Microhm Bridge (Avtron T207)

It's got to be some kind of AC bridge - low resistances generate low voltage drops and thermocouple voltages around the circuit make it impossible to measure anything useful at DC.

This means that there has to be an oscillator somewhere in the box to turn the DC output from the D cells into AC. It will probably run at a fairly low frequency - my guess would be that your transformers are iron-cored, and if it was used for air-craft work they'd use the relatively thin laminations employed in 400Hz aircraft AC.

The oscillator is most likely a Royer inverter on the printed circuit board - a Baxandall class-D oscillator producing a decent sine wave would be better, but scarcely necessary for testing earth bonding - which would incorporate one of those two transformers on the board.

That would drive a lot of turns on the high current transformer - probably through a tolerably stable resistor - to give a much higher current (at a lower voltage) from a second winding to drive through the resistance being measured. The voltage generated across that resistance would drive a relatively small number of turns on the second transformer, to be stepped up to a higher voltage across a winding with many more turns on the same transfomer.

This voltage would then be compared with a voltage derived from a resistive bridge set up the potentioneters behind your big dial, and the difference might then be amplified with a few transistors and most likely demodulated by transistor switches (driven by a voltage picked off the original oscillation) before being fed into you null detector.

You can also use a moving coil meter to detect alternating current, if the magnetic field in which the coil rotates is excited by an alternating current of the same frequency. Readily available moving coil meters use a permanent magnet to set up the magnetic field, so I'd be surprised if your gadget exploited this effect.

I wouldn't expect to see Hall effect sensors in a device this old. The extra connections from the transformers to the printed circuit board are more likely to be a low current high output drive from the board into the transformer and possibly a demodulator drive from the transformer back into the board - there will be some phase shift between the high-voltage drive and the high current output, and it might make sense to derive the demodulator drive from the phase- shifted signal.

-- Bill Sloman, Nijmegen

Thanks for the information. Once I finish work on a couple of major = projects=20 I may come back to do a little more testing on this unit. I don't think = it's=20 worthwhile to gut it for parts or the enclosure, so either I'll get it=20 working as a low resistance meter or I'll just get rid of it, although I =

truly hate doing so. However, I have many other pieces of equipment that = may=20 be more valuable and worth keeping and fixing.

If anyone has any links to similar instruments, and especially = schematics,=20 I'd greatly appreciate it.

For my proposed design of a DLRO, I plan to use a battery and/or a high=20 value low voltage "ultracap" to provide current through the shunt. I = figure=20 on about 2 VDC and a maximum current of 100 amps (which is 20 mOhms). = For a=20 one second TC (which should be enough for a reading), a 50 farad = capacitor=20 would be about right. Otherwise a battery of about 20 A-H rated at 5x = would=20 work.

A 25 F 2.7V Ultracap is only about $5 each, and can provide 16A peak=20 current, so about 6 in parallel would do nicely. For simplicity, I = planned=20 to use resistors to limit the current, and use a ratiometric reading,=20 probably with a PIC. But I might also consider a current regulated = supply.=20 For low duty cycle readings a linear regulator would be feasible at = about=20

There may be some issues with thermocouple voltages using DC, but that = is=20 why many DLROs have a reverse switch and the average should null out = such=20 effects. AC current may have problems with inductance. Not a big problem = for=20 shunts, but sometimes these meters are used to read resistance of=20 transformer coils or other inductive loads, and in fact it is often=20 necessary to wait a while as the current stabilizes, and then "dump" the =

energy into a load before removing the current leads. Otherwise you can = get=20 quite a spark!

Paul

cts

t's

may

re

ld

True. Technically this is still alternating current excitation, but people tend to call it "reversing DC".

et

With "alternating DC" it may still pay to use a sine wave - but a relatively slow one. The practical way of generating very slow sine waves is by direct digital synthesis, and - IIRR - some of the Analog Devices DDS chips can be programmed to generate quite low frequency sine waves. The AD9835

seems to be able to go down to 2.3mHz - a seven minute period - when driven by a 10MHz clock. This isn't the cheapest way of making very low frequency sine waves with a low harmonic content - Don Lancaster's "magic sine waves" are tolerably practical at low frequencies and can give you a very cheap system - but it is pretty painless.

Of course, if you want to separate the inductance and resistance of an inductor, you use an impedance bridge, and measure the phase shift between detected voltage and driving current - which is to say you have separate in-phase and quadrature detectors. The in-phase component is the voltage drop across the resistive part of the impedance and the quadrature component gives you the voltage driop across the reactive part of the component.

-- Bill Sloman, Nijmegen