EEVblog 1522 - BM786 Multimeter Repair PART 2
ฝัง
- เผยแพร่เมื่อ 3 ม.ค. 2023
- Part 2 of the BM786 troubleshooting repair, proving it's the processor at fault.
And Dave goofed up BIG TIME, did you spot the ERROR in the previous video?
Forum: www.eevblog.com/forum/blog/ee...
Support the EEVblog on:
Patreon: / eevblog
Web Site: www.eevblog.com
Other channels:
EEVblog2: / eevblog2
EEVdiscover: / eevdiscover
T-Shirts: teespring.com/stores/eevblog
#ElectronicsCreators #Repair #Multimeter - วิทยาศาสตร์และเทคโนโลยี
![Hacking Reality [Official Film]](http://i.ytimg.com/vi/vJi3_znm7ZE/mqdefault.jpg)
![[TH] Esports World Cup : Grand Final](http://i.ytimg.com/vi/sedbKhFYDMo/mqdefault.jpg)
![[UNCUT] ลูกหมี-ลิลลี่ เหงียน เปิดใจฟาด ปู มัณฑนา I คนดังนั่งเคลียร์ I 15 ก.ค.67](http://i.ytimg.com/vi/FKQ60Y7bJ6s/mqdefault.jpg)
My BM869 shows that insertion error when the battery is really low. Might be the same issue here.
The original batteries on my BM235 where slightly smaller than standard ones. When the MM got vibrations or slight shocks, it showed intermittent InEr . I thought it was a faulty device, but after replacing the batteries, everything was fine!
I am glad that you followed through on the diagnoses.
I work at an electronics manufacturer as the rework/repair tech, I've seen a lot of cases where it LOOKS like the MCU is dead, but there's another fault making the board fail, like a backwards chip on the same power rail as the micro that loads the line down when powered, broken resistors going to the programming header, lifted leads on the micro, ect
But given this unit has worked at one point and you were able to program a donor micro on the faulty board, it was definitely a dead micro in this case as you had suspected earlier.
The resistor change definitely is a bit odd though. Maybe a voltage divider that was changed to go hard to the voltage rail or ground? No idea.
get some kittens and send one to Dave.....
Seen something similar with TI and Microchip where, depending on what else is being powered up when a programmer is connected, the power rail and signals would dip slightly and that was sufficient for the micro to not respond. Only by increasing the programmer voltage by 0.5V did the micro 'appear' (but then you have to be careful with max voltages).
I like that you don’t hide the goof, it happen to all being in a unknown territory.
Sometimes an STM32 will reconfigure its programming pins as GPIO either corrupted flash or intentional code options (program pins are alternate functions on GPIO) .. to recover you have to reset the chip and activate the pins in programming mode immediately before the application completes initialisation.. Which involved manually holding it reset then releasing with a normal programmer polling the SWD pins.. Then often needing a bulk erase from STs own utility rather than a third party programming application.. also the BOOT0 boot mode pin that invokes a boot loader is susceptible to noise and will crash the processor at any time if toggled over half the rate of peripheral clocks despite only being "used" just after reset to choose bootloader flash. We hit it with 10ns glitches by mistake and it crashed in one design.
at 32:52 No not dumb or embarrassing. As a late in life beginner into electronics it is MOTIVATING! This teaches me that if even you can make mistakes. So can I and learn from them!
That was awesome, I was invested wanting to know what the issue was. Was glad you had a donor board just to confirm it! Thanks for keeping all your goofs and process in the video, really nice to learn that way following your train of thought. I never fault you for goofs or anything, and I don't think others do... almost certain people prefer to see them like I do. There's a lot to learn in goofs!
Most definitely almost impossible to work at your normal capacity while thinking about the video you are making at the same time. Normally you'd just be silent completely absorbed in the debugging task, which absolutely would result in many fewer errors. Again, glad you leave that stuff in - rather than try to make your videos a showcase of your talent like others do. Anyone who holds that against someone's "skills/knowledge" simply hasn't ever tried making an engaging uncut video while working.
You were invested with Dave.
Anyone who claims they are 'goofless' (to coin a phrase, lol) is a liar.
I prefer the warts and all videos too
@@arongooch Yeah i guess bad wording :)
I always like the repair videos
Thanks for the mention too !
I appreciate the repair videos, and esp. the time you took to remove all the pauses/etc.
The "non working" chip may just be locked though.
If so, depending on how the protection works, it may need to be fully reset/erased or it may not be a reversible process when they blew the security fuses/locked it.
I also winder if the InEr is due to the fiddling with the xtal for the instrumentation chip.
You measured the processor voltage on the supply side or R150... I would be checking on the pins instead in case the resistor is open.
I did. No problem.
Glad that you were able to solve the problem
If it so hard repair a electronic device for a guy that has " in theory" all information about that device, imagine how hard is repair for a guy that has no one information about that device.
In Part 1 Dave says that brymen did not give him all the info for the board, so he has to trace out all the lines to find out where everything goes.
The harder it is the more rewarding the experience, if it was easy it would be boring
This is why we need to fight for Right to Repair.
Apart from knowing what the micro is, I have no other technical information.
@@felixcosty ahh and? Most people also dont have schematics either - but using a miltimeter trace down the componenet. Dave just seems to jump all over the board instead of startiing at one point and working through it. Has he even checked a short to ground and traced to were its the least resistance? Look through his past videos - not alot of successful fixes...
So cool, hands-on after learning amazing theories education on electronics from sir Dave.
God bless sir Dave.
My boss once spend several houra debugging a memory leak in his code when the production team populated a 16K FRAM instead of a 64K FRAM.
It happens. :P
NEVER, ..EVER, underestimate, the Human capacity, for plain ..STUPIDITY..! 🙄
With that dmm beeps it sounds like Dave swears time to time 😂
Great work.
Dave, I always enjoy watching your videos! You are a very respectable member of the electronic community in my opinion. Thanks for showing us your barbaric reworking skills!😁
Good work.
Hey Dave thank you for the repair video. Now I wonder if swapping out the 000R resistor at R104 to 10K would resolve the problem sense that is the multi meter ic. And if that doesn't work then adding 150k resistor to R105.
Hi ,
Sorry , I was looking out for you !! I thought you’d be following Virgin Orbit flight? Ohh well.
Sorry I didn’t know where to post, so I snook in here😳😁👍
Thanks!
As a plumber I really enjoy your videos. I have a favorite plumbing flux too, other fluxes don't work the same. I've always wanted to fill some assholes electronics flux dispenser with plumbing flux. They'll love it at first....... then in about a month the copper will be all gone from anywhere the flux was left on. The good part is there will be no traces to fix.
There was a TV studio, I think in the Green Bay area, that back in the 80's had a tiny fire in a trash can. It was quickly extinguished and people thought that was that. When they returned after the weekend, there was copper oxide dust all over the place and not surprisingly, nothing worked.
I'd love to know what was in the fire extinguisher.
And I too am picky about plumbing solder. I don't do a lot of it, like it isn't my job. But I've been picking away at home renovations. The water-soluble stuff is awful. But I'd love a recommendation on a brand that comes in fairly small containers. I'm not too bothered it if is relatively expensive as a small quantity will last a long time. It is the frustration of needing to rework, especially after the pipes have been pressurized with water that I'd like to avoid. I can't always pressure test with air.
@@strehlow I never test with air but I almost never have leaks on small pipes 1/2" up to 1". Larger than that is easier to have leaks, 2" I always have a few leaks but I also put in really good quality unions so I can remove entire branches to dry them out to rework a joint. As for flux I prefer Oatey brand. I'm really picky about my plumbing flux and I've had fluxes that don't really work for me for some reason. Oatey visibly starts to boil and sizzle when it's hot enough for solder so it's a great way to know when you're hot enough. As soon as it sizzles start applying solder. The biggest thing though is to thoroughly clean the pipe ends, clean clean clean then apply flux, same with electronics. If you have to rework a joint also clean the black shit off the outside of the fitting so it won't contaminate nearby clean copper. My reworks are nearly 100% if I do that. Overheating is probably the #1 mistake to have bad joint, heat just enough until the solder melts rapidly then remove the heat. I use propane too and that seems to help prevent sooting vs acetylene or Mapp gas. Just a green can of camping propane works really well and it's cheap.
I've still got the protective film on my uni-t 139c
When you make electronics videos you're great!
"The controller chip for our water purification system has given up a ghost. We can't make another one, and the process is too complicated for a workaround system..."
19:38 look at them pesky little solder bridge buggers on the upper side! I'm sure (at least hope so) it was fixed before firing the meter up :)
By the way, the acronym for "catch you next time", but with "you" contracting to U... haha!
Yes, I wicked those up.
@@EEVblog sounds wicked... nice :)
See you next Tuesday.
Exelent video series, tks 🎩 Mrs. Jones,.
Send it to Scotland Dave, I'll have a crack at it.....😀
part 2! yay
Is the bus contention actually good? Implying some degree of logical progression/ life in the thing?
I've lost many days of engineering time to stupid errors, the happen, don't fret it, accept three hard earned wisdom and nice on with your life.
Nice continuity beep. No high annoying one. Like the on screen scope and meters a lot
Wow. You already anticipated the dead chip and now, you nailed the coffin.
Notice difference between 17:08 (faulty unit) and 17:11 (donor unit)?
Faulty unit seems to have metallic hexagon screw type of head? shorting out something perhaps? or its just metallic color.
and donor unit has black plastic cap/head?
can i ask you ... on silkscreen, the thick traces ... its for additional isolation or just info graphics?
As for the InErr, is it possible that since the processor from the donor board had a different resistor, that perhaps that resistor needs to follow the processor to the fix board?
If mem serves, the fix board had an upside down 0 ohm resistor while the donor board had something like a 1K ???
Perhaps the mod was for a slightly diff. proc chip!???
sometimes as a first guess I'm using IR camera or if there are not many components just checking all of them one by one. it's less thinking but can give fast lucky find. specially for people with limited electronics knowledge before they will put the part in garbage bin
Can you do a video on antenna design?
Very interesting, thanks for sharing :-)
no thats the sign off an ace stay cool
Shorted pin on top row at 19:40
Question about meters in general. One of my meters measured 350 ma of current draw on my car battery. I switched it over to the 10 amp scale and it read 3.5 amps. Is that meter broken? I thought it should read .35 amps.
are all the pins well soldered? maybe one pin is not making proper contact
Just a thought, did you change the resistor values on the repair board to the values you found on the donor board?
You said it's not a repair video but the title clearly says repair 😉
Obvious logic would say with 2 boards if components had different values, one was bad. However, this could mean the 2 boards were different revisions. More likely the first. Shorting on board would be a big problem and possibly also destructive.
Part 3!
Did you switch the clock with the chip?
Good to see you make the same kinds of goofs as me! Be really helpful if you had the schematic ;-)
Honest to god, Dave, I was convinced the other QFP was the processor roughly half-way through the last video, simply by the huge number of traces going to it. I never commented on it as I'm not an electronics engineer and didn't want to make a fool of myself, to be brutally honest with you.
I always get picked up on errors like this but in this case not a single person watching the previous video noticed it!
شكرا على الفيديو .من أي شركة او معمل صنع مقياسك bm786
Did it lose internal ground?
Hi Dave, i think the error is Initialisation Error, maybe from ic replacement and wrong calibration.
From the manual: "The meter beeps as well as displays “InEr” to warn the user against possible damage
to the meter due to improper connections to the uA, mA, or A input jacks when another
function, especially a voltage function, is selected."
@@Another8Years Um, how about "no connections" to the jacks?
@@Another8Years Um, how about "no connections" to the jacks?
"Just the tip"
Have you ever been instructed to bake a PCB for 24Hrs before desoldering a QFP. It prevents any moisture, absorbed by the plastic, from out gassing and damaging the QFP.
Yes, popcorning as its called. Why those chips have a shelf life before installation, reducing when the hermetic vacum foil packaging is opened on the tray or reel of chips.. Worse in humid countries, so designs out sourced from Europe to Asia were often delivered with wrecked ICs where similar handling in Europe did not result in popcorning...
Baking is part of the recovery process for those chips before reflow soldering.
could it be that the 0-ohm resistor vs the older version of the board causing the error? as if this version of the processor working correctly with the other arrangement?
Doubt it. That's part of the LCD driver by the looks of it.
A cap and a 0 ohm sounds like an error. Then the power supply or chip may be stressed and U know the board fails after a week.
@@danc2014 Yeah it's pretty strange to have a zero ohm on one board and a 10k or so on the other. Big difference. I notice the other resistor was 01C marked. It could be mistaken for 000 if you weren't careful, I suppose. Now was that factory rework, or user rework...?
According to the datasheet, that's VLCD. Change in the liquid crystal chemistry?
looked like one though , cool
That's the Porsche 911 of fluxes! 🤣
Is the meter just saying that you have the leads in the wrong ports or not at all, the common port looked like a split one for sensing.... maybe??
love to the other ausies
Wallaby hairs corrupting the ppbus.
It's ALWAYS the ppbus
@@EEVblog ppbus_g3h, the words that louis rossman has burned onto my brain
@@EEVblog lmao
Could the Insertion error be a false alarm related incorrect DMM input jack connection?
Бывает при грязных токовых гнездах.
Have you tried a hard reset after the firmware update ?
Yes, makes no difference.
@@EEVblog Well it was a long shot.
or maybe you could wipe the chip and reinstall the update. its another long shot you never know it might work. if it doesn't it probably has EMP damage from what caused it maybe from a microwave. lol 😂😂🤣🤣😂😂🤣🤣😂😂
What Instrument using for data on screen?
I just screen capture a National Instruments meter
Did that Multimeter somehow get shocked? I mean a silicon failure and then the Insertion error..... Yeah this looks to me (non expert guy) like a failure due to Overvoltage / ESD / EMP
Yes I'm not an expert, you've seen that correctly. This doesn't stop my interest in such technical stuff tho.
The chip is very well protected against such things.
You never know. It should be well protected. But you can never be sure. I killed a WiFi router once with ESD 🤣 luckily it was EOL already.
@@EinGamer22 : Can't compare consumer grade crap to measurement instruments, not even in the same ballpark. Haha...
@@bigcheese781 actually true. Like I've said I'm no expert in this stuff.
Why I'm thinking about overvoltage: the silicon lab should check every single chip for defects. So there should only be good chips. A bond wire could have broke tho (idk why I mean that chip won't get to 70+ degrees Celsius.)
Also how is the insertion error occuring. Is that a defective part on the board or just a simple Software issue?
@@EinGamer22 I think when the chips pass the tests at factory, some of them are really close to be defective (poor bond wire contact, badly formed gate on silicon, thinner traces at some places...) but they're just good enough to pass the tests. Then they end up failing early in real use. I guess there is no solution for that problem unless there is some testing that checks the whole thing at a molecular level rather than just testing its functionality
11:12 looks cortex m3 ,likely stm32f1o0xx?
Nope.
What a faulty meter
3:15 Come on, you really don't know? U2 makes music, and have been doing so from a long time.
Poor beepers. I've melted a few.
very nice repair video 🥳 I just purchased a cheapy $57 HoldPeak_HP-770D CAT III Auto Range 40000 Counts DMM BarGraph MultiMeter, seems to work OK on hobby stuff?
Now for the important news. April 20 2023, 11.27am local time, Total Solar Eclipse over Western Australia’s Coral Coast. Are there any venomous snakes out where the tourist will be?
Дейв, нали знаеш че си гений.
What the hell is wrong with those crystals? why the jumper wire? design error? I can also spot some vias drills almost directly on top of each other. why?
Dave fitted that crystal in the last episode because the oscillator wasn't working and he thought it might be a bad crystal. He didn't have the correct part so kludged that one on.
Watch part 1, A pad was lifted removing the crystal in the previous video.
Diodeb D7 is broken🙃
I guess the next step is find out what chip it is, ie, it's probably a relabelled off-the-shelf chip, that way people can repair their DMM when that failure prone chip dies.
Why did you sssume that "InEr" = Insertion? It is Initialisation, obviously. As far as I cam see, you soldered the chip poorly, some pins are probably loose.
This is 'Input error'. Swears at current inputs.
It got damaged by a cosmic rays 😆 😂 😆
Теперь все знают , что чип это Stm32.
No one would be watching the video frame by frame to catch up the goof up....
I literally spent like 10 minutes talking about it and poking at it. Plenty fo time to notice it's the wrong chip, but not a single viewer noticed it!
shame on you 4 seing that
😁Hi
Hi
Your audience expected more from you too apparently
If even you can make mistakes then I feel a little less bad about mine
Brymen should release firmware updates to public!
Это политика страусов.
После случая с измерением сопротивлений.
@@funcool8884 именно! Они конкретно облажались! Ежу понятно, что любая прошивка имеет баги, и не давать возможность обновлять их бесплатно, выглядят очень подло со стороны производителя.
Thank you Dave. This is why I pay the $$$ for Fluke. Both professionally and privately. Their goof up, would be honored, repaired and returned a month and a half.
Or just buy the Bryman, and if it fails buy a new one which would be delivered well within 1 1/2 months. And you'll still be quids in.
The BM786 was under warranty (it was 9 months old according to a post in the forum). So presumably the buyer got a replacement. The Brymen does have a shortish warranty (1 year) - but you're paying at least $150 more for an equivalent Fluke, so the Fluke warranty isn't exactly free.
wait hpne 4 a beeer be back loojs cooool
It's fixed! Like a neutered cat.
A whole lot of useless components. It's time to make something with a normal color TFT LCD with a built-in electronics and a lot of measuring modes. In the same space, it is possible to integrate a signal generator and signal logger with 10x precision with internal calibration tables - there is no problem with built-in memory for that nowadays and normal MODERN multimeter functions, not defined by marketing, such as L measurements with low current milliohm meter. Like what is the problem ? It is solvable by modern integrated circuits.
Use ChatGPT to help you get the data you needed for repair?
To ChatGPT: "Write a poem describing how to fix my multimeter. Then draw some images showing giant multimeters rampaging through Sydney."
Good thing you had them put your name on theses pieces of crap and convinced everyone to buy them.
That’s super sloppy.
I am studying engineering and I can't understand it
Maybe it’s not for you 😂
Keep at it, don’t give up.
What part can't you understand?
Keep going components
Don’t worry..! NO, Engineers, can understand it..! That’s ..WHY, they ..always employ ..Repair Techs…! 😉😁😂😂😂
I have never seen you fix anything. You need a better more logical fault finding approach perhaps.
1st of all, an multimeter it’s the main tool in EE lab. You should purchase a trusty piece of equipment and not “Made in 3rd world” POS. OK, your budged is very tight, then buy an 5 - 20 bucks DMM, save money and buy a trusty equipment designed/made in normal world.
Otherwise you will spend more time in troubleshooting the multimeter than the real problem… Think twice and buy a quality product like Fluke, Keithley, Keysight, etc.
Can you trust the EEVblog branded DMM? Even the creator(??? I doubt he even know the design...) / sponsor doesn’t have the schematics !!!
Is Brymen better than KAIWEETS? ...the same crap, only in different package, and the price is different :)
> Is Brymen better than KAIWEETS?
Yes.
Brymen are not POS (and Taiwan is not 3rd world), but to each his own. Fluke, Keithley, Keysight are much more expensive. I have some Flukes, and I have some Brymens. I use the Brymens more often. They're more compact and more readable. I use a Fluke for car work because I have a magnetic strap for it.