26 W, while the real power is 25.1 W. Thus, where both waveforms are pulsed, have ripple or contain noise, the power can only be measured with a wattmeter How do these RC meters calculate? You seem to propose they can just take average current*avg voltage & give an accurate watts. They need to use RMS waveform values to at least have some hope.. For non reactive loads, they must include a power factor. Use rms voltage and/or rms current to calculate average power, resulting in meaningful power values. and using RMS is exactly what I put in my first post, then that seemed to be immediately questioned! « Last Edit: September 26, 2020, 08:29:35 am by eevcandies » #### HKJ • Super Contributor • Posts: 1480 • Country: ##### Re: wattmeter needed « Reply #11 on: September 26, 2020, 08:25:18 am » All meters use averaging on DC, also RMS meters. The RMS function is only active when measuring AC. If a meter did use RMS on DC it would not be able to show negative values! When measuring power the RMS function is not really used, instead the voltage and current is multiplied and then averaged. This can be done by either a analog multiplier or by sampling and then multiply the results. The sample method must sample at kHz before it can be used to measure AC power. #### MosherIV • Super Contributor • Posts: 1500 • Country: ##### Re: wattmeter needed « Reply #12 on: September 26, 2020, 08:30:44 am » Hi. I will explain this from a different point of view. TrueRMS for multimeters mean that they provide a RMS value when measuring AC volts. This makes a difference when measuring non sine wave AC waveform, the rms voltage of a square wave and sine wave with the same peak to peak voltage is different. I am not sure if they do measure RMS for AC current, maybe but I am not sure. If a dmm does have True RMS and power measurement, RMS is most likely not for the power measurement. Some others have found meters specifically for dc power, go with one of those. #### eevcandies • Regular Contributor • Posts: 92 • Country: ##### Re: wattmeter needed « Reply #13 on: September 26, 2020, 09:10:00 am » Some others have found meters specifically for dc power, go with one of those. Such as which meters? Again, this is not for a line powered AC application, it is powered by a switching DC source with ramping & curves up & down. It would make no sense for a module to use average volts & amps to calculate delivered power, but it would for the meter to use rms amps and volts (or to do instantaneous integrating). « Last Edit: September 26, 2020, 09:15:11 am by eevcandies » #### joeqsmith • Super Contributor • Posts: 6754 • Country: ##### Re: wattmeter needed « Reply #14 on: September 26, 2020, 02:19:42 pm » Athough you can technically, mathematically, 'RMS' anything, I don't think you want 'RMS watts'. ... To specify a meter to do a reasonable job of this, we would have to know the the voltage, current and bandwidth involved, as well as the nature of both the source (what 'waveforms' do you refer to) and the load. Anytime I see or hear RMS power, I think back to my first year out of high school. We know the voltage and current but only the OP knows the bandwidth. Maybe the Gossen Energy with an external DC clamp can't measure it or it costs too much. Other general ideas would be a shunt, two meters and some software. LEM would be good to about 100KHz. Another option would be a shunt, scope, some signal conditioning along with some software. Maybe they need to handle really high common modes and want a handheld for that. How electrically robust is your meter?? https://www.youtube.com/channel/UCsK99WXk9VhcghnAauTBsbg #### bdunham7 • Super Contributor • Posts: 1170 • Country: ##### Re: wattmeter needed « Reply #15 on: September 26, 2020, 02:59:48 pm » I'll give you partial credit, you should not do the square and square root, like a normal rms calc, however, you must use integration. If you can't/don't want integration & just use two signal readings, they must be RMS measurements, not avg measurement. So a meter that is going to simply multiply 2 values (Volt & amps), must use RMS to be accurate. Thank you for the partial credit. Did you read what I wrote and look at the links? First, if you just 'solve' your equation you'll find that what comes out is just the mean of the function or samples. The integral of instantaneous power over a time period is energy, not power. Energy divided by a time period is.....average power. It's like magic. As for multiplying Vrms and Irms to get power, please get it through your head that it isn't 'accurate', it doesn't work at all unless the load is purely resistive. Quote Of course the best way is to instantly read and mult I & V to form watts in real time & get an overall reading. I suspect chapo meters are not doing this...they might just take the avg of each & multiply those, giving poor results. I don't know the basis for your assumption that they don't do this. There really is no other way to get a meaningful result, although there are other factors that could affect accuracy and I wouldnt' be surprised if a$10 wattmeter was less than highly accurate.  The missing information is the sample rate, which I don't see in any of their spec sheets. One just lists "SAMPLE RATE: sample/s".

Quote
How do these RC meters calculate?  You seem to propose they can just take average current*avg voltage  & give an accurate watts.  They need to use RMS waveform values to at least have some hope.. For non reactive loads, they must include a power factor.
Use rms voltage and/or rms current to calculate average power, resulting in meaningful power values.   and using RMS is exactly what I put in my first post, then that seemed to be immediately questioned!

Again, please reread--I said nothing about using average volts or amps.  They way a wattmeter works is 1) take samples of instantaneous voltage and current 2) calculate the instantaneous power at each sample time point and then 3) average the calculated power results over a set time period, this is displayed as active or real power in W, Watts.

To calculate RMS current and voltage over the same time period, the MCU would calculate the square of each sample, then the mean of those squares, then the square root of that mean.  RMS = Root of the Mean of the Squares.  The meter can then display Watts, RMS Volts, RMS Amps and, by multiplying RMS Volts and RMS Amps it gets what is called the apparent power, typically known as VA or Volt-Amperes.  It can also display PF, Power Factor which is W/VA, and if you wish reactive power, VAR, which is VA-W.  That's how they work.

The important point in all that is that the power calculation is not derived from RMS measurements, period--and it cannot be derived from them alone.  A bare-bones watts-only meter would not bother calculating or measuring RMS at any point. However, once you have all those sample points and an MCU with nothing else to do, it doesn't add much cost to add on all those extra readings.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people.

#### eevcandies

• Regular Contributor
• Posts: 92
• Country:
##### Re: wattmeter needed
« Reply #16 on: September 26, 2020, 03:18:39 pm »
Here is what I ended up getting:

https://www.amazon.com/gp/product/B079JVGRSL/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1

Now my entire point is, if they just are taking an RC avg of the current  & displaying that  & taking an RC avg of the voltage & displaying that, then multiplying & displaying watts, that will work only for fairly steady signals.  Mine are shaped with curves, so they need to use RMS values for the volts & amps to be reasonably accurate, if they are merely multiplying 2 results.  The V & I signals are fairly close in phase, so the power factor is nearly one here & not a consideration.
So does this meter 1) just mult two avg values 2) is it using two rms values, 3) is it doing a realtime integration?  #1 will not work

If I use two multimeters to manually do it, I use two RMS meters & get accurate results, but can't multiply in realtime all day long.  Want a meter that does so for me (or does #3).

#### bdunham7

• Super Contributor
• Posts: 1170
• Country:
##### Re: wattmeter needed
« Reply #17 on: September 26, 2020, 03:56:05 pm »
Here is what I ended up getting:

The V & I signals are fairly close in phase, so the power factor is nearly one here & not a consideration.

So does this meter 1) just mult two avg values 2) is it using two rms values, 3) is it doing a realtime integration?  #1 will not work

If I use two multimeters to manually do it, I use two RMS meters & get accurate results, but can't multiply in realtime all day long.  Want a meter that does so for me (or does #3).

Have you asked the seller what it does?  Your $12 wattmeter does list a sample rate--2 per second. You'll be lucky if it even averages, it probably just takes two samples and displays the results. I'm not sure what you expect from cheap trinkets like that. However, you can easily test it if you can vary the power waveform you send through it. Just compare it with your two-meter method and see how close it is or isn't. If the PF is 'nearly one', then using 2 RMS meters will be 'nearly' accurate--that's true. But the odds of finding a wattmeter using this method is nil outside of something exotic. A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people. #### eevcandies • Regular Contributor • Posts: 92 • Country: ##### Re: wattmeter needed « Reply #18 on: September 26, 2020, 04:25:13 pm » But the odds of finding a wattmeter using this method is nil outside of something exotic. I wouldn't exactly think so. I just want a single multimeter to do what I'm now doing with 2 multimeters---so a dual channel would suffice, I'd think. All they need is a setting to do the multiply (or do the actual integration). I did find this meter, though couldn't find a power setting in the manual (yhthough the LCD shows some "power" icons) #### joeqsmith • Super Contributor • Posts: 6754 • Country: ##### Re: wattmeter needed « Reply #19 on: September 26, 2020, 04:26:17 pm » " Would like 0.1 watt resolutions (ex: 132.6 W) & 0.5% accuracy or better." Odd you would go that route. I was guessing you also wanted to run down to 0V. I wasn't able to find the continuous max current or the accuracy. The specs on most handhelds for 10A is pretty limited. 15 seconds 10, 5 minute cool sort of thing. Nothing near 10A continuous. CEM and UNI-T have some handhelds that may have worked as good. When you get it, it would be interesting if you could do some sort of review on it. Not an unboxing, what's included but rather fill us in on the details. « Last Edit: September 26, 2020, 04:28:17 pm by joeqsmith » How electrically robust is your meter?? https://www.youtube.com/channel/UCsK99WXk9VhcghnAauTBsbg #### bdunham7 • Super Contributor • Posts: 1170 • Country: ##### Re: wattmeter needed « Reply #20 on: September 26, 2020, 04:46:47 pm » I wouldn't exactly think so. I just want a single multimeter to do what I'm now doing with 2 multimeters---so a dual channel would suffice, I'd think. All they need is a setting to do the multiply (or do the actual integration). Well, good luck and let us know when you find it. The meter you show is both pretty exotic and discontinued. And it appears that even this rare and discontinued meter won't work for you because it doesn't do TRMS AC+DC, which is what you need. You could use a PC-connected DMM that can read TRMS AC+DC V and I simultaneously (I can't name one offhand-can you?) and a program like LabView to do what you want by multiplying the two values. What I can't understand is why you insist on this method. Power measurement is not a black art--people all over are doing it all the time. The reason they don't do it your way is that it doesn't work except for special cases. A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people. #### eevcandies • Regular Contributor • Posts: 92 • Country: ##### Re: wattmeter needed « Reply #21 on: September 26, 2020, 05:49:08 pm » I was guessing you also wanted to run down to 0V. In this case, I never go below 12V, or am not interested if I do...the voltage jumps around between 12V & maybe 26V, tops, at 1 KHz max when it's undulating. TRMS AC+DC V and I simultaneously That should be easy for a dual meter..it only has to be faster than myself looking at two meter & digging my calculator 100 times. I'll give the little module I ordered a workout & see which of the 3 methods it is using. What I can't understand is why you insist on this method. I am not insisting at all. The only thing I DON't want is a reading that is merely multiplying the avg current times the avg voltage, that would be very inaccurate. With that in mind, what multimeters or modules should I consider? « Last Edit: September 26, 2020, 05:57:11 pm by eevcandies » #### Wytnucls • Super Contributor • Posts: 2951 • Country: ##### Re: wattmeter needed « Reply #22 on: September 26, 2020, 06:05:06 pm » « Last Edit: September 26, 2020, 06:15:21 pm by Wytnucls » #### bdunham7 • Super Contributor • Posts: 1170 • Country: ##### Re: wattmeter needed « Reply #23 on: September 26, 2020, 06:14:04 pm » That should be easy Things often look easy when you don't fully understand them. Quote With that in mind, what multimeters or modules should I consider? Multimeters, I can't think of anything. Power meters, if you really want the precision and accuracy you claim to, you need to find a manufacturer that actually provides the specifications you need to determine the suitability for your purpose. Sample rate and/or bandwidth, for example. Accuracy (for the power reading) vs. bandwidth for another. Here is one that actually seems to meet your criteria: https://www.bkprecision.com/products/multimeters/5335B-power-meter.html I suspect you'll manage to live with less! A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people. #### eevcandies • Regular Contributor • Posts: 92 • Country: ##### Re: wattmeter needed « Reply #24 on: September 26, 2020, 07:58:17 pm » Have a look at this one: http://www.twtes.com.cn/en/product/pro62.html https://www.aliexpress.com/item/32964404017.html « Last Edit: Today at 06:15:21 pm by Wytnucls » WELL WELL WELL....that is quite interesting. When I have been on my bench stool the past 6 months, what has been about 2 feet from my knee buried under some digikey boxes? I had to dig around to find where the rascal was at... look at the picture...I use this occasionally for line power measurements (very rarely)....I never had any inkling this could be used for non-AC-line powered setups. I got this for about$150 seven years ago & almost forgot I had it, since it's rarely been used.

You will also note it clearly says:  EXTECH TRUE RMS POWER ANALYZER  !!

Time to give this puppy a whirl

« Last Edit: September 26, 2020, 08:03:50 pm by eevcandies »

#### Wytnucls

• Super Contributor
• Posts: 2951
• Country:
##### Re: wattmeter needed
« Reply #25 on: September 26, 2020, 08:08:47 pm »
Well, that's fortunate. Hopefully, that's all you will need!
Extech doesn't make anything. They slap their name on all sorts of meters from other brands.

#### mjkuwp

• Supporter
• Posts: 237
• Country:
• mechanical engineering defector
##### Re: wattmeter needed
« Reply #26 on: September 27, 2020, 01:30:47 pm »
Athough you can technically, mathematically, 'RMS' anything, I don't think you want 'RMS watts'.
...

member bdunham7 has the correct position and the entire post is good.
https://www.eevblog.com/forum/testgear/wattmeter-needed/msg3250004/#msg3250004

My interpretation is that you desire instantaneous readings of Volts and Current (multiply to get power). and you want this done at a very high data rate.  I understand the desire for high quality measurements but still I don't see any reason RMS comes into play here.

Another idea from left field is to use an analog multiplier with a high bandwidth and put your meter at the output of this.  Parts such as AD633 will do this.

and there is a brief explanation of a project here:

https://www.circuitspecialists.com/blog/measuring-power-with-a-digital-panel-meter/

maybe there are some dev boards or other projects that would be available for purchase.

#### eevcandies

• Regular Contributor
• Posts: 92
• Country:
##### Re: wattmeter needed
« Reply #27 on: October 01, 2020, 03:54:55 pm »
I don't want it "instant" (2-3 readings a sec is fine) I just want to  multiply RMS volts times RMS amps to get VA , or Watts, if the power factor is near unity.    That's the standard equation.  So a dual multimeter that measures both should be able to do the trick (since I do the same thing using two rms meters and a calculator).

#### mjkuwp

• Supporter
• Posts: 237
• Country:
• mechanical engineering defector
##### Re: wattmeter needed
« Reply #28 on: October 08, 2020, 01:13:49 pm »
I think Dave's 121GW meter does this with a 3-wire connection.
....

.....

The meter you mention looks pretty interesting...I wish it gave selectable watts rather than just VA (since it monitors both I & V in one unit, it should be able to measure watts)..well, it's close!   I wonder if there is one that measures output as well, this is for power supply efficiency testing (input & output power)..could use two of these meters, perhaps.

I guess I am not sure what you are after then because some statements don't go together.

The units of V*A is Watts and 1 Volt x 1 Amp == 1Watt so if you don't care about power factor and want to assume a purely resistive load then this meter would do that.  The display might have an indicator stating VA or mVA but you can read this as Watts.

The quantity you would be measuring I think is Apparent Power as opposed to Active Power because it is not accounting for phase differences between the waveforms.

It could be helpful if you share with the group what you are trying to achieve - what is the outcome you want.  It could be an interesting discussion; I think context always helps when discussing tools.

Smf