Wall wart instead of dual batteries.

Thread Starter

zophas

Joined Jul 16, 2021
165
I am hoping to try out the following circuit. Mainly because I'd like to add a milliohm meter to my equipment and because I already have the Digital Panel Meter. A kit I bought and built years ago based on the ICL7107 with 4 seven segment LED readout. The designer of the below circuit does not explain why "separate" batteries should be used. I would like to power the whole thing with just one 9v DC wall wart. Can this be done? Thanks.
 

Attachments

Ramussons

Joined May 3, 2013
1,404
No, you will need 2 wall marts.
If you can get the schematic of the Panel Meter, we can analyse and see if we can manage with 1 battery.
 

Ramussons

Joined May 3, 2013
1,404
Using 2 wall warts is even worse than using two batteries. The Panel Meter schematic below uses an ICL7107 chip.
That chip has no common link between the sense inputs and the power connections. It requires a Floating Power supply, hence the independent battery.
 

Ian0

Joined Aug 7, 2020
9,668
The 7107 has differential inputs, so you’re probably on to a winner. I suspect that when it was used In the digital panel meter one of the differential inputs was commoned with ground. Removing the link between negative input and ground may be all that is necessary.
 

Thread Starter

zophas

Joined Jul 16, 2021
165
The 7107 has differential inputs, so you’re probably on to a winner. I suspect that when it was used In the digital panel meter one of the differential inputs was commoned with ground. Removing the link between negative input and ground may be all that is necessary.
It was a kit I put together years ago thinking I would try to put it into my home made bench PSU but I never got around to that. Now I'm hoping to use it for a milliohm meter. I have about 10 different designs for milliohm meters and most of them I use my DMM on the mV scale, which is fine. This design appealed to me because I can finally make use of the Panel Meter. I was just hoping to avoid the cost of PP3 batteries which is quite high where I live.
 

AnalogKid

Joined Aug 1, 2013
10,986
We don't know how much you know about what you are attempting. Here is a short course on the Kelvin connection.

https://en.wikipedia.org/wiki/Four-terminal_sensing

You can cheat this by combining the low side excitation connection with the low side measurement connection. This eliminates the common ground problem that requires dual, isolated power sources, but makes the measurement accuracy dependent on how close the low-side connection is to the body of the resistor being measured. This always is a problem for the measurement connection, but now you have to get the excitation wire (which can be much fatter in some applications) jammed in there.

If you want to try this, then bring the two low-side wires together as close as possible to the probe tip. Every millimeter that the two currents share a conduction path degrades the measurement. You can try both power schemes and chart the differences in readings to get a feel for how much of a problem this actually is with your setup.

ak
 

Thread Starter

zophas

Joined Jul 16, 2021
165
We don't know how much you know about what you are attempting. Here is a short course on the Kelvin connection.

https://en.wikipedia.org/wiki/Four-terminal_sensing
Thanks, @AnalogKid . I'm not really attempting to do anything serious. I just would like a milliohm meter on my bench. I ran into this the other day when I realized that my cheap DMM could not read very low resistances. Always good to have something that can. Yes, I have looked at a lot of info thanks to the internet about low resistance measuring. I am trying to keep this as simple as possible and make use of my old Panel meter kit too.
 

crutschow

Joined Mar 14, 2008
34,283
Here's my take on using one supply (LTspice simulation below).
I added a Zener to float the resistor UUT voltage to about 2.7V above ground which can then be readily measured by the 7107 differential inputs.

You will need a four-wire connection, of course, to minimize the effect of connection and wire resistance.

1628621864480.png
 

Ian0

Joined Aug 7, 2020
9,668
Exactly what I was thinking of, but you posted it first.
I'd probably add a couple of back-to-back schottky diodes across the probes, because I seem to have developed a habit of looking for shorts on a pcb without remembering to turn off the power supply.
D1 could be replaced by a 25Ω resistor.
 

Ian0

Joined Aug 7, 2020
9,668
Agree. How did I miss that all these years? I have a large, adjustable power resistor (the kind with a linear adjustable slider), but nothing nearly this low a value. Duh.

ak
Brass for the lower resistance ones, stainless steel for higher values.
both have lower temperature coefficient of resistance than copper.
I can’t imagine that anyone makes constantan threaded rods.
[edit] or maybe people who make clock pendulums do.
 
Top