Slightly OT: Diagnosing AC and DC systems simultaneously.

This is going to sound pretty simplistic to some of you, but my expertise is in networks and controllers, not in electrical troubleshooting, so I'm looking for some advice.

I have a system that includes a sturdy Murr Elektronik 10A 24VDC power supply with 120VAC single-phase input.

The system is undergoing a cyclic instability in which the 24VDC supply to the controllers is dropping out and coming back on again repeatedly.

What I need to figure out is whether the 120VAC supply side of the power supply is dropping, or whether the 24VDC side is overloading, protectively shutting itself down (or just reducing the output voltage), and recovering.

What sort of instrument can I use to simultaneously capture or trend a single-phase 120VAC signal and a 24VDC signal ?

Can I do this with a digital storage oscilloscope, with two separate, isolated probes ?

You can do it with a scope depending on the scope and how long you need to store for?

My guess would be the 24Vdc rather than the 120, can you swop the 24vdc out to see if the problem goes away ?

If money be no object,

http://us.fluke.com/usen/products/features.htm?cs_id=31768(FlukeProducts)&category=LOGINSTR(FlukeProducts)

I really like the Fluke meters, and every dime I've spent on one (well, that my employer has spent on one) has been well worth it.

You can set up a digital stroage scope to trigger on an event. Like the voltage dropping.

Having been thru this a few times, I suggest you get your clamp on that reads DC current and measure the 24VDC amps. Compare the reading to the power supply rating.

If you do not have a DC clamp on, get a high wattage 1 or 0.5 ohm resistor and put it in series with the supply. Then you can monitor the voltage drop across the resistor to get the actual current.

IMO you will find a 24 VDC problem and not an input supply one. Usually if you are having an AC input power problem, some other AC powered device is also having issues.

My guess would be that you have an intermittent short on the dc side. Most power supplies shut themselves off when a secondary short is detected, and turn back on automatically when the short circuit is removed.

probably the power supply. most 24vdc power supplies are regulated, that is they try to keep their output voltage at nominal rating, even with input variance. if possible can you measure the current output to make sure there is not too much of a load on the circuit?

I have used two fluke 189 dmm which have a datalogging function that has pretty fast response time (not oscilloscope fast, but maybe you can get a couple of these on loan or pretty cheap), to measure to different circuits for voltage drop. I just used the very nice datalogging software you can get from Fluke to sync the time base of the meters, downloaded the recorded data and was able to tell what happened first, which in my case was, did a tension transducer output spike, or did the drive surge and cause this tension response? If you don't need an extremely fast response time, these could work.

controlled said:

My guess would be that you have an intermittent short on the dc side. Most power supplies shut themselves off when a secondary short is detected, and turn back on automatically when the short circuit is removed.

Click to expand...

I agree completely.

SIMPLE TEST: Remove the 24VDC load and let the power supply sit there. If all is okay, add branches of loads to see when the problem occurs. (Of course this may not be practicle depending on the machine or automation. Perhaps you can run this in manual with all loads added one-by-one until the problem occurs.)

You did not mention all the aspects of the system. If you have an oscope then yes it will tell you BUT if not a cheaper option would be connect the controller(s) with another 24vdc power supply (independent) and monitor if/when online i.e. input a signal from the suspect PS and time stamp if lost.

If both PS seem to fail then the 120vac may have a problem, a UPS may be connected and provide info to show 120v is missing and UPS is online.

Just depends.

Rig up a micrologix (or whatever you can get your hands on ) Hook the 120 supply up to an input and have it latch a bit if the input turns off.

Whether or not this will be fast enough I guess would be a source of debate.

I like the fluke 190 series as well. I have had both the 120 series and 190 series meters. The 120 series is about 1/3 the price. The MAJOR difference is that the 120 has a shared common for both inputs (great for blowing stuff up be careful), a smaller b/w screen and is slower (20 or 40MHZ). So in this case it would not be able to monitor both the 120 and 24 simultaneously.

The max/min function on many common multimeters might be good enough for what you want to do as well

I'm going to guess you already have inputs to a PLC or something. If it is DC then hook the power supplies DC side to an input. Then take another cheap power supply on only hook it to the plc input to monitor your 110VAC power. Monitor with the PLC

Thanks for all the excellent ideas, guys. I especially like the idea of a DMM with a data logging function; I will look into that. I am unfamiliar but intrigued by DC clamp-on ammeters.

Leadfoot and Ron; I should have put more information in about the system overall, but I was trying to focus on the measurement method rather than the troubleshooting technique.

The problem is repeatable and not intermittent. This is a system that switches from line power to a generator. When the generator comes online, the system automatically restarts many of the loads (fans and pumps, mostly) and about 3 seconds later the 24VDC power fails. It comes back on almost immediately, but then 3 seconds later it fails again. This cycle repeats every 3 seconds, for 5-15 minutes.

I really only need about 3 seconds of data to show what's going on.

For commercial reasons, I need to prove directly and not by inference that the 120VAC supply is stable at the time the 24VDC supply fails. I only supplied and specified the 24VDC side of the system. If it's my power supply failing despite a sufficient supply voltage, I need to replace that supply and maybe change the design.

By the way, I begged the designers to put the 24VDC supply on a UPS when they built this thing. I get one chance to fix it, as it's very remote and very complicated to shut down.

Thanks again !

Since you mention a generator, you might also look at the HZ. Is the generator bogging down and dropping below the allowable HZ of the power supply?

May be able to use a plain old Fluke 90 series for this. They have a frequency meter, true RMS, and DC capabilities but back when I worked with gen sets I liked using plain old D'Arsonval meters when checking for voltage fluctuations. The 90, or any frequency meter should offer measurements to provide if stable, the genset should have one mounted on the panel.

Power supplies will usually work over a wide range, if a genset is fluctuating that much a 120vac light will show it. When synchronizing gensets I just used 2 light bulbs connected across the breaker, when speed was matched the light(s) would slowly go out...you engaged the breaker when off. If out of phase the 2 lights alternated.

I know it does not sound professional but ifn it was me I would have a light bulb on the 120vac, a true RMS meter with frequency on the 120vac, and a D'Arsonval type meter on the 24vdc and watch it during a changeover....assuming you can have them make the switchover for the test.

Not knowing what the genset supplies it could be anything, harmonics from electronics can affect a PS, the main system may not be as affected as genset supply. It could be a poorly designed ground system allowing feedback. I would not suspect your PS if it works fine with utility power.

Another option may be to "load" and "quality" test the genset system. This can normally be done while not being connected to the machine in question.

I am just throwing ideas around, if a genset is fluctuating causing problems it should be very noticeable. If it a real power issue then a scope may not offer the answer unless you have an idea where too look, recording scopes ain't cheap either.

rsdoran said:

Power supplies will usually work over a wide range, if a genset is fluctuating that much a 120vac light will show it. When synchronizing gensets I just used 2 light bulbs connected across the breaker, when speed was matched the light(s) would slowly go out...you engaged the breaker when off. If out of phase the 2 lights alternated.

Click to expand...

Ingenious but probably the best way. Did that come from your days at Doranland?

Is it possible to power the 24 vdc power supply from a different source when the power transfer takes place? I would put the dc power supply on a stand-alone ups then switch to generator power. The dc power supply shouldn't blink. If the problem is on the 120 vac side, I would be looking at the transfer switch. The latest and greatest ones have a bunch of parameters to configure depending on the generator model.