Very low mili ohm DVM attachment DIY Works Great! / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Very low mili ohm DVM attachment DIY Works Great!

1 2 next

Move Thread

LAN_403

Wavetuner

Sun May 24 2009, 02:37AM

Registered Member #1500 Joined: Sat May 24 2008, 04:38PM
Location: Ojai, Ca.
Posts: 44

Ever try to measure very close winding to unravel a seriuos multi tap transformer wired like an autransformer? Here's what worked for me: For less than $20 and a trip to Radio Shack your in for miliohm measurements in an hour or two. I now have my mystery transformer wired for 2000v - 14400v in eight steps. Than there are also two HV windinds that can be paralled for more power or in series for full voltage. The primary was the challenge. It is conservatively rated at 5KVA and has a Hugh core! Great for a plate transformer. Thanks goes out to Bob Nuckolls, Sr, Engineer, Ratheon Aircraft.

It works by using the 200mv setting on your DVM which usually steps in .1mv increments and a lm317 and simpe ohms law! Here's the link:

Very handy! I bet it would be good for making shunts too.
Hey, they use it for aircraft maintenance.

Myke

Sun May 24 2009, 05:15AM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

Neat device. I'm curious, how does the transformer to relate to this project?

You could also use a Wheatstone bridge for measuring low resistances. It would be pretty easy to zero it and make a series of markings on the front panel so that you don't have to use a calculator for ohms law.

Hazmatt_(The Underdog)

Sun May 24 2009, 06:02AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Cool thanks.

A friend of mine makes a lot of 0.3R nichrome coils and this is going to be helpful. I can make him a couple of birthday gifts out of this.

Dr. Slack

Sun May 24 2009, 08:33AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

You don't have to use a calculator for ohm's law if you make the excitation current 1A, then mV == mohm. If it's battery powered then 10mA or 100mA only needs shifting the decimal point.

The "practical" arrangement shown in that link must use very good croc clips if the residual is down to 0.2mohm. It really is much safer to use a real 4 wire measurement all the way to the component, with a pair of current clips and a pair of voltage clips where possible, this eliminates the contact resistance. The practical arrangement shown eliminates the lead resistance, but leaves the contact resistance uncorrected. Contact resistance varies with pressure, corrosion, surface finish. If anything, variable contact resistance is more of a measurement problem than the reasonably stable and zero-able lead resistance.

You could make a "2 wire" croc clip by gluing a bit of copper clad pcb inside one jaw. Then when it grips a wire, there is one independant contact on each side of the wire. A pair of those would allow a full 4 terminal measurement without the hassle of having to place 2 clips on each terminal.

If you don't make the measurements very often, then a current-limiting power supply and a mV meter a very quick to hook up.

One thing I have made effectively 4 terminal measurements for in the past (registered electrical regulation technicians please cover your eyes now), is to trace the path of UK ring mains, prior to adding extra sockets. At the meter, remove the two lives of the ring from the breaker, connect one to neutral, connect the other to my 4A limited power supply. Then I go round the house plugging my meter into each socket in turn. The gradual drop of mV identifies the order that the ring drops onto the sockets. Two sockets with the same potential must be on a spur. Pulling current from either socket reveals which is on the ring and which is on the spur.

Warning, very low DC voltage can still be confused by thermoelectric potentials on the voltage connection. The highest quality mohm meters use AC excitation to reject this error, but it makes for a more complicated circuit.

Wavetuner

Mon May 25 2009, 02:47AM

Registered Member #1500 Joined: Sat May 24 2008, 04:38PM
Location: Ojai, Ca.
Posts: 44

I used gold plated, sharp clips singles and gold bannana plugs. The measurements were pretty stable. It solved my mystery connections as they broke out of the primary in one place This was a beta transformer from Harder company. It is rated very conservatively rated at 2000-14400 5KVA. The core is almost 200 lbs! It camne from San Diego and was ship gray. It had hugh horns on it. I'm cleaning it up and bringing the now know progessive pairs through the usual phenolic plates. The secondaries can be paralelled too. Serial #1 :-^)

Proud Mary

Mon May 25 2009, 09:57AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Isn't nichrome resistance wire calibrated to within 5% per metre, or so?

This would do if one didn't mind mucking about with a Wheatstone Bridge.

But there's more than one way to skin a cat!

...

Tue May 26 2009, 06:22AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I second the point that using single crock clips completely defeats the purpose of a 4 wire measurement, it does not actually increase the accuracy of the measurement over using normal probes (since it is trivial to zero out the resistance of the probes, the error comes from the contact between the crock clip and workpiece (which in this example was about 2mR). If you need precision better than an ohm or so (2 contacts gives you about .5ohm, so I wouldn't recommend using this technique for anything less than about an ohm), you should use a real 4 wire measurement, which lets you read down to the microhm or even down to picoohm depending on the test current and accuracy of your meter.

My favorite example of a 4-wire measurement is the copper thickness probe I made last year to measure the thickness of various pieces of copper clad I had acquired

Bored Chemist

Tue May 26 2009, 07:40AM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

Can I just remind those of you who keep saying that this measurement technique isn't ideal that it did the job and the transformer windings are now identified.

BTW, I like the idea of using double sided board in a crocodile clip as a 4 wire terminal.

Incidentally, since the original purpose was to measure winding resistances of a transformer the use of AC might be troublesome.
You can sometimes eliminate the effect of thermal emfs by reversing the current and taking an average of the 2 readings.

Bjørn

Tue May 26 2009, 08:22AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

For those that measure inductors and transformers with this method, don't let the low voltage give you a false sense of security. If you disconnect the wires with the current still applied you may get a pretty bad shock from the back EMF if you get the fingers in the way.

klugesmith

Tue May 26 2009, 05:05PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Bjorn wrote ...
For those that measure inductors and transformers with this method, don't let the low voltage give you a false sense of security. If you disconnect the wires with the current still applied you may get a pretty bad shock from the back EMF if you get the fingers in the way.

Good point, to which I can attest from experience. Long ago in high school, made a mild electric shocker out of a little filament transformer, a 1.5-volt AA battery, and a pushbutton switch. Only used one winding, probably the secondary. When you released the button it would produce a jolt that could be felt by everyone joined hands in a circle.

EDIT: just figured you could produce a quite-possibly-lethal shock with a sub-C size NiCd cell and a practical human-body-MRI-size superconducting magnet, say with L = 10 H. Normally this runs with hundreds of amps and stores hundreds of kJ. When you connect the battery, current will build up at 1.2 / 10 = 0.12 amperes per second. Battery voltage will drop under load, but after 100 seconds we'll have at least 10 amps flowing. Now if you pull the wire away from the battery, but still have one hand on each wire from the magnet, you'll get a 500 joule jolt starting at 10 amperes, lasting on the order of 10-100 ms depending on skin contact resistance.

Harry wrote ...
Isn't nichrome resistance wire calibrated to within 5% per metre, or so?

We're straying from the original thread topic, but... nichrome has a relatively low temperature coefficient, where copper is notoriously high at 4% per 10 degrees C. Couple years ago I bought some thermocouple wire just for its constantan, an alloy originally designed for low TCR. The ebay vendor had the wire gage wrong by a couple of AWGs, a detail immediately apparent when I measured the milliohms, then confirmed with a mechanical measurement.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob