Some questions about power analyzers

[mindcrime]

Gang:

My protestations to the contrary in the TEA thread notwithstanding, I might buy a power analyzer at some point. I had not even considered this until recently, as I had thought (to the extent that I think about power analyzers at all, which is nearly nil) that a power analyzer was a VERY specialized, esoteric bit of kit, very niche, with little day-to-day application.  But Dave's recent video on the Hitachi TV repair inspired me a bit.





I see that being able to observe the power consumption of a device even when it's not displaying any obvious activity could be useful for debugging something that's misbehaving. And now that I'm thinking about it, I can imagine other times when you would want to know details about the power consumption of a device. So now a power analyzer seems like something that I could use, although clearly not something that would get used every day.

That said, I don't know much about the things, so I'm not sure what I would buy. I poked around Ebay a bit and it seems that the main brands that are "out there" are Voltech (what Dave was using in the video) and Yokogawa (CRAZY expensive). There are lots of Voltech PM3000's around, but the price range is very broad. I saw one as low as $500 and some as much as several thousand dollars (USD). The guy with the one listed for $500 offered it to me for $400, but it's listed as "needs repair / parts only", and I don't really want to buy something that might not be working when it's something I know bugger-all about repairing and don't know how hard it is to service, or know anything about availability of parts, schematics, manual, etc.

Some obvious options I see are:

1. Buy a cheap, possibly not working Voltech (probably the PM3000 since they seem so common, but maybe another model?) and risk needing to service it. Worst case, I wind up with an expensive paper-weight. Any thoughts on how much would be reasonable to spend on one of these that isn't known to be working?

2. Spend a little bit more for a Voltech that's listed as being in working shape. If I went this route, do you folks have any opinions on what a good price would be for for, say, a PM3000? Or if there's another model you recommend, what a good price would be for one of those?

3. Forget the used market and buy a new power analyzer. I was looking at the Rohde & Schwarz HMC8015-G and that looks like a decent possibility if I went the brand new route. Thoughts?

4.Something else altogether, depending on what you lot tell me. Since I "don't know what I don't know", I'm probably missing some really obvious options here. Fluke and Hioki seem to be significant players in this space as well, so maybe something (new or used) from one of them? Other?  Any and all suggestions or input are greatly appreciated.

[HKJ]

I do not know what you call expensive, I use 3 different devices to analyze power:

1) Textronix PA1000, a box the size of a bench multimeter, can show power and harmonics.
2) BK Precision 9801 AC generator, supply up to 300VA at up to 250V and can show power but not harmonics.
3) Gossen MetraHit Energy, handheld multimeter that can show power and harmonics in addition to all the standard DMM functions. This meter can easily measure standby power.

They are all single phase and works for standard mains voltage in both EU and US, two of them also works with DC.

They all have their good and bad sides.

[mindcrime]

I do not know what you call expensive,

Fair enough. That Rohde & Schwarz unit is about $2800 USD, and that's right about the upper limit of what I'd be wiling to spend. Maybe round off and call it $3000 USD, max. Anything above that is probably too rich for my tastes right now.

I'll look into the ones you mentioned and see what the options are out there. The Tek one is definitely out of my price range for a brand-new one, but maybe something is available on the used market...

Thanks for the info!

[coppice]

You need to work out the functionality you need. The big spread in power analyser prices reflects their spread in functionality. Your budget would buy you a new Yokogawa WT300E, and that will let you do a lot with good accuracy. It won't do as much as the higher end Yokogawas, though. The older Yokogawas are good instruments, and they can be a reasonable price. Tektronix bought Voltech, and the PA1000 is an updated version of the Voltech PM1000+. You can often find second hand Voltechs at reasonable prices.

[egonotto]

Hello,

David can do the repair also with energy meter.

You can buy a differential voltage probe and a current probe and a deskiew fixture and a scope with Power analysis.
Perhaps a Siglent SDS2104X Plus with Power analysis.

Best regards
egonotto

[mindcrime]

The big spread in power analyser prices reflects their spread in functionality.

Sure, that part I get. What isn't obvious is why there is so much variability in the prices for instances of the exact same model of analyzer.

Your budget would buy you a new Yokogawa WT300E, and that will let you do a lot with good accuracy. It won't do as much as the higher end Yokogawas, though. The older Yokogawas are good instruments, and they can be a reasonable price.

I'll have to go back and do some more target searching. My first pass through a quick ebay search seems to have all the Yokogawa units north of $9,000 USD! Definitely out of my range. 

Tektronix bought Voltech, and the PA1000 is an updated version of the Voltech PM1000+. You can often find second hand Voltechs at reasonable prices.

Makes sense. I'm just still trying to work out what a reasonable price for one of the used Voltech units would be. The floor price for a PM3000 that isn't tagged "parts / repair only" seems to be around $1,000. Does anybody have a feel for how reasonable (or not) that is?

[coppice]

The big spread in power analyser prices reflects their spread in functionality.

Sure, that part I get. What isn't obvious is why there is so much variability in the prices for instances of the exact same model of analyzer.

Condition matters a lot, as does a current calibration certificate. I think you said one cheap unit you found needs repair. Schematics aren't always easy to find for equipment like this. Once you've repaired it, will you be able to recalibrate it? People will pay a lot more for an easy life.

[mindcrime]

The big spread in power analyser prices reflects their spread in functionality.

Sure, that part I get. What isn't obvious is why there is so much variability in the prices for instances of the exact same model of analyzer.

Condition matters a lot, as does a current calibration certificate. I think you said one cheap unit you found needs repair. Schematics aren't always easy to find for equipment like this. Once you've repaired it, will you be able to recalibrate it? People will pay a lot more for an easy life.

Yeah, easy is good!    The thing is, I don't mind repairing a piece of old equipment, if doing that is a fun and educational project in its own right. It's just that with something like this where I know very little about it a-priori, it's hard to know how hard it will be to find parts, service manual, etc. That makes me a little skittish about going the route of something that (definitely|probably) isn't already in a known good state.

I'm not necessarily worried about a current calibration certificate. Obviously I want a piece of test equipment to be as accurate as possible, but I'm just a hobbyist, so I don't need NIST traceability or accuracy to within 10^-12 percent or anything.

The good thing is, this isn't a pressing need, so I can take my time and do more research, and then bide my time waiting for a good deal to present itself. 

[mindcrime]

I feel like I'm experiencing the "Red car effect", where you start seeing red cars everywhere if you're thinking about a red car for some reason. Now that I'm thinking about power analyzers, possible uses / applications keep popping up. For example, I'm working on this reflow oven conversion thing, and the datasheet for the SSR I'm using recommends a fast blowing fuse on the load side. Sensible enough, but what amperage fuse to select? I can use the fact that the oven is a nominal 1500W oven, and that the mains voltage is a nominal 120V, and do the math and arrive at *a* value. But it occurs to me that plugging the thing into a power analyzer and empirically observing the current draw would be very nice in a setting like this. Especially for detecting things that the math doesn't easily tell you like, "is there a turn-on surge that's significantly higher than the steady state draw?" or something like that.

And yes, of course I realize that you don't need a fancy power analyzer *just* to do a current measurement. But if I had one in the lab, it would be a handy way to do something like this.

[Stray Electron]

Gang:

My protestations to the contrary in the TEA thread notwithstanding, I might buy a power analyzer at some point. I had not even considered this until recently, as I had thought (to the extent that I think about power analyzers at all, which is nearly nil) that a power analyzer was a VERY specialized, esoteric bit of kit, very niche, with little day-to-day application.  But Dave's recent video on the Hitachi TV repair inspired me a bit.





I see that being able to observe the power consumption of a device even when it's not displaying any obvious activity could be useful for debugging something that's misbehaving. And now that I'm thinking about it, I can imagine other times when you would want to know details about the power consumption of a device. So now a power analyzer seems like something that I could use, although clearly not something that would get used every day.

That said, I don't know much about the things, so I'm not sure what I would buy. I poked around Ebay a bit and it seems that the main brands that are "out there" are Voltech (what Dave was using in the video) and Yokogawa (CRAZY expensive). There are lots of Voltech PM3000's around, but the price range is very broad. I saw one as low as $500 and some as much as several thousand dollars (USD). The guy with the one listed for $500 offered it to me for $400, but it's listed as "needs repair / parts only", and I don't really want to buy something that might not be working when it's something I know bugger-all about repairing and don't know how hard it is to service, or know anything about availability of parts, schematics, manual, etc.

Some obvious options I see are:

1. Buy a cheap, possibly not working Voltech (probably the PM3000 since they seem so common, but maybe another model?) and risk needing to service it. Worst case, I wind up with an expensive paper-weight. Any thoughts on how much would be reasonable to spend on one of these that isn't known to be working?

2. Spend a little bit more for a Voltech that's listed as being in working shape. If I went this route, do you folks have any opinions on what a good price would be for for, say, a PM3000? Or if there's another model you recommend, what a good price would be for one of those?

3. Forget the used market and buy a new power analyzer. I was looking at the Rohde & Schwarz HMC8015-G and that looks like a decent possibility if I went the brand new route. Thoughts?

4.Something else altogether, depending on what you lot tell me. Since I "don't know what I don't know", I'm probably missing some really obvious options here. Fluke and Hioki seem to be significant players in this space as well, so maybe something (new or used) from one of them? Other?  Any and all suggestions or input are greatly appreciated.

   Yokogawa is the Hewlett Packard of Japan. Same HIGH quality and the same HIGH price. In fact, in the 1960s, HP and Yokogawa used to sell each other's equipment in their respective countries.  I used to have an HP catalog printed completely in Japanese!

   I quickly read your post but I'm not sure exactly what your needs are. Do you need to monitor AC power lines or DC?  Do you need to hook it up and let it run for days or weeks until it sees a glitch? Or do you just need to look for sags and surges in line voltage?  How GOOD quality do you need? Does it need to record all of the readings while it's running or only the events of interest? Does it need to be able to monitor both current and voltage or only voltage? And how many phases simultaneously?  Does it need to be Certified and Lawyer proof or just hacker grade?  If it doesn't then you could use a cheap bench meter and have it log to a computer via RS-232 or USB.

       My mother had AC power problems in her house and I could tell by her description that she had a floating neutral which let one half of the incoming line voltage sag while the other half surged, all depending on what loads were applied. I told her to call the power company. She did, SEVERAL times. But they just blew her off. I told her to call them back and demand that they put a power line analyzer on her incoming AC line. The power company complained that they would have to "order one" and that they couldn't do that. This is a MAJOR power company that serves several million customers in this area!

   I got fed up and I went to a surplus place that I knew that had an old Drantz power line analyzer (a model 658 as it turned out) and I bought it for $40 and took it home and hooked it up and started playing with it. I have to say that I was very impressed. I could connect it and let it monitor the line and look for any glitches (adjustable thresholds).  I could let it run for days with nothing found. But I could take one of my old (cheaply made) Astron switching power supplies and turn it on anywhere in the house and the monitor would instantly see an about 235V spike on the (110v) power line.

  After a few days i was comfortable with it so I called my mother and told her that I would bring it over in a few days and told her what it was and the make and model number.  She then called the power company and told them what I was planning and asked for a technician to come out and witness the test. Literally the very next day, a crew came out and pulled the power meter and the complete box off of the outside of her house and replaced all of it with brand new equipment.  She never had any power problems after that and I've never had the chance to use the analyzer "in combat" since then. But it's a handy thing to keep around! It's rather fascinating to use and to be able to see everything that can affect your home power system, and the mere threat of using it seems to get an immediate response from the power companies!

  If you want to monitor AC power then I seriously suggest that you buy a Drantz or similarly built equipment that is made specifically for that. First, it will be much safer than anything that you could build. That would also limit your liability if something happened. Second, it's such specialized gear that there's very little demand for it in the secondary market so you can buy it CHEAP. Third, it will have features that you probably haven't even dreamed of but that will be necessities when you start using it. Even an old model like the 658 is going to be a VAST improvement over a bench meter or some poorly designed and cheaply made Chinese kludge.

   I suggest that you go look at some of what's available on Ebay and then go look up some of those models and see what features they have and if there is an operators manual available for anything that you are considering.  Read the manual and be sure that it will do what you want and also that anything that you're considering from Ebay is complete and has any special probes or other accessories that are required to make it useful. One of the things that I do like about my old 658 is that for voltage monitoring, you don't need any kind of probe, you simply connect a wire to a terminal on the back of the unit. But (IIRC) the 658 only monitors voltage and not current. I have used power analyzers in the past in an industrial setting to monitor the current consumption and phasing of big electric motors but so far I haven't missed the lack of current monitoring in the 658.  Your needs may vary.

   The manual and brochure for the 658 is or used to be online. I also have copy and can E-mail it to you if you need. Drantz also makes plug in option modules that will let you monitor temperature, current and/or DC voltage. You just have to figure out what capabilities you need and what model(s) will fit that need and then look for a  unit that has the needed options or for the individual modules.

   My suggestion is that the very first thing you should do is to go find a manual for a 658 or something similar and read it until you understand the basics of power line monitoring and what they're looking for and get some idea of what the analyzers are capable of.  Even if you are working with DC power, the same principles will apply.

   

[Stray Electron]

  I looked at the OP again and I have to say that I wouldn't touch a used bit of specialty equipment that doesn't work and is "as is" unless I had a VERY GOOD idea about how it works, and only if the price is rock bottom cheap. There are also sorts of tricks used to "Guard" measurements, provide extremely high isolation and safety and a hundred other things that you're probably not even aware of. Even with a schematic (unlikely!) you probably couldn't repair it. And you have to consider the almost certainty that many of the parts in it were made specifically for it and are made of unobtainium.

   You don't need to buy NEW (unless you want to use it commercially). There is plenty of used equipment on Ebay that is guaranteed to work. Keep in mind that your time and labor are worth something and go buy a piece of gear that works and then get on with your projects instead of spending time and money trying to fix a piece of gear that you probably don't fully understand and is very likely impossible to repair.

[threephase]

Depending on your needs, there is also the option of a power clamp meter, that will cover single phase AC mains, there are also solar clamp meters that will measure AC and DC power.

Likely it will not be as accurate as the bench devices you have mentioned. Appa technology make some quite nice units, you can find them under RS Pro or the Amprobe banner. Chauvin Arnoux are another option, I believe they are under Amec in the USA.

As the others have pointed out, its defining your needs that will decide your best option.

[mindcrime]

  I looked at the OP again and I have to say that I wouldn't touch a used bit of specialty equipment that doesn't work and is "as is" unless I had a VERY GOOD idea about how it works, and only if the price is rock bottom cheap. There are also sorts of tricks used to "Guard" measurements, provide extremely high isolation and safety and a hundred other things that you're probably not even aware of. Even with a schematic (unlikely!) you probably couldn't repair it. And you have to consider the almost certainty that many of the parts in it were made specifically for it and are made of unobtainium.

   You don't need to buy NEW (unless you want to use it commercially). There is plenty of used equipment on Ebay that is guaranteed to work. Keep in mind that your time and labor are worth something and go buy a piece of gear that works and then get on with your projects instead of spending time and money trying to fix a piece of gear that you probably don't fully understand and is very likely impossible to repair.

Yep. I'm leaning towards "new" for this the more I think about it. Or if I do go used, I'll definitely go for something that is listed as "working" and by a reputable seller who mainly sells test equipment and related gear.


As far as your earlier question about requirements.... well, I'm a hobbyist and learning as I go, so I don't really have "requirements" in a very formal/strict sense. I know I'd like to be able to take a piece of AC powered gear (a TV, a reflow oven, etc.) that runs on 120VAC, connect it up and see things like the power usage in watts, peak current draw, instantaneous current draw, etc. Just single-phase, bog-standard North American 120V mains, and DC would be all I'd ever want to inspect.

I don't need NIST traceability, lawyer-proof whatever, or anything like that. I probably also don't need to be able to monitor something for weeks, although I can see how that would be useful, especially if the metrics can be logged to a PC via USB / GPIB / etc. Actually, the more I think about it, the more that sounds pretty useful. But I'm a data analytics / ML guy anyway, so my mind always starts telling me "yes, yes, yes...." in response to anything that involves collecting / logging metrics and data. 

[tautech]

Don't discount how much of this you can do with current and voltage probes and a modern DSO.

[joeqsmith]

....I know I'd like to be able to take a piece of AC powered gear (a TV, a reflow oven, etc.) that runs on 120VAC, connect it up and see things like the power usage in watts, peak current draw, instantaneous current draw, etc. Just single-phase, bog-standard North American 120V mains, and DC would be all I'd ever want to inspect.

I don't need NIST traceability, lawyer-proof whatever, or anything like that. I probably also don't need to be able to monitor something for weeks, although I can see how that would be useful, especially if the metrics can be logged to a PC via USB / GPIB / etc. Actually, the more I think about it, the more that sounds pretty useful. But I'm a data analytics / ML guy anyway, so my mind always starts telling me "yes, yes, yes...." in response to anything that involves collecting / logging metrics and data. 

P3 P4400 Kill A Watt Electricity Usage Monitor
https://www.amazon.com/P3-P4400-Electricity-Usage-Monitor/dp/B00009MDBU

Using a my old DSO to measure mains power...

[_Wim_]

Tektronix bought Voltech, and the PA1000 is an updated version of the Voltech PM1000+. You can often find second hand Voltechs at reasonable prices.

Indeed, this seems a very reasonable offer:

https://www.ebay.com/itm/274479751625?epid=6024760413&hash=item3fe844a1c9:g:OtwAAOSwt2lfTn00

[Gregg]

Mindcrime,
I have a Dranetz 626 that you can have for the shipping cost. There are manuals available on line to see if it may do what you want.  http://manuals.repeater-builder.com/te-files/DRANETZ/DRANETZ%20626%20Series%20Operator.pdf
PM me if you want it; I would like for someone to be able to use it
-Gregg

[james_s]

Don't discount the low end, there is a lot you can do with cheap modern technology. I borrow a proper power analyzer once in a while but most of the time I can get what I need from one of those inexpensive Kill A Watt devices. When I need to look at the actual waveforms I use my scope.

[BrokenYugo]

Another vote for starting cheap. A kill a watt P4400 is a very handy thing to have and will do most of what you're asking for.

[coppice]

Another vote for starting cheap. A kill a watt P4400 is a very handy thing to have and will do most of what you're asking for.

The Kill-A-Watts (I mean the actual brand Kill-A-Watt. That name has become like Hoover, and is often applied as a generic name for a mains socket based power meters) I have tried are excellent products. If you just want to monitor power consumption, they are pretty accurate over a reasonable dynamic range and a wide range of power factors. Other makes are a bit or miss. Its cheap to use a chip designed for utility energy meters, which will give pretty good results. However, a number of these mains socket based power meters are utter garbage, offering poor accuracy, and no allowance for power factor.

[joeqsmith]

Comparing the Kill a Watt P4460 with my old DSO. 

[coppice]

Comparing the Kill a Watt P4460 with my old DSO. 

I've no idea why you talk about CAT rating. The Kill-A-Watt only needs to meet the requirements of any other through feed product that plugs into a mains socket, like a time switch.

The quoted 10W consumption is probably 10VA. Its powered by a cap drop supply, so the power factor is really horrible, and its sub 1W consumption can easily be several VA. The low power accuracies you obtained look worse than the Kill-A-Watts I tried a few years ago. They were better than 1% accurate over quite a wide range of powers and power factors. Did you have the LeCroy attached upstream of the Kill-A-Watt, where it would be measuring the self-consumption of the Kill-A-Watt?

[joeqsmith]

Did you have the LeCroy attached upstream of the Kill-A-Watt, where it would be measuring the self-consumption of the Kill-A-Watt?

Yes.

[Stray Electron]

  Here's a cheap used one that works. https://www.ebay.com/itm/203019478378?hash=item2f44e7656a:g:ywsAAOSwiE9e44WO Yes, that's a floppy drive on the right hand side.

  Here's another of the same model with with the display shown. But this one is being sold for "parts" so it's not guaranteed to work. https://www.ebay.com/itm/153285701784?hash=item23b08a3098:g:UukAAOSwAQVcArcm

  A Kill-A_Watt isn't much of an analyzer but if you just want to see how much power you're pulling at the moment or the line voltage, they're good to have. 

 

[mindcrime]

Do any of these plug-in-a-plug Kill-A-Watt type devices have USB or any other way to pull data over to a PC? And can any/all of them give instantaneous current draw and/or peak current draw? Those two values seem like two of the things I'd be most likely to be interested in more frequently.