EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: cvanc on January 30, 2018, 04:16:49 pm
-
I know tons of meters have a min/max function, but do any of them actually have the ability to catch and document that brief dip of the AC line when you turn on a heavy load?
I am under the impression that most DMMs do not respond fast enough to do this, but i'd love to know if there are any that do. Thanks!
-
I know of some benchtop Flukes which have selectable sampling rates (with also changes the resolution). Like the Fluke 8808A.
The sampling rate -and this is from the top of my head- is like 10 or 15 a second, so it could catch some brownouts if it also has a min/max function.
-
Depends on how long the brown-outs that you are looking for lasts, several seconds or longer could be detected by a "normal" dmm's (3 or 4 readings per second).
If shorter you need a higher sample rate, perhaps a ac transformer+resistor divider and a audio recording device?
-
Gossen Metrahit Energy has some power quality logging, I have never used it and do not now exactly what it can and cannot do.
-
Fluke 289 has quite an interesting way of logging data. It doesn't just take an instantaneous sample at each interval; instead, it continually makes measurements even between the regular sample times, and can store extra data points which indicate when something interesting has happened.
This could be ideal for your needs. Set it to sample, say, every 10 seconds just to get a general trend, safe in the knowledge that if there's a brief interruption in suppply, it'll get captured too.
It's an expensive DMM, so I'd recommend testing it out for yourself before committing to it, obviously.
-
A bench DMM like the 3446XA should be able to do this. Get on AC Volts and set a trigger on AC Volts lower than nominal. It can digitize way faster than 50/60Hz so you should be able to get a nice waveform (need to check if you can start digitizing before the trigger - haven't looked into it).
Otherwise, you can use a scope with a HV probe and draw a template and trigger on not meeting the template.
Laurent
-
What about one of the many graphing multimeters?
-
Keysight's U1272A/1273A and U1282A have datalogging, which can be programmed for a fixed rate (at every "X" seconds) or via AutoHold (if the input signal varies more than a given number of counts).
Both could be used to detect brownouts, although the second method would have much less data to deal with.
-
I went through the same thing. Found out soon it would take an very expensive meter to do this. I ended up recording min max every 24 hours for a year using a $50 meter (5 samples per second). I think I "found" two brownouts in a year. But there were hundreds of brownouts during this time. Finally the utility fixed the problem. My point is it is almost impossible to find them using a normal DMM.
I think the "scope" method would be the cheapest solution, even if you had to buy a $400 scope. I was going to do this but then the brownouts stopped.
Good luck
-
I just checked the manual of the Energy, the PQ mode is basically made for this type of stuff:
The following types of disturbances are recorded:
• Undervoltage (LoVoLt) and overvoltage (HiVoLt) with start time,
duration and extreme value.
• Under and overvoltage of the half-period RMS value (dips and
swells) with start time, as well as minimum and maximum
values.
• Momentarily exceeded values with a duration of greater than
1 ms (peak) with time of occurrence and maximum value
• Steep transients with a rise time of 0.5 to 5 ms within a range
of 200 to 1000 V including time of occurrence, relative voltage
value and the previous 1 ms instantaneous value
-
I have a Dranetz 626 that you can have for the shipping cost. I got it from a going out of business deal along with a bunch of other stuff about 15 years ago and never powered it up. I’ll have to open it up and see if it is worth shipping. There was a thread about this on the forum that showed some extensive battery damage. If you want it, PM me; I can take some pictures after I dig it off the bottom of the pile.
-
I went through the same thing. Found out soon it would take an very expensive meter to do this.
I think the "scope" method would be the cheapest solution, even if you had to buy a $400 scope.
$1 of resistors as voltage divider, an old PC, a copy of Audacity.... :popcorn:
-
I use a HIOKI brand meter to make these measurements. Not something you would want for a home hobby use but if you really need something like this, I would recommend checking out their products. I have been very happy with the ones I have used.
-
I used a Mooshimeter to measure brownouts / voltage sagging at my house.
-
People are talking all about sampling rates and logging but are not considering the very real implications of how AC voltage is measured by DMMs.
Most DMMs use what is effectively an analog computer to convert RMS voltage into an equivalent DC voltage. There are various techniques to achieve this but all involve some averaging of some intermediate signal (to realize the "mean" part of Root Mean Square). The time constant of this averaging operation will limit how fast the RMS converter can respond, as well as the lower useful limit on frequency. Even RMS voltmeters based on thermal converters have exactly the same issue, with the thermal time constant replacing the electronic averaging. Computing RMS meters, which sample at a high rate and directly compute RMS from many samples, can also have a long averaging time, though it may be configurable to be as little as one cycle. Even then, you can only know what happened as function of the average effect to one cycle at best.
If you really want to catch a loss of a cycle or two (e.g. between mains loss and UPS kick-in) or a dip or spike lasting a cycle or two, you probably can't really depend upon an RMS conveter to respond quickly enough. Even if it does show a change, it will not represent the true nature of the deviation in the AC waveform. The RMS reading is not inaccurate per se, since RMS voltage is defined as the square root of the average of squares, over some time period.
So instead of measuring RMS voltage and looking for a change, you probably want to measure the waveform and look for a deviation from the norm.
With that in mind, I'd look at using an oscilloscope, with an envelope triggering function. Basically, you define an envelope around the normal AC waveform (maybe +/- 10 V). The oscilloscope triggers normally and continuously, but will stop and capture the waveform data if any part of the digitized waveform goes outside that envelope boundary. Make sure to use a safe probing technique and consider a low pass filter of some kind to prevent high frequency spikes from causing false triggers. Obviously don't make the low pass filter corner frequency too low or it will cause the same problem as the RMS converters.
-
A matter of using the right tool for the job. The HIOKI I use has different ways to collect the data. I can say for example, collect data once every hour but if something falls out of spec (and those triggers can be fairly complex) then collect that. The unit can measure three phase for both current and voltage and has a spare channel (which has saved me more than once). While I benchmark handheld meters for how electrically robust they are by exposing them to ever increasing transients, the HIOKI is not only designed to survive them but capture them. An analyzer like this is not something I would have a use for in my home hobby but when I work on CAT III looking at the AC line, this is my tool of choice.
-
I went through the same thing. Found out soon it would take an very expensive meter to do this.
I think the "scope" method would be the cheapest solution, even if you had to buy a $400 scope.
$1 of resistors as voltage divider, an old PC, a copy of Audacity.... :popcorn:
:--
This would be *NOT SAFE AT ALL* unless one sets up a high impedance differential input using both stereo channels each as one of the differential inputs, or (the right approach) uses a divider of some kind and a reasonably wide bandwidth and high voltage isolating transformer.
-
The Dranetz Power Disturbance Analyzer is the way to do this job. We used to use them to monitor computer facilities back in the day. There are a bunch of them on eBay.
I would be concerned about the availability of printer ribbons and paper.
-
That thing looks cool. From looking at a picture of one I'd say there's a reasonable chance that it's just standard thermal paper as used in receipt printers. If that's the case then it should be widely available still.
-
$1 of resistors as voltage divider, an old PC, a copy of Audacity.... :popcorn:
:--
This would be *NOT SAFE AT ALL*
unless one sets up a high impedance differential input using both stereo channels each as one of the differential inputs,
or (the right approach) uses a divider of some kind and a reasonably wide bandwidth and high voltage isolating transformer.
LOL, the popcorn indulging gent did specify 'an old PC' to nuke,
a 'copy' of Audacity in case the original turns to vaporware,
and total DIY mythbuster cost $1 ;D
Anyway, whatever way used to pad down the AC signal to one volt or less, preferably via a 120 > 12 volt isolated wall wart FIRST etc
one had better use a fully floating 2 wire in- 2 wire out audio transformer (stepdown or 1:1 etc) to feed the concerned old PC :scared: or hesitant laptop :o
with the appropriate lead/cable adapter to feed into the 3.5mm socket and stand back
-----------------
FWIW I've caught brief and long brownouts (and simulated ones with loads, variac and switch scenarios) etc easily with both Fluke 289 (and Fluke 189 with it's logging system)
For field use without any extra gear, 289 with its graphical trend display is a winner for straight up 240 volt monitoring.
And coupled with another meter set to min-max or PEAK min-max, you can pretty much catch any elusive or intermittent extreme brownout or surge etc good enough for Australia :-+
Just don't expect major meter miracles to happen if you're not familiar with the Fluke 289 logging modes,
and or rock up to an extended period logging job without a pack of fresh batteries = short SOL session :-[
-
How about you get a small 2nd hand UPS which will both allow you to protect equipment from the brownouts and log the line voltage. Also it will be safer. I'm sure you would be able to do this from the apcupsd open source software for APC UPSs. It does have logging built in which I expect can be configured to log these events. You could use a Raspberry Pi (or equiv) if you want to avoid having a PC on 24/7. Perhaps a new-ish model of UPS may log these events itself. You can pick up a 2nd hand UPS cheaply, usually just need the battery replacing.
-
How about you get a small 2nd hand UPS which will both allow you to protect equipment from the brownouts and log the line voltage.
My higher end UPS logs exactly when it has to switch to online mode but it does not keep details about what the power anomaly was.
-
I'm sure you can do it with apcupsd, but there is a lot to read through!
The status message includes min/max line voltages since the last status message. The UPS also records when it last switched to batteries and a reason code, e.g. "LASTXFER : Low line voltage"
http://www.apcupsd.org/manual/manual.html#apcupsd-status-logging (http://www.apcupsd.org/manual/manual.html#apcupsd-status-logging)
-
The Agilent (Keysight) 34461A and it's older sibling, the venerable 34401A both set integration time for a measurement in power line cycles (NPLC). I know both go below 1 NPLC but that may be for dc. For ac, the 34461A can do 50/sec and my old 34401A may be close to that.
Should give a good picture of the dropout. You could even watch the load current (on a separate run) to see how it behaves (after checking out the specified limits, of course).
-
I opened the Dranetz 626 that I offered and it looks very nice inside. There is an internal battery consisting of 10 sub C NiCads that hasn’t leaked. It has a power supply module that takes 115 / 230 volts switchable and a 12VDC proprietary looking receptacle that could be changed. It has a 3 phase monitoring card and a communication card with RS232. But wait, there’s more; room for more cards, that is plus it powers up. And a part roll of printer paper that appears to be thermal paper.
-
And a part roll of printer paper that appears to be thermal paper.
It a very standard size of thermal paper. Any good office supply store will have it.
-
A matter of using the right tool for the job. The HIOKI I use has different ways to collect the data. I can say for example, collect data once every hour but if something falls out of spec (and those triggers can be fairly complex) then collect that. The unit can measure three phase for both current and voltage and has a spare channel (which has saved me more than once).
Hi Joe, I know this is an old thread but do you mind telling me the model of the Hioki meter you are using?
-
Back in the day, they used a recording voltmeter/chart recorder.
This was usually an analog voltmeter guts driving a pen on a continuously moving chart.
It required the replacement of the ink & paper roll from time to time, but gave you a permanent record
with elapsed time since it was turned on.