weird DMM ohms readings on transformer

I came across a weird issue I've never seen before which is baffling.

My scope is down for the count so I can't look at the waveforms coming out of the meters which may explain this.

I was measuring the primary resistance of a 10kVA transformer. I tried a Fluke 87 V. The meter freaked out and was showing negative OL and all sorts of garbage. With autorange on, it was unable to sort of measure the approximately 315 ohms. Manual ranging worked, part of the time.

Ok, time for a new battery. Still no good.

weird, the meter must be blown out. Next I tried the known good Fluke 73 series 2. Same problem. It can't autorange at all, but will sometimes read from 200 to 300 ohms if manually ranged, and if you let the meter sit for about 10 seconds.

Any other transformer or anything with resistance works can be read fine.

Now it's time for the HP 34401. No problem, the X1-X2 resistance in 315 ohms, in autorange or manual mode.

So is the resistance range on a DMM really some sort of AC signal?

I tried the same test with the H1-H2 terminals shorted out. Neither Fluke meter can autorange, but if set manually, they show about 600Ohms which then drops to about 310-320ish after a while.

The HP meter shows the same.

Without the scope, I'm guessing the ohms range is not really putting out a clean DC current, or there's some interference going on with the integrator in these meters and whatever the inductance is of this one transformer.

Has anybody come across this before? It's never crossed my mind to second guess ohm readings on a DMM as long as the DUT is off.

Reply to
Cydrome Leader
Loading thread data ...

Yup, seen it. Try shorting the secondary before measuring the primary DC resistance. That should get rid of all but leakage inductance.

It's down to rate of current rise versus integrating time.

--
"Design is the reverse of analysis" 
                   (R.D. Middlebrook)
Reply to
Fred Abse

Why would a 10 KVA transformer have 300 Ohms resistance? That is WAY too high! It should be well less than one Ohm, I'd think, unless this is for UHV transmission lines (and they don't mess with tens of KVA on those).

Jon

Reply to
Jon Elson

I didn't say it should. You attributed someone else's quote to me.

--
"Design is the reverse of analysis" 
                   (R.D. Middlebrook)
Reply to
Fred Abse

It's the 14.4kV winding in a distribution transformer so several hundred sounds ok, but doesn't quite jive with the %Z rating unless I'm doing something wrong with the math

What's a tech if the field supposed to do when they hit the "improbable" combination of something their meter just doesn't work with? Heck, in all the fluke sales stuff it's a man in a raincoat probing some sort of motor control at an oil refinery or something rugged sounding.

I'm now wondering if there are magic capacitances that cause bogus results or intense confusion with these meters.

Cameras sometimes have published "grey zones" where metering or certain modes functions just don't work as expected. They should have these for test equipment as well.

Reply to
Cydrome Leader

Sorry, it was the OP (cydrome leader) who gave that very odd reading.

Jon

Reply to
Jon Elson

Shorting the secondary before taking a reading seems to belp, but it still takes over a minute for the reading to settle down. The lower ohms ranges settle much faster than the higher ones.

I need to replace the fuses in the analog meter and see how fast it settles. I'm going to guess it runs at a higher voltage and will build up the magnetic field faster than whatever these digital meters are doing.

What other goofy effects have you seen with these other instruments?

Reply to
Cydrome Leader

OK, that would have a LOT of inductance, then. I imagine many digital meters may screw up with that big an inductor. Probably any D'Arsonval meter should do a much better job.

Jon

Reply to
Jon Elson

Back in the day, the most frequent howler was attempting to set the thyristor-controlled field current (5 to 10 amps), of a big motor controller, using a DMM. People in the know would have a moving-coil, or moving iron, ammeter just for that job. The man in the raincoat would blow expensive fuses.

My Fluke 87V gives really silly readings on capacitance, with capacitors which have seriously deteriorated ESR. It applies a known charging current, and measures time to charge. It's capacitance function isn't of much use.

Knowing how your instruments work, and what they *actually* measure is essential.

In your case, with that particular transformer, I'd use a DC PSU, a voltmeter, and an ammeter. I'd short the secondary, too, to reduce the nasty inductive spike when disconnecting the setup.

--
"Design is the reverse of analysis" 
                   (R.D. Middlebrook)
Reply to
Fred Abse

I have a couple Protek meters that I really like. But, the autoranging can sometimes screw up quite massively. One time, I was checking the

+5 V on a board with some logic. The meter gave an astounding reading of 125 V or thereabouts! I knew this had to be wildly incorrect as the circuit was mostly working, rather than exploding in flames. I locked the range and it gave a somewhat more reasonable reading, but still quite off. I used another meter and got 5 V on the nose. Well, the fact the Protek went crazy on this circuit told me there might be some high frequency noise on the +5, so I added some caps.

That't the most ridiculous reading I've ever gotten from a DVM.

Jon

Reply to
Jon Elson

Come back AVO8, all is forgiven !

Reply to
Baron

I still have one, and it's still within calibration tolerance.

--
"Design is the reverse of analysis" 
                   (R.D. Middlebrook)
Reply to
Fred Abse

Fred Abse scribbled thus:

Yup. There's a reason that they are/were industry standards. If I remember correctly the model 7 was the low Z one, 200 OPV.

--
Best Regards: 
                     Baron.
Reply to
Baron

I need to rescue a Simpson meter. I'm no old timer, but I'm quite skeptical of these digital meters. There's apparently so many weird things that they don't cope with. I've "discovered" problems with each and every one I've used. Analog meters just seem to lack the surprises.

It really is.

I don't doubt any analog meter ever made would have no problems either.

It's just sort of sad that what should be a simple test can still require a lab style setup to get a reading.

I should try some power supplies in constant current mode on this thing to see if any get confused due to their switching power supply and whatever microcontrollers runs the front panel and some magic resonant frequency that causes self destruction.

Once I ran into a bizarre resonant condition with an induction motor and a variac. At at certain voltage (or maybe motor speed) the variac saturated. This was light dimming, horrible buzzing and a blown fuse, but only at a certain dial position.

Reply to
Cydrome Leader

That's pretty impressive. Did it give equally amazing AC readings too?

Reply to
Cydrome Leader

No, I think the AC mode has a low frequency cutoff, and this was likely very high frequency noise.

Jon

Reply to
Jon Elson

Never had a 7. ISTR they were 1000 OPV. Certainly too low for electronics.

I was fortunate enough to get an 8 MKIII, on a trip to England. Bakelite case, stud-and-leaf switches, and tropicalized resistors. Later models had thick film resistors and printed circuit switches.

--
"Design is the reverse of analysis" 
                   (R.D. Middlebrook)
Reply to
Fred Abse

Fred Abse scribbled thus:

Hi Fred, Happy new year. I used to have a model 7 until it got borrowed. Currently I have two Model 8, a MKII and MKIII. I did have an ex mill one in a leather case, that one got stolen when I was burgled. The swine also took my scope and spectrum analyser along with a lot of other gear.

Never did get any joy from the police, but that's another story.

--
Best Regards: 
             Baron.
Reply to
Baron

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.