DMM Continuity Tester V1.0 - Better than clicking probes together!

[IanJ]

Hi all,

Watching various YouTube videos where Multimeter reviewers reach the stage where they test the continuity beep mode of the meter by 'clicking' the probes together as fast as they can to see if they can beat the 'beep' into submission.....I thought "there must be a better way than that".

Not to mention I was moving from Altium CircuitStudio to KiCad V6 and so needed a quick first KiCad project to test the workflow........the DMM Continuity Tester was born.

It's operation is kinda self-explanatory, but there are a couple folks out there I have pushed units out to given I had built up 3 or 4 so I guess YT videos will appear if indeed it's found to be useful.

I've made it open source, it's based around an Arduino Pro Mini (5V) a small Zoltech LCD........all the files are here:
https://www.ianjohnston.com/images/stories/IanJ/DMMContinuityTester/DMM_Continuity_Tester_PCB_and_CODE.zip
(zip scanned with BitDefender)

Update 24/06/23:
I have attached the HEX file for the Arduino......some folks have been having problems compiling the source.
https://www.ianjohnston.com/images/stories/IanJ/DMMContinuityTester/DMMcont.ino.with_bootloader.zip
(zip scanned with BitDefender)

The performance of some DMM's are quite surprising, here's a few:

Example MARK Examples (minimum) at 250mS SPACE:
- Brymen BM257 = 38mS
- Fluke 85 III = 0.9mS
- Fluke 87 = 0.9mS
- Agilent 34405A = 25mS
- Keysight 34461A - 1mS (needs Space = 300mS)
- Agilent 34401A = 20mS, but will work lower but gives continious tone
- Brymen BM786 = Better than 0.01mS
- V&A VA38 = 0.1mS (beep varies in volume to point it's hardly audible though, obviously not using a fixed beep length)

Only one meter (BM786) beat the minimum that can be set with the tester, I guess I should re-write the code using interrupts to try and squeeze a few more uS from it.

Ian.

[tooki]

Very interesting!

Are you able to set the resistance value? Continuity testers vary significantly in their responsiveness at different resistances. (Fluke is excellent not only for being fast, but for being fast at the resistance caused by cheap test leads. Many meters suck with cheap probes, but do well with good probes.)

[IanJ]

Very interesting!

Are you able to set the resistance value? Continuity testers vary significantly in their responsiveness at different resistances. (Fluke is excellent not only for being fast, but for being fast at the resistance caused by cheap test leads. Many meters suck with cheap probes, but do well with good probes.)

It's a rather simple circuit, a p-type FET basically shorts the DMM as it would if you were clicking the probes together. RDS(on) with the FET I happened to use is about 1ohm.

Ian.

[tooki]

Well that’s the thing, when touching probes lightly, they’re often significantly above 1 ohm. Like way more. The situations where really fast continuity is most useful (like when dragging across a row of connector pins to find what’s connected to what) are often those where you’re not ensuring good contact, and thus reacting quickly to, say, 20 ohms is important.


Oh yeah, what would also be really interesting is to be able to test intermittent connections, that is, how does the meter respond to brief gaps in continuity. IIRC, the Flukes not only latch on continuity, but on discontinuity, ensuring an audible gap in the beep.

[IanJ]

Well that’s the thing, when touching probes lightly, they’re often significantly above 1 ohm. Like way more. The situations where really fast continuity is most useful (like when dragging across a row of connector pins to find what’s connected to what) are often those where you’re not ensuring good contact, and thus reacting quickly to, say, 20 ohms is important.

Oh yeah, what would also be really interesting is to be able to test intermittent connections, that is, how does the meter respond to brief gaps in continuity. IIRC, the Flukes not only latch on continuity, but on discontinuity, ensuring an audible gap in the beep.

Well, it's open source so I guess it would be interesting to see anyone else add a few more FET's each with their own series resistor for example, or add more functions to the software like adding gaps in the continuity like you said or even randomizing the output. I guess there's lots that can be done.......just as long as I don't see reviewers clicking DMM probes together by hand and calling a DMM a pass or fail....LOL!

Ian.

[Kleinstein]

One could just add a series resistor to the circuit. So no change needed to the tester.

As long as the contact time is shorter than some 10 ms, I would not even care. So the 0.01 ms steps are well good enough.

[Haenk]

Well that’s the thing, when touching probes lightly, they’re often significantly above 1 ohm

I'm happy with my Probemaster probes, with gold plated tips, they make really good contact, way better than those standard probes.
They make a perfect birthday gift

[thm_w]

Smart.

Do we care about delay as well? Could have a microphone to detect the beep and display delay from activation -> beep.
But maybe its not an issue, and the most important thing is what you've measured here (minimum pulse width).

[inse]

A very specific piece of test equipment test equipment that you have designed there.
The task could also be accomplished with a much more versatile function generator and a MOSFET.
For example my very low cost feeltech generator is able to produce pulses with the required parameters.
My results:
Keithley 2000: 3ms
Fluke 75III: 130ms
PeakTech 3720: 150ms
For other continuity testers its hard to determine whether it is a click or a beep.

[pstimpel]

Ian, is there a special reason for the use of exactly this P-Mosfet? Or asked different: what of its properties are key to success in this circuit / for this purpose?

BR,
Peter

[Kleinstein]

The FET to make the connection should not be critical at all. Not sure why he used the extra p-mosfet. The n-FET MOSFET to drive the gate should do as well.
The ROn shoud be resonable low to be safely detected as a short - so like < 20 ohms or so.

[tooki]

Well that’s the thing, when touching probes lightly, they’re often significantly above 1 ohm

I'm happy with my Probemaster probes, with gold plated tips, they make really good contact, way better than those standard probes.
They make a perfect birthday gift

They’re my favorite probes, too. But that isn’t the point:

1. Any idiot can design a continuity tester that reacts well with Probe Master probes shorted against each other. It takes more skill to design one that works well with more ordinary probes, or ones that are dirty due to field use.

2. In real life, we aren’t testing the continuity of our probes, but of some other connection. And that may be very oxidized.

3. In fact, we might not be using probes at all. We might be using alligator clips, connectors, etc.

4. In the wild, I might not be able to use my Probe Master probes. Like at work, where I’ll be using what they have.

[IanJ]

Ian, is there a special reason for the use of exactly this P-Mosfet? Or asked different: what of its properties are key to success in this circuit / for this purpose?

BR,
Peter

It's just a mosfet i had in stock and use on other products like my PDVS2mini. It's RDS(on) is 1ohm approx.

In this design there is no high Vgs voltages to worry about that might negate some mosfets.
Ps. Even if that was the case an additional 15v zener and a 10k resistor would sort that out anyways.

Ian.

[IanJ]

Next version if I get the time.....

- Add a microphone to measure response time of DMM's.
- Increase the speed so the likes of the BM786 or other fast DMM's can be measured (possibly use 16bit PWM or perhaps an STM32).
- Add some arbitary type modes, i.e. mimic running along 10pins of a connector.
- Add more output FETs with various different series resistances to mimic crap probes versus quality etc.
- Anything else!

Ian.

[NiHaoMike]

Might it be possible to implement variable resistance by operating the MOSFET as a linear device rather than a switch? Might not be the most accurate (unless you add a feature for self calibration?) but I'm not sure if there's much value in accuracy for this application.

[IanJ]

Might it be possible to implement variable resistance by operating the MOSFET as a linear device rather than a switch? Might not be the most accurate (unless you add a feature for self calibration?) but I'm not sure if there's much value in accuracy for this application.

Might be possible but I guess a bit of a pain to implement reliably or without getting more complex. Doesn't help matters that the voltage across DMM probes when open vary quite a bit across different models/brands in my testing.

Ian.

[Mike G]

Nice idea Ian, tapping the probes together never struck me as a definitive way of judging the response of a DMM
I have now got it running on a Grove Beginner Kit with oled display, must admit a nice bit of coding
I noticed a timing glitch in the spaces but disabling the on board led flashing appears to have eliminated this, otherwise quite usefully accurate.
Another option is to use the BB3 and a couple of FETs, gives an equally accurate output.
Mike G
P.S. looking forward to Pt3 of the Keithley repair

[Bud]

Well that’s the thing, when touching probes lightly, they’re often significantly above 1 ohm

I'm happy with my Probemaster probes, with gold plated tips, they make really good contact, way better than those standard probes.
They make a perfect birthday gift

Sorry, what is the point to have gold plated tips? A few hard presses on oxidized solder joints and gold will be gone from the pointy tip surface.

[tunk]

A short youtube test by DefPom: https://youtu.be/CgyCS5RizwE?t=240

[IanJ]

Nice idea Ian, tapping the probes together never struck me as a definitive way of judging the response of a DMM
I have now got it running on a Grove Beginner Kit with oled display, must admit a nice bit of coding
I noticed a timing glitch in the spaces but disabling the on board led flashing appears to have eliminated this, otherwise quite usefully accurate.
Another option is to use the BB3 and a couple of FETs, gives an equally accurate output.
Mike G
P.S. looking forward to Pt3 of the Keithley repair

Nice catch on the on-board LED. I'll try that myself.
Implementing PWM etc would of course sort out any timing issues allowing for other stuff to be done, but heck it was only V1.0.
Ian.

[tooki]

Sorry, what is the point to have gold plated tips?

That they don’t oxidize, the same reason we use gold on connectors, card edges, etc.

A few hard presses on oxidized solder joints and gold will be gone from the pointy tip surface.

Definitely not true. My Probe Masters are at least 5 years old now and have probed many a thing, including oxidized solder joints, and the gold still looks like new. Whatever their gold alloy is, it’s hard enough to last.

[Bud]

Still, none of multimeter brands make these. Apparently for a reason or two. Probe Master is not an industrial manufacturer.

[tooki]

Brymen does.

What, you think Probe Master is in someone’s garage or something? It may be a small company, but it’s been around for years and years and is clearly legitimate, and definitely makes things for serious use. They have probe extenders (CAT rated IIRC) and a lot of their older stuff was clearly focused on the telecom market, which easily qualifies as light industrial.

My assumption is that cost is the primary reason gold plating isn’t universal. I’m sure it’s less durable than chrome plating, but it’s certainly not as dinky as you claimed. (Ever seen a gold card-edge connector! Yeah, you can plug those in more than once without ruining the plating.)

[Bud]

What, you think Probe Master is in someone’s garage or something? It may be a small company, but it’s been around for years and years and is clearly legitimate, and definitely makes things for serious use.

That is exactly what i think. I recall Probe Master is a one man band. And as such i believe the guy does not make anything, just a smart person having access to internal production market in China and picking quality products for resale.

[Kleinstein]

Chrom plating of probe tips is more like a stupid idea. Chrom tend to build up a hard protective oxide layer quite fast.

If they coat just the tip (e.g. 1mm) a gold coating would not even be that expensive.