OFFICIAL RESPONSE FROM AGILENT:

"We have determined that a short on our LAN/VGA module is what caused the EEVblog unit to stop working. The short caused a resister on the acquisition board to fail, bringing down our 3.3V supply path and causing the unit to no longer boot. After further investigation, it was determined that a small run of early production LAN/VGA modules were susceptible to this short. Although the issue is relegated to this small initial run of LAN/VGA modules, we are replacing all LAN/VGA modules in customer hands and have implemented additional testing procedures on our LAN/VGA manufacturing line to eliminate this issue from happening again."

What does $12000 magic smoke smell like?
Murphy ensures that Dave always gets the outliers on the production bell curve for review!

Hi Welcome to the Eev blog an Electronics Engineering Video Blog of interest to anyone involved in electronics Design. I'm your host Dave Jones Uh-oh Looks like we have something wrong I Was just playing around with this new Agilant 3000 Series and look not a it started to lock up. The first thing I noticed was that the menu options just did absolutely nothing. and then I started getting all this garbage and tearing on the display and now trust me, that ain't a signal cuz there's nothing going in there that is something has gone horribly wrong with.

like the acquisition uh, frame buffer or the capture engine. something like that that's actually put in all that garbage on there. I Have no idea what, but something has gone horribly horribly wrong. Let's try a reboot and see if it fixes it.

Wait a bit, it switch it back on. It'll take a while because uh, these things take about 30 seconds to boot. No look, look, it's completely. It's completely died.

It has completely died. No. hang on. I Saw a little flicker of Life there.

No, no, it is completely and utterly dead. Let me disconnect the main plug it back in. Shouldn't have to do that because the uh power switch is actually a real power switch. It's not just a soft module power switch.

No, it is fail. It's completely and utterly died. What a loser! When it first failed I Could have sworn that I smelled something that characteristic electronic burning swell where the magic smoke has escaped and well I just can't smell it anymore. Try the fan.

Not really, it's it's it's gone. So I don't think I was imagining that I've smelled that a million times in my career. It's just I reckon something went poof. That's $12,000 Magic Smoke Love it! Well, the first rule of troubleshooting is they shall check voltages and that's exactly what I've done.

I've checked all the rails. There's plenty of different rails in this thing and they all look spot on. So uh, I don't think it's um, at first glance it doesn't appear to be anything to do with the power supply. I'd love to troubleshoot this further, but quite frankly I don't have the time.

so I'll just ship it back to Agant with a big fail sticker on it and they can investigate it and let us know what went wrong with it. $122,000 Scope brand spanking you out of the box work for a little bit and then poof. H Oh well, happens. See you.


Avatar photo

By YTB

26 thoughts on “Eevblog #147 – agilent 3000 $12000 smoke”
  1. Avataaar/Circle Created with python_avatars Fix The Tempo says:

    Good old nand memory corruption)))

  2. Avataaar/Circle Created with python_avatars Junker Zn says:

    It looks to me like the external data bus for the cpu has noise on it. The ram content itself is ok (otherwise the 'static' would be more permanent). how interesting! I've seen this effect before on some of my old hobby projects. I pulled the character generator ROM out of my old Commodore PET replaced it with a character generator static ram on a perf-board, with very long wires from the old socket to the perf-board, and got similar looking static on the screen due to the very long wires I was using. Since it was the character generator rom it didn't crash the machine. It was quite fun to play with it!

    But what could cause something like that on a scope? The frame buffer is in main memory most likely. Something in the CPU / digital-logic related to the CPU for certain. But all the voltages were ok? Hrmm…. maybe a crystal oscillator went poof. The static on the screen clearly corrupted the running program too since all the scope controls locked up. Points towards main memory or the main memory address or data bus.

    Hmm. Even a single bad address line could cause this sort of static, too.

    -Matt

  3. Avataaar/Circle Created with python_avatars DAN HOLMES says:

    I know what went horribly horribly wrong….. you paid too much for it to begin with.

  4. Avataaar/Circle Created with python_avatars zokonjazokonja says:

    In company we have lot of Cap/Tan meters from HP/Agilent (HP 4278 / Agilent 4288), and what I can say is that we have periodical problems with agilent units, they usually fail calibration at some point or just fail generally.
    On the other hand, old HP4278, heavy truck bastard, which some of them are even 20 years old, working without problem.
    They seem nondestructive, I think they would work even if atomic bomb explode near it, and would still measure inside specification.
    Once we open one HP4278 because fan didn't work, it was clogged with dust, it was almost one bucket of dust inside, when I clean it with compressed air I had to go outside on open air, when I activate air it was a huge cloud of dust which could be probably seen 1km away. Closed back hatch, send to calibration lab, pass all test, back to production.

  5. Avataaar/Circle Created with python_avatars mark says:

    where's the smoke u said smoke i want to see smoke

  6. Avataaar/Circle Created with python_avatars among us says:

    Here comes the algorithm

  7. Avataaar/Circle Created with python_avatars Atetus94 says:

    Shit happens, CYA Agilent! 😀

  8. Avataaar/Circle Created with python_avatars Snout Husky says:

    Not having the riches to buy one of these, if I did get one and it went poof – instant panic! >.< Heh. "You get what you paid for, and this time it was a chonkin' big jumpscare." Would have been even more amused if it actually did leak out a big puff of smoke in its dying gasp. 🙂

  9. Avataaar/Circle Created with python_avatars punker 4 real says:

    it cost 1,000$ per min to run it

  10. Avataaar/Circle Created with python_avatars punker 4 real says:

    TV screen? went

  11. Avataaar/Circle Created with python_avatars theSomething says:

    One died in the lab 2 weeks ago during startup…

  12. Avataaar/Circle Created with python_avatars N5KDA says:

    Infant death syndrome.

  13. Avataaar/Circle Created with python_avatars Paulius Mscichauskas says:

    180 no scope.

  14. Avataaar/Circle Created with python_avatars CHECK_ENGINE_ says:

    I HAVE THIS IN MY HOME

  15. Avataaar/Circle Created with python_avatars Jordan Johnson says:

    Good response from Agilent

  16. Avataaar/Circle Created with python_avatars niceguy60 says:

    Dingo did it.

  17. Avataaar/Circle Created with python_avatars manu bali says:

    Wouldn't it void your warranty?????? You opened it yourself

  18. Avataaar/Circle Created with python_avatars Dave unknown says:

    Hai!

  19. Avataaar/Circle Created with python_avatars Dalek Caiomhin says:

    Shit happens 🙂

    Got me an ancient Hameg HM312(3) because it didn't work, put in three pots, a proper mains receptacle astern, tested it out with a squarewave oscillator at 1MHz, and it works great. Cost of repair: about $15US. What sucks? It's 10MHz single trace. What else? Hameg sent me the only manual they have, and it's in German, photocopy PDF, making it impossible to easily online translate it.

    Okay, I know what I'm doing with a screwdriver and voltmeter, and now it works on all settings.

  20. Avataaar/Circle Created with python_avatars Peter Hodgkinson says:

    Well, it could have been used as a nice paperweight or doorstop, surely ?

  21. Avataaar/Circle Created with python_avatars Ernst Stavro Blofeld says:

    You broke it – you pay for it.

  22. Avataaar/Circle Created with python_avatars John Andrews says:

    After working with Agilent equipment in my lab for almost eight years, I've found them unreliable. I've had an E5250A catch fire (while idle), an expansion box for my 4156C failed twice. The power supply for our 4073A failed. The 81110A failed. And a power supply for an HP C3600 failed, but they only used to be the same company. And before I took over the lab, our expansion box for the 4156A failed, and the 4156A's VMU failed, The list of Agilent equipment in my lab that didn't fail is shorter.

  23. Avataaar/Circle Created with python_avatars Mystickneon says:

    Wun Hung Lo Buird Scope Makels Intelnationar, Inc.

  24. Avataaar/Circle Created with python_avatars iamdarkyoshi says:

    "I popped it! No more scope"

  25. Avataaar/Circle Created with python_avatars Son of a Zombie says:

    thumbs up for the video, thumbs down for Agitent.

  26. Avataaar/Circle Created with python_avatars KitCox says:

    On rare occasions I've beaten The Law Giver. Can you spell rare?

Leave a Reply

Your email address will not be published. Required fields are marked *