EEVblog #170 - Agilent U1272A Multimeter FAIL

Dave has rotten luck with test gear for review failing. What's up with his Agilent U1272A Multimeter?
FOLLOW UP: No, I did NOT damage the meter with ESD or anything else, and this is not an isolated incident. Other people on the EEVblog forum are reporting the same problem:
FOLLOW UP #2:
Agilent have admitted the problem is with the range switch glitching and have fixed it in firmware. They are now offering either a replacement unit or a FREE USB cable to all affected users so they can update the firmware. An excellent response from Agilent, well done!

[EEVblog]

@AgHandhelds Thanks for the reply Yip Wee. Good to see Agilent directly commenting on Youtube. I regret to inform that it is not an isolated incident, no less than 3 other people on my EEVblog forum have reported the same or similar issues. I have already reported this to Division via my Australian Agilent contact and they are looking into it.

[EEVblog]

@Deckardsvr Pretty sure it's not a soldering problem. That doesn't fit the symptoms of such a fault, and the soldering in this unit is first class.

[EEVblog]

That's the LowZ resistor (measures 1.7K) in the Agilent loading down the 9V battery I was using as the test voltage. IMO it shouldn't switch that in in OFF mode, only LowZ mode.

[EEVblog]

@enginerdy What am I supposed to do, get a calibrated ESD gun and systematically zap every joint on the board with a human body model charge?
I do like to think it is kind of a useful (although unspoken) part of the test, even if it is completely uncontrolled and vary by orders of magnitude. Same with taking stuff apart without the service guide or correct tools or procedures etc.

[JarkkoHautakorpi]

Came here cause my U1273A halts sometimes like for 5 minutes or so. Does not measure any voltage, removing battery does not help, it's just not working for a while. Tried all different modes, and no it's not in 'hold' mode. It must be a software bug. Super annoying, never happened with any other multimeter.

[MoeyfromBankstown]

An Agilent multimeter should be full proof. A company that is renowned in the production of high quality equipment, I must say I'm disappointed Agilent.

[rpdigital]

Looks like ACV selection makes it crazy, this is why I do like the ON/OFF button. 1st I select what I will measure, 2nd I will switch multimeter ON.

[Monpjc]

@stridermt2k I don't know your background but would suggest some reading of ESD failure modes in production and chip considered. ESD is not as big as risk as chip manufacture would have you think. Proving ESD failure is difficult - I know I have been though this is TI and paid large sums to have chips analysed.
As for Dave's handling of the board there is a very low risk. The devices will have its pins loaded externally so would dissipate most of the ESD. This issue is not ESD related.

[eurokid]

I've never been particularly impressed with Agilent multimeters. They're buggy and seem to have weird problems. I have a U1241A that does some weird stuff every now and then, doesn't instil a lot of confidence. Fluke is the only way to go if you want a multimeter that you can rely on day in and day out. Agilent really needs to step up their game if they want to compete with the big boys at Fluke.

[GreatJoe]

Before looking inside, I'd guess it's a floating joint, either on a passive or more likely one of the legs on a chip. It's impossible to tell with the naked eye, you'd need a beefy microscope just to stand a chance. If an error like that came to me, I'd start by using a solder wick on the pins of the main processor and then resolder the whole thing. Works on the PICs I work with.

[igsaturation]

Hi Dave, I've upload 2 videos that are close to the problem you discovered. I have not taken my meter apart, but noticed some wierd responses after hours of testing. It did not reoccur, but looked like your issue. So I tried moving from range to range until it suddenly locked up, and isolated it to the ON switch. Video responses are enclosed.

[emactan]

Had a similar problem with Sanwa CD771 couple years ago. I never even opened it up except to change batteries. I could it get to work again if I turn it off and on again several times. I'm not sure what caused it, but I occasionally forget to switch back to volts mode after measuring resistance. Perhaps the meter wasn't robust enough to withstand voltage on resistance range. Got fed up eventually and got myself my first Fluke.

[MarcelGlasgow]

Hi, Dave I love your videos and you are the best on YouTube so far:-) good luck with you channel.I was deciding between Fluke 87V and Agilent U1271A or bench DMM, but after seeing your videos about the Agilent DMM I am leaning towards the Fluke 87.thank you for saving me hassle with Agilent.

[mashersmasher]

Agilent really seems to be dropping the ball in your tests. I'm really glad you did the test on these

[RandyLott]

That's interesting. Maybe a ton of noise reaching the instrumentation amplifier and amplifying it creating error? Either way, it's not acceptable.

[sonicbaz]

Agilent should thank you for capturing that.

[stephenborst]

looks like it is a bit warm in your shop 185c

[winniemay]

Don’t worry, Agilent are very quality conscious. Just give them a call and they will handle it properly for you.

[superdau]

Just out of curiosity (and most certainly unrelated to the problem): why does the voltage on the Fluke drop by 0, 5V everytime you switch OFF the Agilent? I can't imagine a reason for that. If the measured voltage should change at all, it should go up a little, since in my understanding the off-position means completely disconnected probes (and magnitudes larger impedance) instead of the MOhms that the measuring circuit has.

[TerminalJack]

I have to laugh about people critiquing your test here. They obviously haven't seen the drop test you put all of those 50 dollar meters through or the canyoning test of that Fluke. This is Dave F'n Jones, people! LOL.

A person would think meter circuitry would be robust enough to survive someone monkeying around inside the case without an ESD strap--not that I'd push my luck.