BM789 continuity behaviour

[Electro Fan]

No issue with 786.

I wonder if it is firmware version related. 

Not that it matters since Bremen will not provide end user firmware upgrade. But at least it may help for those who are shopping for a used 789 if it is firmware related.

78905 versions should command a higher price in the resale market for users seeking occasionally extra high sensitivity.

[mwb1100]

BM786 firmware 78608 doesn't show this issue.

[sonpul]

Or it does, and you just learned to accept it?

I usually focus, start work and finish. The call is used for troubleshooting, partial reverse engineering, checking your work, etc. It doesn't seem to interfere at all. Otherwise, I would not have bought a second exactly the same.
On my devices 78911 firmware. I can observe such a feature on them. As a rule, if I pay attention to this, then the very first touch to the board of one of the probes, for the whole day or the ringing cycle, may be.
Or the devices do their job immediately without a short beep.

[Veteran68]

OK, this thread has me testing more.  Mine does the same as Caliaxy describes:

random very short beeps when touching metal parts of various equipment with only one probe (the other probe in the air). Looks like some static charge pick-up - touching repeatedly the same metal part does not result in multiple beeps; you have to wait a bit or touch something else before being rewarded with a new beep.

I can reproduce this ^ .

I can reproduce this as well with my eevBlog BM786 (FW 78607). I just touched the red probe to a metal shelving unit while holding the black probe in the air, and it beeped once. Touched it again and nothing. Then touched it to a heatsink on a PCB on my bench maybe 2 seconds later, and it beeped once again.

Pretty sure I've seen this before with other meters besides the Brymen. In fact I watch a LOT of YouTube repair videos and am sure I've seen meters beep once like that on first contact, and they weren't Brymen meters either. However I just tried with a couple of my other meters at hand -- a Fluke 117 and a Brymen BM869s -- and couldn't reproduce it. I have a lot of other meters that I may try here and there.

I hadn't really thought of it as a problem and it's never caused me any concerns. I chalked it up as some kind of capacitance behavior or something but never noticed before that it would also happen with one probe in the air.

EDIT: Added firmware version

[LuisBe]

Hello everybody again, sorry for the delay but my holidays are over and I can't play with multimeters for so long.
I would like to say that I really like the BM789, its size, its construction, its functions and simply in this thread we talk about small details and let the community know about them without bad faith.

About the short beep when touching the chassis of my equipment, I have seen that my installation must not be right and some of my equipment must introduce some kind of noise in my ground and I think that makes my multimeter behave like that.

https://youtube.com/shorts/v3YMqNoi91U?feature=shared


About the short continuity beep on a Chinese Arduino board, it strikes me that my BM235 doesn't behave like that. Maybe it is because the BM789 is much more sensitive so the continuity function is better on bigger boards than a small Arduino board.
My firmware version is 78912, you will see in the video after displaying it, it turns off (I don't know if it's normal operation but I think it's not a problem).


https://youtube.com/shorts/BC1WUbZbf5E?feature=shared


Greetings to all. LuisB

[tomwilkinson]

My BM789 Firmware 78910 exhibits this problem intermittently.

[giosif]

Thanks to all of you who tested and reported here!

For reference, my unit is running FW 78911.
So, I don't believe the issue is firmware related.
Could they be different HW revisions maybe?

[s_n]

I don’t have a 789, but I’ve seen similar behaviour on other meters related to charging capacitors. I can reliably reproduce it by shorting a capacitor to discharge it, then measuring continuity across the capacitor. Measuring repeatedly will not trigger continuity again until the capacitor is discharged. I would assume that this behaviour is related to the continuity open circuit voltage and the speed of the continuity mode.

[LuisBe]

About the short continuity beep on a Chinese Arduino board, it strikes me that my BM235 doesn't behave like that. Maybe it is because the BM789 is much more sensitive so the continuity function is better on bigger boards than a small Arduino board.
My firmware version is 78912, you will see in the video after displaying it, it turns off (I don't know if it's normal operation but I think it's not a problem).


https://youtube.com/shorts/BC1WUbZbf5E?feature=shared


Greetings to all. LuisB

Here is what Brymen told me about the behaviour of BM235 and BM789:

"BM235 or BM789 Continuity function outputs around 1.7V and 2.3V respectively to test. [...] BM235 continuity response time is around 15ms and can not catch its SHORT CIRCUIT moment. BM789 continuity response time is very fast and is less than 100μs. It can catch its SHORT CIRCUIT moment and truly reacts what it measures.
BM789 is a professional meter. Its Continuity design is with much faster response than most of other meters to target for the professional applications of needing very short-time Continuity tests. If this user would like to have slower Continuity response, maybe he could try to use BM789 Diode function to test"

[Electro Fan]

- Just as I was about to hit upload on this post, I saw LuisBe’s post – thanks for following up with Brymen, LuisBe.  We can certainly add Brymen’s response to the mix – but per some thoughts below it might be interesting to confirm via further testing what is happening with respect to the “one probe continuity beep”.

----

FWIW, I think it’s too early to draw a firm conclusion about the test results being reported in this thread.

Perhaps there is unit to unit variation caused by hardware or firmware among 789 units and/or other Bymen models and/or other non-Brymen DMM models.  (And maybe what we are seeing is partially explained by differences in designed-in model to model response times.)

But another theory (hypothesis?) is that what we are seeing in terms of differing test results from users is not necessarily unit to unit variation or model to model variation among the test equipment but something more like “DUT to DUT” variation in the sense that each “DUT” is really not just a different “Device” but a different “Environment.”

To rewind, what users are reporting is that the 789 sometimes emits a beep when only one probe physically touches something – sometimes it’s a conductor in a circuit and sometimes it’s just a finger touch from a user.  The interesting thing is that these seemingly random beeps happen with just one observed physical touch to one of the probes rather than both probes.  Normally we expect continuity to be signaled to the user with a beep when a circuit is confirmed to have some current detected between the probes.

Next, per s_n’s comment (s_n, welcome to the EEV forums btw), it does appear that the Brymen DMMs are behaving in a way that is generally associated with capacitors.  So then the question becomes:  what capacitor(s)?  An external “DUT” capacitor or maybe an external “DUT environment”, ie a capacitive environment, or an internal capacitor or capacitive circuit within the DMM?  Or possibly some combination of capacitance change is being detected by the probe(s) along with some capacitive behavior occurring within the DMM?

At this point, the most interesting thing, IMO, is not even whether the change in capacitance is occurring due to an external condition (in the DUT or in the testing environment), or internally within the DMM but the fact that the change occurs when only one probe is physically touching something (at least in the sense of a normal DMM continuity test with 2 probes where a user applies both probes to a potential circuit). 

This is just a guess but it seems that the amount of charge we are seeing build up and dissipate is relatively small.  Perhaps what we are seeing is test equipment (a DMM with two probes) that has a sensitivity that is on the threshold of detecting and reporting (via the “random beeps") some change that is lower than the change that normally occurs with a physical touch to both probes; instead what we are seeing is a change that is occurring with some amount of electrostatic charge such that when one probe is physically touched the physical touch is just enough to create a circuit when in fact the other probe is also participating in the circuit via an electrostatic charge or discharge.  This would indicate some relatively small amount of capacitor-like activity is occurring somewhere.

So, now we are back to what small change is occurring?  It seems it’s a bit too early to say what is changing but it also seems that there should be some way to measure and observe the change.  In the same way that we can charge and discharge a capacitor and see the results on an oscilloscope it seems that we should be able to observe and measure what is happening with the DMM “random beeps.”  It might require more of a metrology level test setup in terms of both more sensitive (more resolved and more accurate) test equipment than we normally associate with DMM continuity tests but it doesn’t seem that such test equipment and testing technique/know-how is beyond the resources and know-how of at least some EEV members.  In my limited experience I have come to appreciate for example how relatively subtle changes in temperature can change resistance measurements in a simple resistive test.  Such changes are not noticeable (or generally even of interest) when measuring DUT resistance around 1 ohm but as resistance measurements reach levels on the order of 10 milliohms or less temperature changes of a few degrees C result in different resistance measurements.  My guess is that something similarly subtle is happening in users’ test environments and therefore it is perhaps different environments that Brymen 789 users are detecting and reporting. 

Or maybe it really is differences in DMM’s (ie, unit to unit variation) that we are seeing – but it seems that we should be able to devise a test with enough resolution and accuracy that we can explain what is happening.  This last point is sort of important for readers here as what this thread is about, I think, is not complaining but rather explaining (and learning). 

Net, net:  My guess is that the “random” nature of the beeps should be explainable through better testing.  We might be literally in the “noise” and this might not be a good use of some people’s time and energy but for anyone who wants to look further I think this might be a fun and educational project - to see if we can more definitively test and observe what is causing a seemingly one probe continuity beep (but is in fact a normal two probe beep).  I’m guessing it’s something low level but something that can be tested, observed, measured and explained with better test equipment and better test technique/know-how.

[giosif]

@Electro Fan: That was a (long, but) interesting read. :-)
And I agree that, without more in-depth investigation, it is hard to tell what the main contributing factor(s) to this behaviour is/are.

This said, mine is more of a layman's perspective on the matter: in practice, the continuity mode on the BM789 is less useful than that on most decently-fast meters (e.g. 121GW, BM257s, Fluke 87V, Fluke 187/189, etc.).
And just to prove what I mean (and what actually compelled me to start this thread in the first place), I made a recording of me using the BM789 in continuity mode on a given board (the video also shows me using the BM257s, with no spurious beeps present):
https://youtu.be/ZibMm3JXq7I
I suggest paying particular attention to the 0:30 mark in the video, when I am probing a denser area on the board.
As you can see, the amount of spurious beeps in quick succession makes one unable to determine if there is continuity or not, unless one keeps the probe in a location for longer.

Also, based on the response I received from Brymen (basically, the same as LuisBe's), I am to believe this is intentional for "professional applications" where sub-100μs response time for continuity is needed.
I wonder what real-life usage scenarios for a handheld multimeter would require such crazy-fast response time (as opposed to, say, the 15ms for the BM235).
Is a human even able to tell the difference between the two response times in any given scenario?
Also, if the spurious beeps are a consequence of this fast response time, don't they actually negate the value of this feature?

In my latest response to Brymen, I queried if it may be possible to implement (through a FW update) an option for the user to switch between "crazy fast" and "decent fast" response times.
I will report any response I get.

[joeqsmith]

I tried with a 150pF then a 300pF and could not get mine to trigger.    From my spreadsheet, the short circuit detection resistance was as high as 270 ohms.   The BM235 measured 152 ohms.  The BM839, 110 ohms.   869s, 94 ohms.   From the meters I have looked at,  in general Brymen seems to set this resistance higher than other brands.  I suspect this is why you are seeing the blips.  Agree that this would slow you down when tracing out a board. 

[tautech]

They can't all play Van Halen.   

Joe you have a standardized continuity test using an AWG to sense both min pulse width and max frequency which I have used too testing bench meters. 

[joeqsmith]

I just didn't see the whole touching the leads thing as being very useful.  Now if we could get some sort of standards for low frequency magnetic interference, low energy transient, chemical compatibility and switch life,  maybe we would have something.   

[giosif]

Hi,

Just to provide an update that Brymen did eventually get back to me with an update on changing the response time of the BM789 for continuity mode.
And the change is a hardware modification (which Brymen called a "hack") - using the attached picture as reference:

• R17 & C9 on the factory BM789 are:
    R17: soldered with a 0Ω resistor
    C9: spare (without any capacitor soldered)
• The modification consists of:
     R17: replace the 0Ω resistor with a 25kΩ one
     C9: install a 0.1μF capacitor

I have implemented the above modification on my meter and I can confirm the continuity behaviour is now similar to that of other meters mentioned here (e.g. BM257s, 121GW). 
I have asked Brymen and they confirmed this change impacts only the continuity measurements on the meter and no other function.

Hats off to Brymen for looking into this and coming up with a resolution.

[mwb1100]

Hats off to Brymen for looking into this and coming up with a resolution.

Indeed - major kudos to Brymen!  I don’t think many vendors would take that kind of effort.

I just have to ask:  is the expected continuity response still fast?

[giosif]

I just have to ask:  is the expected continuity response still fast?

From my observation, it is indistinguishable from before the modification, in terms of response time.
Some people with better hearing (or faster neurons ) might notice a difference, though.

[NoisyBoy]

Wow, their response surprised me, in a good way. 

The only odd thing is if they provide hack to change hardware components, why the insistence to not allow user firmware upgrade?  It has less risk than swapping out SMD components. 

I like their DMMs, and I would love them even more if they allow user firmware updates. 

[Antrus]

The lower the resistance R17, the higher the speed. Right?  The multimeter controller is very similar to the MSP430F.

[Neutrion]

So FW 08 here and can't replicate the touching anything and it beeps stuff. Metall non metall etc. tried.

Although I am not living in front of some big radar dome or having a big Tesla coil in the flat so might our circumstances differ a bit as others have pointed out. Watching politics sometimes, also doesn't seems to charge things up around me enough to have it detected by the meter.

But the behaviour with some capacitance in circuit sometimes gives false positives that is true, so I am curious how this recomended mod changes that. I hope someone does the mod to test it before me.

Also not sure if one can access the parts properly because of the can on the 789 what the 786 doesn't have. Don't have picture in front of me of its guts.

[SanFable]

Hi,

Just to provide an update that Brymen did eventually get back to me with an update on changing the response time of the BM789 for continuity mode.
And the change is a hardware modification (which Brymen called a "hack") - using the attached picture as reference:

• R17 & C9 on the factory BM789 are:
    R17: soldered with a 0Ω resistor
    C9: spare (without any capacitor soldered)
• The modification consists of:
     R17: replace the 0Ω resistor with a 25kΩ one
     C9: install a 0.1μF capacitor

I have implemented the above modification on my meter and I can confirm the continuity behaviour is now similar to that of other meters mentioned here (e.g. BM257s, 121GW). 
I have asked Brymen and they confirmed this change impacts only the continuity measurements on the meter and no other function.

Hats off to Brymen for looking into this and coming up with a resolution.

Great news,

You mention R17 and C9, in the picture we can see marked R12, any idea why it got marked too?
Did you miss something?

[Neutrion]

That is a good question. I hope giosif will get back to them.

[giosif]

I hope someone does the mod to test it before me.

As I mentioned in my previous post, I have already implemented this mod and it is working "as advertised".

Also not sure if one can access the parts properly because of the can on the 789 what the 786 doesn't have. Don't have picture in front of me of its guts.

Yes, I had to desolder the metal shield covering the part of the circuit where this mod needs to be applied.
However, the shield is soldered to the board only in a couple of locations, so it's not that painful to remove.

You mention R17 and C9, in the picture we can see marked R12, any idea why it got marked too?
Did you miss something?

The picture I attached came directly from Brymen, so don't know why R12 is marked as well.
In any case, this modification doesn't touch R12.

[tor47h]

BM786 - I had the same problem.
R17 25k + C9 100nF = excellent and correct solution 
Thanks for sharing.