Handheld meter robustness testing

[MiroS]

After two Part 1 and 2   not sure if that multimeter make any sense over e.g. BM867s.  I am starting to thing that it may be actually a kind of lower cost (for Brymen) variant.  I can see only one good thing so far - location of bargraph  is a way more optimal than in BM869s/BM867s. Do both use the same chips ?

Brymen should start to think less conservative way and add e.g. LED display to make this thing more attractive. 
Maybe at least the price will be really a long way below 150.

[EEVblog]

After two Part 1 and 2   not sure if that multimeter make any sense over e.g. BM867s.

It's smaller, it costs less, bigger display, it has AutoHold, and it has an EF tester.

[rsjsouza]

After two Part 1 and 2   not sure if that multimeter make any sense over e.g. BM867s.

It's smaller, it costs less, bigger display, it has AutoHold, and it has an EF tester.

I agree. The new meter seems to fit in a category that is more concerned about a general purpose use (lab/field) and does not care for dual display - something similar to choosing a BM857/9 versus a BM867/9. Besides, the 60000 counts can come in handy if you use 5V precision systems (the extra room above 5V can be of help).

[joeqsmith]

Talk about responsive, Brymen is working on it.  They suspect a problem with the new firmware.  Apparently earlier versions don't have the problem.  It sounds like they have a good handle on it.   I plan to hold off on further testing until they have it sorted out.     

[rsjsouza]

Talk about responsive, Brymen is working on it.  They suspect a problem with the new firmware.  Apparently earlier versions don't have the problem.  It sounds like they have a good handle on it.   I plan to hold off on further testing until they have it sorted out.   

Funny how, in the past, that is what we would call it "beta testing": an almost finished product released to a few key customers to be trialed by fire. Nowadays this is almost the exception: alpha stage products released to the broad market where the issues are fed back to the devs through customer support. Oh well... Congratulations in reporting the bug.

Now, if others could get a clue and follow the same pattern of development cycle and customer support...

(BTW, your findings prompted me to resurrect an old project of mine for a HV low current Power Supply to test DMMs in more daring conditions - as well as exercise several HV capacitors).

[MiroS]

After two Part 1 and 2   not sure if that multimeter make any sense over e.g. BM867s.

It's smaller, it costs less, bigger display, it has AutoHold, and it has an EF tester.

I agree. The new meter seems to fit in a category that is more concerned about a general purpose use (lab/field) and does not care for dual display - something similar to choosing a BM857/9 versus a BM867/9. Besides, the 60000 counts can come in handy if you use 5V precision systems (the extra room above 5V can be of help).

Really I would not be sure , 857s price is 133EUR  also below 150 USD  ... that would maybe sense if that would cost below 100 or so.
857s seems to be far more universal one.

[MiroS]

Talk about responsive, Brymen is working on it.  They suspect a problem with the new firmware.  Apparently earlier versions don't have the problem.  It sounds like they have a good handle on it.   I plan to hold off on further testing until they have it sorted out.   

It looks that my rule 'stay away from fresh multimeter' may work again 

[rsjsouza]

After two Part 1 and 2   not sure if that multimeter make any sense over e.g. BM867s.

It's smaller, it costs less, bigger display, it has AutoHold, and it has an EF tester.

I agree. The new meter seems to fit in a category that is more concerned about a general purpose use (lab/field) and does not care for dual display - something similar to choosing a BM857/9 versus a BM867/9. Besides, the 60000 counts can come in handy if you use 5V precision systems (the extra room above 5V can be of help).

Really I would not be sure , 857s price is 133EUR  also below 150 USD  ... that would maybe sense if that would cost below 100 or so.
857s seems to be far more universal one.

Exchange rates vary, but today 133 EUR = 157 USD, so there is no difference.
 
From the page I see at TME, the BM857 is 145 USD, which even still puts it in the same category of the other two (BM867 shows for me at 156 USD), making the difference too small to be considered another class.

The BM857 and the BM786 are quite similar with an edge for the BM786 if you don't need to do battery or trend measurements (the 500k counts mode)

[MiroS]

After two Part 1 and 2   not sure if that multimeter make any sense over e.g. BM867s.

It's smaller, it costs less, bigger display, it has AutoHold, and it has an EF tester.

I agree. The new meter seems to fit in a category that is more concerned about a general purpose use (lab/field) and does not care for dual display - something similar to choosing a BM857/9 versus a BM867/9. Besides, the 60000 counts can come in handy if you use 5V precision systems (the extra room above 5V can be of help).

Really I would not be sure , 857s price is 133EUR  also below 150 USD  ... that would maybe sense if that would cost below 100 or so.
857s seems to be far more universal one.

Exchange rates vary, but today 133 EUR = 157 USD, so there is no difference.
 
From the page I see at TME, the BM857 is 145 USD, which even still puts it in the same category of the other two (BM867 shows for me at 156 USD), making the difference too small to be considered another class.

The BM857 and the BM786 are quite similar with an edge for the BM786 if you don't need to do battery or trend measurements (the 500k counts mode)

Oh me ...  I meant not BM657, but BM867S ... sorry for the mess.

[joeqsmith]

Talk about responsive, Brymen is working on it.  They suspect a problem with the new firmware.  Apparently earlier versions don't have the problem.  It sounds like they have a good handle on it.   I plan to hold off on further testing until they have it sorted out.   

Funny how, in the past, that is what we would call it "beta testing": an almost finished product released to a few key customers to be trialed by fire. Nowadays this is almost the exception: alpha stage products released to the broad market where the issues are fed back to the devs through customer support. Oh well... Congratulations in reporting the bug.

Now, if others could get a clue and follow the same pattern of development cycle and customer support...

(BTW, your findings prompted me to resurrect an old project of mine for a HV low current Power Supply to test DMMs in more daring conditions - as well as exercise several HV capacitors).

I'm looking forward to seeing what you come up with for your own test standards and equipment.    When you start working on it and running them, feel free to expand on this thread. 

[joeqsmith]

Sounds like Brymen may have a fix.  I'm surprised that they responded this quick, but it also sounded like they had a good idea where the problem was when I first wrote them.   

While I offered to change the microcontroller,  they have shipped a new meter.   I should have an update in a few days for you. 

[EEVblog]

Sounds like Brymen may have a fix.  I'm surprised that they responded this quick, but it also sounded like they had a good idea where the problem was when I first wrote them.   
While I offered to change the microcontroller,  they have shipped a new meter.   I should have an update in a few days for you.

Sorry, haven't been following, is this the BM786?
If so what was the issue?

[MiroS]

Yes it is, minus EEVblog blue case (see reply #3770) An issue with AC+DC was found, so we get the benefit of Joe's testing.

I think testing is not finished yet,  Joe may find new one

[joeqsmith]

Yes it is, minus EEVblog blue case (see reply #3770) An issue with AC+DC was found, so we get the benefit of Joe's testing.

I think testing is not finished yet,  Joe may find new one

It's possible.   With Brymen sending a whole new meter,  most likely I will run through all the basic checks off camera.   I doubt viewers will want to sit through that boring checkout a second time.   Assuming it works, I'll make mention of it and then show it running through all the problematic tests.  Most likely, I will then run the meter through a series of waveforms.   By the end of Part 3,  I want to viewers as well as myself, to be confident that this problem has been resolved.   

I just checked the tracking and it looks like it is due to arrive in just a few days, so it won't be long. 

[gnavigator1007]

Will you still run the problem meter through the rest of the tests?

[joeqsmith]

Will you still run the problem meter through the rest of the tests?

Knowing it has a problem, I am not sure what we would learn.  People are already confused stating it's an AC+DC problem and it has nothing to do with that mode.  Adding another meter to the mix would surely add to the confusion.

What I have thought about do is to pull them both apart and verify there is no difference beyond the new firmware.    The life cycle test requires several days to run.  Transient testing both meters would also take a fair amount of time.   

If you have some thoughts about it, feel free to chime in.  If for example you were really only interested in seeing both meters on the waveform test, I could maybe paint one of the meters blue or something to clearly show which one is which.   It wouldn't be any effort in this case to run the two in parallel.  But again, considering it has a problem and the whole point it to show it was corrected, I don't see what we gain.

[gnavigator1007]

Mostly just curious, but also thinking it might be a chance to see one fail early while the other manages to run most of the gauntlet, similar to what happened with the 87v tests. Been too long since I watched those now so I can't remember if there were actual differences in the board or not. Maybe not fair as one already is known to have a problem even if it is just firmware issue. Guess I was really just wanting to see the maximum number of meters fail (that are not mine) and I'm just being greedy.

[joeqsmith]

Mostly just curious, but also thinking it might be a chance to see one fail early while the other manages to run most of the gauntlet, similar to what happened with the 87v tests. Been too long since I watched those now so I can't remember if there were actual differences in the board or not. Maybe not fair as one already is known to have a problem even if it is just firmware issue. Guess I was really just wanting to see the maximum number of meters fail (that are not mine) and I'm just being greedy.

The 87V that failed was quite old but I don't think that had anything to do with it.  You may recall after running a brand new one and seeing how well it held up,  I carefully ran the old one a second time going through each stage trying to sort out why it failed.  In the end, I had no answer.  I repeated the test on that same meter and it held up really well.  It seems when I dug into it, even some stray contamination on the board didn't make a lot of sense. 

In the case of the 87V, we had a hardware failure.  Running a second new one did at least give me enough confidence to have a second look at the old one. 

Now we apparently have a firmware problem and no hardware damage.   I think we have looked at enough Brymen products to be fairly confident that they know how to design a switch that will last and a front end that can take a fair bit of abuse.   I highly suspect we are looking at some fairly boring tests moving forward.   Of course, I thought that when I started too. 

I do have a few old Harbor Freight meters if you just want to see some hardware damage.     

[Chalton_trc]

Is the auto-range problem fixed on the Brymen 786? Does the 789 suffer from the same problem or other problems?

[EEVblog]

Does the 789 suffer from the same problem or other problems?

It would be the exact same code base, so almost certainly. Brymen said they have fixed it.

[joeqsmith]

Is the auto-range problem fixed on the Brymen 786? Does the 789 suffer from the same problem or other problems?

Welcome to the forum.   

I'm not in any sort of rush and am giving Brymen time to sort it out.  Until then, I have no plans to proceed with further testing.   It sounds like they have told Dave it's all working now, so maybe we can get back into it soon.

[EEVblog]

Is the auto-range problem fixed on the Brymen 786? Does the 789 suffer from the same problem or other problems?

I'm not in any sort of rush and am giving Brymen time to sort it out.  Until then, I have no plans to proceed with further testing.   It sounds like they have told Dave it's all working now, so maybe we can get back into it soon.

They said they fixed the issue you raised and sent you a new meter. Not sure about the other stuff they hinted at.
Expect maybe a few years and thousands of units in the field to get all the small issues ironed out, that's what the BM235 took.

[bc888]

Nope! My BM235 was perfect and I got an early model. In fact I'm still in love with it although Tektronix is my regular one I have on the dinning room buffet...to the wife's annoyance. I want one of these new Eevblog Brymens for a potential son in law's Christmas present. Some folks are just hard to buy for and I don't want to give up my bm235. 

Dave, please convey to Brymen that Christmas is near!!!! Joe, (on occasion and this is one) you rock buddy!

[joeqsmith]

They said they fixed the issue you raised and sent you a new meter. Not sure about the other stuff they hinted at.
Expect maybe a few years and thousands of units in the field to get all the small issues ironed out, that's what the BM235 took.

No doubt they are busy.  I haven't heard anything from them in a while.

[EEVblog]

Nope! My BM235 was perfect and I got an early model.

It's been through probably four firmware revisions since it was released fixing small things. The issue that Joe found on the new meter most people would never notice, and I've seen that on other meters on the market. There is actually an issue (confirmed non-fixable though, Brymen call it a design limitation) in the BM235 that to my knowledge no one has encountered yet. Also present in some other meters on the market.
There are just so many permutations of signals and ranges and functions possible it's likely you could eventually find an issue with any meter on the market. Like it took a decade before someone found you could brick a Fluke 87V with a GSM signal.