One Phase Power Meter/Analyzer

[Martin72]

Hi,
This is supposed to be a “loose collection” of model suggestions, i.e. I am interested in a power meter/analyzer, but I don't want to buy one immediately.
I would like to have an overview of the affordable models on the market first.
By affordable, I mean devices well under €1000.
I don't mean built-in models or the infamous plug-in devices for energy measurement either; I'm talking about fully-fledged desktop measuring devices.
What it should be able to do is measure current, voltage, power, power factor, frequency, THD (current/voltage).
The following would be desirable:
-Power factor display cap/ind.
-Reactive power/effective power/apparent power
-Display of Wh/Wsec

In addition, a basic accuracy of below/equal to 0.5%, a bandwidth of at least 1kHz.
What I have seen of the models on the market so far is rather sobering.

PeakTech 2510 (370€)
Uni-T UTE9802+ (387€)
Uni-T UTE9806+ (458€)
Uni-T UTE9811+ (410€)
Matrix MPM-1010 (166€)
Matrix MPM-1010B (196€)
GwInstek GPM 8213 (910€)

The models from “Matrix” are too cheap to be any good, I don't trust Uni-T and Peak Tech, so that only leaves the GwInstek in this list – but its price is already at the limit.
If there is no other way, I would pay the price - but maybe you have other models to suggest.

Martin

[Fungus]

[Kean]

Big Clive's Hopi meter doesn't meet the OPs request for a bench meter, or provide THD measurement.

I have not done research on latest models, but I would suggest checking out the Tonghui TH3300 series, or maybe TH3400.
Specifically the TH3312 if you want up to 20A, or the TH3321 if you only need up to 2A.  See attached datasheet image (the PDF I have is in Chinese).

It seems to be selling on Amazon and other places for under USD700, but some sites say it is discontinued.  The Tonghui website doesn't indicate that.
I have a TH3321, for which I paid USD620 (about AUD1,100 including shipping) in 2019.  I'm quite happy with it, although I never got around to any writing python code for report generation.

https://en.tonghui.com.cn/product/TH3300-Series-413.html
https://en.tonghui.com.cn/product/TH343XTH344X-Series-415.html

[2N3055]

EU source for Tonghui.

https://www.sourcetronic.com/shop/en/digital-power-meter/

[Kean]

Ah yes, Sourcetronic will be a little more expensive that buying from AliExpress or Tonghui but you will get local support.
Pretty good support too based on previous feedback here, and the prices on that page seem reasonable.

I am certain that Martin knows Sourcetronic as a Tonghui reseller/rebadger from prior threads re TH/ST LCR models.

[Martin72]

Thank you all so far.
It seems that the TongHui is “in the lead”, and there is probably a price to pay for that.
In addition to the €690, VAT has to be added, which makes it €821, so in the range of the GwInstek.
The cheapest TongHui costs $650, plus $190 for shipping, customs and sales tax... That should be just under €1000.
And then I wouldn't have the service that GWInstek or Sourcetronic offers.
As it stands now, it comes down to GWInstek or Sourcetronic.

[Martin72]

I am certain that Martin knows Sourcetronic as a Tonghui reseller/rebadger from prior threads re TH/ST LCR models.

At work, we have LCR and high-voltage testers from Sourcetronic.
What's more, Sourctronic is based in the same city...

[Kean]

When first looking for a device myself, I considered an R&S HMC8015 as I was also buying my RTB2000 and some other gear from R&S, but the HMC8015 was too expensive.  Its replacement the NPA series is even more expensive, especially if you want the enhanced features beyond the basic NPA101.

The other brand I looked at was Everfine - e.g. PF9810/9811.  They were recommended to me by Dave's mate Doug Ford.  I don't think you will find a local distributor, but maybe check them out at least to see the features they offer especially the test report generation.
https://www.everfine.net/Product_62.html
https://www.everfine.net/SolutionDesc_77.html

I was assuming this was for business use, but if you wanted something for personal use you could consider a second hand unit...
e.g. Voltech PM100 used to be popular and often appear on eBay.

Most of the Voltech models only have a simple LCD display for UI, but can be queried over GPIB or RS232 (depending on installed options) for additional info including harmonics.  Dave even did a video on this single phase model.  The devices are no longer supported and the software is no longer available, so getting additional data might involve significant research & programming.  I also never found any protocol/command documentation when looking at these several years ago which is why I looked elsewhere.  I am not sure if you could query the harmonic dataset using something like TestController.

FYI, I wrote simple python code to automate LED driver testing by controlling a HMC8042 PSU (0-10V dimming control) and querying the TH3321 and some HMC8012's or a 34972A for performance/thermal/efficiency data.  AC power was from the mains or via my IT7321 programmable AC power supply.  Everything was controlled via SCPI over LAN, except the TH3321 which was over RS232.  That was all pretty simple to do.  As mentioned, I had planned to generate nice reports like the Everfine software can generate, but I never got around to querying THD data.  I just reviewed the harmonics graphs on the display.

[danils]

What about HIOKI smaller models? Second hand on the bay at low prices. They have LAN connection and some decent analysis software.
I have a 3336 but it's above 1000€

[DaJMasta]

What about HIOKI smaller models? Second hand on the bay at low prices. They have LAN connection and some decent analysis software.
I have a 3336 but it's above 1000€

The 3332 I've got isn't!  Very handy even in general troubleshooting, and the associated pass through fixtures are generally brand-agnostic.

[jaymac]

I know you haven't asked about it, but I have the GW Instek 8310 although I looked at the 8213 as well.

The main reason for going with the 8310 instead of the 8213 was the higher test frequency bandwidth (the 8213 has a bandwidth of DC - 6KHz, the 8310 is DC - 100KHz with a 300KHz sampling rate). Both models have RS232/USB/Ethernet and support SCPI. The 8310 is a neat unit that covers my needs very well.

I'm only a hobbyist but after a debacle at work involving power measurement piqued my interest I decided to do some testing at home. I looked at a bunch of others but ended up buying the 8310 new. FWIW I'm very pleased with it. It's simple to use... capable... accurate and it will meet my needs for a LONG time to come.

I don't think you'd regret the purchase of either of the GW Instek models.

In the UK the 8213 is £692 ex vat, the 8310 is £1219 ex vat (Farnell UK).

Cheers.

[Martin72]

Hi,
Yes, it will come down to a “race” between GwInstek and the Sourcetronic models.
I'll have to work my way through their data sheets.
A used Hioki or Yokogawa would also be appealing, but that would be a game of chance in terms of availability at the moment I want to buy one.

At work, I'm used to something completely different; they would be overdressed in any case.

https://www.zes.com/en/Products/Precision-Power-Analyzers/LMG671

https://www.zes.com/en/Products/Precision-Power-Analyzers/LMG450

[joeqsmith]

None of your scopes offer software for this?   

Stand alone, I have used HIOKI and Dranetz.  They won't meet your target price but it's not all about cost.

***
Or just roll your own custom software like I did.

https://youtu.be/04I7nHA_HxM?t=451

[2N3055]

Hi,
Yes, it will come down to a “race” between GwInstek and the Sourcetronic models.
I'll have to work my way through their data sheets.
A used Hioki or Yokogawa would also be appealing, but that would be a game of chance in terms of availability at the moment I want to buy one.

At work, I'm used to something completely different; they would be overdressed in any case.

https://www.zes.com/en/Products/Precision-Power-Analyzers/LMG671

https://www.zes.com/en/Products/Precision-Power-Analyzers/LMG450

GPM-8213 is simpler 5kHz model. 
ST3321 (and it's peers) have 8kHz BW and have more measurements, waveform mode etc. They have pretty much all measurements that GPM-8310 has, except GPM-8310 has 100kHz BW.
From what I can see GWInstek has much more refined UI, an both GW instruments have better resolution.

GPM-8310 seems to be very capable instrument. OTOH ST3321 is low current model, with 1µA resolution on it's 1mA full scale range. It goes to max 2A.
Both GW Instek have better resolution even with larger current range.

Both GW Instek and Tonghui do DC measurements also.

But as Joe asked, if this is for your home lab, with adequate probing solution you can get lot of these measurements on a scope.
Harmonic, distortion, power measurements for instance. I know you know that, better than me.

What is specific thing you want to solve with this? Seamless current range changes and built in isolation? Standby power measurements?
Simplicity of connecting and having all the measurements at the same place? Better precision?

[Martin72]

Yes, yes and yes.

Of course, you can do most things with the software already installed in the scope.
Build a few adapters for that, a current clamp.
In this case, however, I would like to have a stand-alone solution for devices that are already “finished” because I am only interested in their power quality and I don't need to block an oscilloscope for it.
The PA option in an oscilloscope will not be unemployed, it is indispensable for development work on switch-mode power supplies, because, for example, higher (switching) frequencies and the power losses at semiconductors or chokes/transformers cannot be measured by even the best power analyzers, at least not “out of the box” and then the bandwidth also plays a role.

[tautech]

Yes, yes and yes.

Of course, you can do most things with the software already installed in the scope.
Build a few adapters for that, a current clamp.
In this case, however, I would like to have a stand-alone solution for devices that are already “finished” because I am only interested in their power quality and I don't need to block an oscilloscope for it.
The PA option in an oscilloscope will not be unemployed, it is indispensable for development work on switch-mode power supplies, because, for example, higher (switching) frequencies and the power losses at semiconductors or chokes/transformers cannot be measured by even the best power analyzers, at least not “out of the box” and then the bandwidth also plays a role.

Or redirect investment to an active current probe to utilize the SAPBUS on your SDS3034cough 3104X HD.


I must get around to a proper play with my SCP5030A....some miserably wet weather is coming....then there's also XY mode and a Ferobend download.....also Mike Wyatts AWG curve tracer and the darn nice 1.5 GHz passive probe to play with.....

[Martin72]

Or redirect investment to an active current probe to utilize the SAPBUS

You can gladly sell me an SCP5030A for the price of one of the GwInstek/Sourcetronic power meters mentioned here, I certainly won't say no to that.

on your SDS3034cough 3104X HD.

3104 is correct.

[joeqsmith]

In this case, however, I would like to have a stand-alone solution for devices that are already “finished” because I am only interested in their power quality and I don't need to block an oscilloscope for it.

The scopes, like any other generic tools I have, I don't dedicate them to one specific task.  To setup and take these measurements,  I connect the scope with two BNC coax cables, run the software and plugin my DUT.  Bingo.  After I make my measurements, I put things away. 

Standalone, I like the Hioki (I've shown it here somewhere).  3-phase plus spare channel (which has been handy).   Can log for weeks to a month at a time (depending what you are after).  For fast transients, can collect at I think 2MHz.  When I first looked at getting one, I explained to their sales group that I was wanting to capture transients with it.  "Sure, it can do that".   It needs to survive.  "Sure it will survive".   Then you won't mind sending me one to run through all the standard IEC tests independently.  "Sure we can do that".   They showed up and I gave them a tour.  We then proceeded to capture several  transients.  We bought a few of them.  About $10k USD/ea.  Not sure today.  They have paid for themselves. 

[pdenisowski]

If you're looking for quality and are flexible on the cost, we (R&S) make a single phase power analyzer - the NPA

https://www.rohde-schwarz.com/products/test-and-measurement/rs-essentials-meters-and-analyzers/rs-npa-power-analyzer_334255.html

I also made a video on how to use it

[pdenisowski]

In this case, however, I would like to have a stand-alone solution for devices that are already “finished” because I am only interested in their power quality and I don't need to block an oscilloscope for it.

It depends on the scope and the power analyzer, but in my experience there are a few power measurement tasks (like efficiency) where power analyzers may be substantially more accurate than oscilloscopes.

[2N3055]

In this case, however, I would like to have a stand-alone solution for devices that are already “finished” because I am only interested in their power quality and I don't need to block an oscilloscope for it.

It depends on the scope and the power analyzer, but in my experience there are a few power measurement tasks (like efficiency) where power analyzers may be substantially more accurate than oscilloscopes.

While I agree with you in general that in practice there is very little practical difference, there is quite more precision to be had, just by virtue of many current ranges and  basic specifications of the instruments.
GS-Instek GPM-8310 is specified at 0,1% on most measurements.

Getting a scope to do that is going to be stretch.

OTOH, I use scope same as you. Because it is good enough for my purpose of what I need to check.

[Martin72]

Getting a scope to do that is going to be stretch.

In addition, the measurement tolerances of the current clamp/transformer and the differential probe add up, so that the “scope solution” does not even come close to the accuracy of a dedicated power meter.
This can easily add up to 5% or more in total, which would be far too much for (e.g.) an acceptance test.
Does this mean that a scope's PA function is just a gimmick?
No, definitely not.
You have to make a clear distinction: the PA function is helpful when developing a power supply.
For example, a power meter can only measure the total losses of a system, whereas the PA function goes “deeper” and can show losses at individual components.
The range is also quite different, a small “trap” for power analyzers that can display ripple.
The PA function can capture most of what a power meter/analyzer captures, albeit with significantly greater tolerance.
Conversely, it becomes difficult or even impossible.
In my opinion, both “systems” are therefore justified in my opinion; it just depends on the focus of the respective measurement task.

[egonotto]

Hello,

RBBVNL9 has a GPM-8310.
There is a thread from him: “https://www.eevblog.com/forum/testgear/(not)-measuring-inrush-current-with-the-gw-instek-gmp-8310-power-meter/”

And a video: “https://www.youtube.com/watch?v=-fdJgzNO0Wo&t=5s”

Best regards
egonotto

[Martin72]

I'm bringing this up again because I haven't bought an analyzer yet.
But maybe I don't need to, depending on how much I can trust what's been lying dormant in my scopes for years...
I tried it out today, but first I built an adapter that can measure current and voltage.
Teko offers various connector housings, and I chose a type that is not too large but can still accommodate the four required sockets.
I used the scope itself as a test object.
It all looks plausible; I just need to verify it with a “proper” analyzer (borrow one somewhere, or take one from work if necessary), then the search might be over.
Input current harmonics can be displayed up to the 40th harmonic, which would be sufficient for me.
Let's see before I consider the matter closed.

Martin