A low resistance ohm-meter

[Skywave]

Overall circuit design is from other peoples' ideas found on the Web, text books and my own thoughts. So no claims for total design originality by myself are made here.

Basic concept is a constant-current generator that sends a known test current through the resistance to be measured. The resultant voltage across that resistance is then fed to a high input resistance amplifier, whose output feeds a moving coil meter. By arranging for a range of constant currents to be selected (which are derived from a stabilized voltage source), a number of resistance ranges are provided. Hence, this is a 'four-wire' resistance-measuring ohm-meter. It uses two 9v. batteries: one for the current source; one for the voltmeter cct. - which is based on a 741-type op-amp. The unit features a switch-selectable monitor of each batteries voltage.
Overall, the cct. is really quite basic - but the unit works as intended.

All the parts came from bits available in-house, including the classic 'junk box', except for the plastic case and the paint for that case. Overall construction is not the neatest I could have done, but then I'm not presenting this as a potential prize winner in a constructors' competition. Just a relatively cheap-to-make piece of useful test gear.

I used a few 'known-to-be highly accurate' resistors for calibration.

Photos as below.

Al.

A good bit of workshop kit, I like it.

[kalee20]

A useful item!

Comments... How low is low resistance? Generally, a high input resistance voltmeter is not needed because the voltage developed across the unknown resistance has got plenty of grunt behind it, so if you are measuring around an ohm, and your voltmeter is as crummy as to load the input with a resistance as low as 100 ohms, it'll still give an error of only 1%.

If you are measuring resistances down to milliohmd, it may be useful to turn the constant current on and off. With current off, you may still get a voltage due to thermocouple effects. Turn on the current, and the change in voltage (not the absolute voltage) is what you want.

[Skywave]

Originally Posted by kalee20

How 'low' is low resistance?

On the lowest R range, 0.05 Ω has been verified as 'reasonably accurate' for the majority of purposes that this item is likely to be used for. That R range will indicate resistances lower than that, but even with a four-wire measurement technique, I wouldn't put too much reliance on readings below 0.05 Ω. Besides, knowing what I am likely to use this item for, I'll be quite surprised in any R below 0.05 Ω (or even that figure) will ever be of interest to be.

Originally Posted by kalee20

. . . and [if] your voltmeter is as crummy as to load the input with a resistance as low as 100 ohms . . .

The meter itself requires 1 mA for FSD and has an internal R of 100 Ω

Al.

[Skywave]

Simply on the off-chance that someone might like to build it, here is the circuit diagram and the set-up and calibration instructions. The quality of the print is not good, I know, but all the essential info. is there.

Calibration.
Constant-current generator. I used an accurately-calibrated milli-ammeter to set up the currents from the constant-current generator. The variable resistors are 10-turn pre-sets
Voltmeter. First, set the mechanical zero of meter. Then join the +Vin and -Vin leads. Adjust the off-set pre-set (RV8) for zero on the meter. (It's adjustable either side of meter zero).
Select FSD range of 100 Ω. An accurately-known 100 Ω resistor is used to set the gain of the op-amp (RV7). It is also worthwhile then to check the Ω FSDs for the other ranges by using appropriate accurate values of resistors - assuming, of course, that they are available: for me, they were.

Possible enhancements & developments
►The fitment of a low-frequency oscillator driving a red LED from one of the power rails might be worthwhile to act as a 'it's switched-on' reminder, since it is easy to forget to switch this battery-powered item off after use.
►As for power sources, there is no absolute reason why batteries should be used. A mains-derived PSU could have easily been used instead. I just wanted something that was 'mains-lead free'.

Al.