Author Topic: Gratuitous Continuity Speed Test and Comparisons  (Read 1221 times)

0 Members and 1 Guest are viewing this topic.

Online KungFuJoshTopic starter

  • Super Contributor
  • ***
  • Posts: 2678
  • Country: us
  • TEAS is real.
Gratuitous Continuity Speed Test and Comparisons
« on: August 11, 2024, 10:53:45 pm »
We were (in classic EEVblog fashion) derailing another thread talking about continuity speed comparisons.

I'd like to consider this further, and have a standardized method to see what parameters to use while testing continuity speed of DMMs with an AWG. Some AWG screenshots would certainly be helpful too.

It was also stated that we should have a second audio-based method to determine the time between detection and an audible signal. This seems simple enough for those of us with DAW or sound editing software. Other ideas?

Discuss. 😉
"Right now I’m having amnesia and déjà vu at the same time. I think I’ve forgotten this before." - Steven Wright
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 17184
  • Country: 00
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #1 on: August 12, 2024, 12:02:59 am »
Use a relay to make contact. Use a microphone to record the click of the relay and the beep. Measure the time between the two.
 
The following users thanked this post: pdenisowski, KungFuJosh

Online KungFuJoshTopic starter

  • Super Contributor
  • ***
  • Posts: 2678
  • Country: us
  • TEAS is real.
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #2 on: August 12, 2024, 01:30:03 am »
I want to get deeper into the methods people are using with an AWG. Some of the claims I've seen are questionable.
"Right now I’m having amnesia and déjà vu at the same time. I think I’ve forgotten this before." - Steven Wright
 

Offline BeBuLamar

  • Super Contributor
  • ***
  • Posts: 1360
  • Country: us
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #3 on: August 12, 2024, 01:43:13 am »
I rarely use the contnuity test as I generally need to know the resistance of the continuity but I would do the test.
 

Online KungFuJoshTopic starter

  • Super Contributor
  • ***
  • Posts: 2678
  • Country: us
  • TEAS is real.
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #4 on: August 12, 2024, 02:01:36 am »
I rarely use the contnuity test as I generally need to know the resistance of the continuity but I would do the test.

My Siglent bench meters show the resistance of the continuity on screen, and you can set the threshold (resistance level) of what the meter considers continuity.

That setting seems to have an effect on the speed of the continuity response as well (at least it does when screwing around with the AWG), which is one of the reasons I'm interested in testing this method further.
"Right now I’m having amnesia and déjà vu at the same time. I think I’ve forgotten this before." - Steven Wright
 

Offline J-R

  • Super Contributor
  • ***
  • Posts: 1215
  • Country: us
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #5 on: August 12, 2024, 02:03:35 am »
Not sure what thread that was, but was this already discussed?
https://www.eevblog.com/forum/projects/dmm-continuity-tester-v1-0-better-than-clicking-probes-together!/
Fixed URL
« Last Edit: August 12, 2024, 03:08:15 am by J-R »
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 17184
  • Country: 00
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #6 on: August 12, 2024, 04:33:51 am »
My Siglent bench meters show the resistance of the continuity on screen

Most of my handhelds do that as well.
 
The following users thanked this post: KungFuJosh

Online Fungus

  • Super Contributor
  • ***
  • Posts: 17184
  • Country: 00
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #7 on: August 12, 2024, 04:36:34 am »
joesmith did some fancy testing.

Look at the spreadsheet here: https://www.eevblog.com/forum/testgear/hear-kitty-kitty-kitty-nope-not-that-kind-of-cat/

There's a tab for "continuity tests" and some matching videos if you look for them on his channel.

 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 27947
  • Country: nl
    • NCT Developments
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #8 on: August 12, 2024, 08:37:42 am »
I wouldn't use a relay as this has a delay in 10's of milliseconds. And there will be contact bounce. Better use a clean signal from a function generator.
« Last Edit: August 12, 2024, 08:39:16 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 
The following users thanked this post: egonotto, KungFuJosh

Online 2N3055

  • Super Contributor
  • ***
  • Posts: 7239
  • Country: hr
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #9 on: August 12, 2024, 09:03:21 am »
I wouldn't use a relay as this has a delay in 10's of milliseconds. And there will be contact bounce. Better use a clean signal from a function generator.

Best is to use signal generator to drive gate of N ch mosfet and use D-S to trigger  meter.
Even 2N7000 will do....
That is so easy to do: you can solder MOSFET directly to BNC connector.
 
The following users thanked this post: egonotto

Offline Antonio90

  • Frequent Contributor
  • **
  • Posts: 376
  • Country: es
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #10 on: August 12, 2024, 11:09:02 am »
I want to get deeper into the methods people are using with an AWG. Some of the claims I've seen are questionable.

I can have a look later for a screenshot. You really just need to setup a pulse with a positive baseline, so that the meter detects voltage (drop) on the terminals, and assumes high resistance. Then the pulse has to drop to around zero volts, often slightly negative, which the meter reads as very low voltage drop, and assumes very low resistance.

The pulse can most likely be set up with any decent AWG, but it is faster to set a regular, positive pulse with negative offset and reverse the leads. That's What I did with all DMMs except for the UT181A, which worked without reversing the leads.

In any case, this doesn't take into account the source characteristics of the DMM, which might make a difference. The proper way to test it is with a MOSFET, as was already stated.

For the delay measurement one would probably need a microphone and a buzzer to check for the inherent delay of the measurement, and then trigger the scope on the AWG pulse while recording the output of the microphone, calculating the interval and substracting the inherent delay.
« Last Edit: August 12, 2024, 12:31:12 pm by Antonio90 »
 

Offline Njk

  • Frequent Contributor
  • **
  • Posts: 302
  • Country: ru
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #11 on: August 12, 2024, 03:22:06 pm »
The fundamental problem is that a standardized method assumes a standard setup. It's a project that will require significant investment of time and effort, to say the least. Will never materialize, I'm afraid.

Continuity test is usually about a resistance measurement. It was suggested to use a generator. The generator generates a voltage pulse, not a resistance pulse. But it's a minor problem, it's easy to convert a voltage pulse to a resistance that the DMM can sense, with FET or something. Usually, input of such a device is quite delicate so a reasonable input protection circuit will be required (a bit more additional components).

In DMMs, the result of the continuity test is indicated by sound so the only non-intrusive way to convert it to a machine-readable form is to use a microphone. Therefore, a microphone amplifier will be required. I think it'll be better to stay away from any sort of digital audio (PC sound cards, etc.) because digital processing is associated with some latency and annoying calibration procedure must be followed to take the latency time into account. A simple analog amplifier seems a better solution (add a few more components).

Anyway the sound is generated by a beeper, internal location and characteristics of which is up to the DMM's manufacturer. That means that for each particular DMM, operator will have to identify optimal spatial location of the microphone relative to the DUT. The chances are in the optimal position, the microphone will be sensitive to a not related environmental sounds which can compromise the measurement accuracy and make the results debatable. That leads to the idea of a standardized fixture, e.g. in the form of wooden cabinet acoustically isolated from the environment. The microphone (or an array of them) are placed inside at a fixed position. The cabinet is of enough internal volume to accommodate a handheld or a bench-type DMM (if we would like to test the latter type as well).

The problem with a stationary DMMs is that many of them are fan-cooled. A high pass filter in the mic amp will be required to filter out a hum from the rotating impeller (even more additional components).

From the electrical perspective, there is nothing complex. A mic(s) followed by an amp followed by a rectifier followed by a comparator that produces a sharp edge at the moment of sound attack, so the time position can be easily measured. BTW the comparator output can be routed back to the input of the voltage-resistance converter device (perhaps through some delay circuit) to create a positive feedback oscillation that eliminates the need for external generator.

It can be more complex to define the test criteria. The DMMs can be very peculiar. The beep first starts for several hundreds of ms, then interrupts for tens of ms, and only then starts sounding continuously. No idea how typical it is but I've one evidence. Another DMMs, like my Fluke 189 generates not a tone, but a sound of quite weird cadence. I think it's a feature. In a really noisy environments, like a server room, that irregular cadence is easy to notice while a musically perfect tone would be a disadvantage. It will be nice if the test setup can differentiate between a well-tempered DMM and a poor one.


 

Offline Demon Xanth

  • Contributor
  • Posts: 27
  • Country: us
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #12 on: August 12, 2024, 05:05:28 pm »
The method I would use:
Set up a DPDT or DPST relay with one pair of contacts going to a set of banana jacks, the other going to an LED.
Monitor with a high enough frame rate camera as you energize the relay.
Review the video to look at the difference in frames between the meter's response compared to the LED lighting up.

The relay's speed would be largely irrelevant since both sets of contacts would have (for all practical purposes) the same delay.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 17184
  • Country: 00
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #13 on: August 12, 2024, 06:19:23 pm »
I wouldn't use a relay as this has a delay in 10's of milliseconds.

Not from the sound of the click.

The problem is knowing when the meter responds, hence using something audible and looking at the audio recording.
 
The following users thanked this post: KungFuJosh

Offline BeBuLamar

  • Super Contributor
  • ***
  • Posts: 1360
  • Country: us
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #14 on: August 12, 2024, 06:36:16 pm »
I wouldn't use a relay as this has a delay in 10's of milliseconds.

Not from the sound of the click.

The problem is knowing when the meter responds, hence using something audible and looking at the audio recording.

You can hook a scope to the 2 test leads and see the voltage accross it drops when the relay closes. On the second channel you feed the signal from the microphone for the beep.
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11871
  • Country: us
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #15 on: August 12, 2024, 06:59:51 pm »
I measure the resistance required for the meter to measure an open and a short.  I measure the short circuit current and open circuit voltage.   For the speed test, I use a 50% duty cycle and increase the frequency until the meter no longer detects a short.  I then use a fixed 2Hz and reduce the on time until the meter can't detect it.   All of that data is recorded and available on-line.

To apply a short for the frequency test, I have a driver that controls three FETs, allowing me to run up to three meters.  The reason for this is some have asked about the meters response time.  With this, I can use a 1000fps camera to watch their LCDs.   

I try to measure when I actually hear a full beep.  Some meters that do not have a latching input can respond so fast, you may hear something but it may not be a beep.   For example, shown is the UNI-T UT90A.  This meter is non-latching and has a very fast response.   This allowed me to modulate a one-shot that has an on-time short enough not to cause the meter to beep but long enough to make it click.   What you are hearing is a series of clicks.   Sorry but I did not take any photos of that circuit (built on a breadboard) nor did I document it.  It was more just for fun rather than used as a metric for evaluating meters.   

***
Just some additional info.

Corrected error above, only 3 meters can be checked at a time.  The 9V and forth FET provide the level shift and drive.   
The resistance is measured using a resistor substitution box.  Mine has a 1 ohm resolution and finding the sweet spot is fairly quick. 
The current and voltage are checked with an HP34401A. 
Normally I am using my Tektronix AWG2401 arb to derive the pulse.   This runs from LabView.
FETs are some very old Siemens BUZ 72 N-channel,   most likely from the 80s.  Not critical and what was in my junk drawer.  Because the meters are all isolated, they self bias.  So no big deal driving them this way. 

Datasheet
https://docs.rs-online.com/e916/0900766b80026b66.pdf

***
A bit off topic.  Dave testing the setting time of a few meters.   This is why I support running more than one meter with my setup.     
https://www.eevblog.com/forum/testgear/hear-kitty-kitty-kitty-nope-not-that-kind-of-cat/msg2786500/#msg2786500

***
Repeat from the last time it came up:
https://www.eevblog.com/forum/testgear/multimeter-with-continuity-test-buzzer-that-responds-immediately/msg3168420/#msg3168420


« Last Edit: August 13, 2024, 10:25:51 pm by joeqsmith »
 
The following users thanked this post: jjoonathan, 2N3055, KungFuJosh, Antonio90

Online KungFuJoshTopic starter

  • Super Contributor
  • ***
  • Posts: 2678
  • Country: us
  • TEAS is real.
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #16 on: August 13, 2024, 12:08:59 am »
Joe, thanks for your reply and info. It seems to me that what you've done to measure the resistance required to set off the continuity is an important factor.

Would you mind drawing a diagram of your test setup? I'd like to setup something standard for more consistent testing.

I'm also curious about how flexible thresholds effect continuity speeds. I don't remember, do you have any Siglent bench DMMs? Or another brand where that can be manually set?

On my Siglent bench DMMs, it seems to me to be significantly faster with a higher threshold set. For example, when the threshold was set to 50Ω, it appeared to be much slower compared to setting the threshold to 500Ω or 1kΩ.

Thanks,
Josh
"Right now I’m having amnesia and déjà vu at the same time. I think I’ve forgotten this before." - Steven Wright
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 17184
  • Country: 00
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #17 on: August 13, 2024, 12:13:05 am »
Joe, thanks for your reply and info. It seems to me that what you've done to measure the resistance required to set off the continuity is an important factor.

Would you mind drawing a diagram of your test setup?

A potentiometer...?
 

Online KungFuJoshTopic starter

  • Super Contributor
  • ***
  • Posts: 2678
  • Country: us
  • TEAS is real.
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #18 on: August 13, 2024, 12:22:36 am »
A potentiometer...?

 :-DD :-DD :-DD

I meant the whole speed test rig, not how he determined the needed resistance. 🤣
"Right now I’m having amnesia and déjà vu at the same time. I think I’ve forgotten this before." - Steven Wright
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11871
  • Country: us
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #19 on: August 13, 2024, 01:05:57 am »
Would you mind drawing a diagram of your test setup? I'd like to setup something standard for more consistent testing.
See attached.   I suggest using no more than a 5 ohm output resistance just based on the meters I have looked at.   The other values are not critical.

I'm also curious about how flexible thresholds effect continuity speeds. I don't remember, do you have any Siglent bench DMMs? Or another brand where that can be manually set?
I don't own any Siglent products.  The Gossen Ultra or Prime as they renamed it, supports this. 
 
On my Siglent bench DMMs, it seems to me to be significantly faster with a higher threshold set. For example, when the threshold was set to 50Ω, it appeared to be much slower compared to setting the threshold to 500Ω or 1kΩ.
Personally, I just want it fast enough to detect when sliding the leads over a large QFP.  Anymore though, that is less of a concern as we mostly work with BGAs and I no longer do much reverse engineering.

Offline MathWizard

  • Super Contributor
  • ***
  • Posts: 1616
  • Country: ca
Re: Gratuitous Continuity Speed Test and Comparisons
« Reply #20 on: August 14, 2024, 11:50:50 pm »
I've designed a few continuity testers for my old Keilthy DMM's, and they didn't go so good. And I learned a few more tricks, and built the latest version, and hooked it up to the DMM. But so far I haven't tested it in circuit, because I can't find the cables that fit this meter. Oh well time to just use alligator clips.

One thing I wish I had, if they make them, is a fine metal brush, to buzz out connections on double side boards, and just big complex boards.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf