Author Topic: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown  (Read 15245 times)

0 Members and 1 Guest are viewing this topic.

Offline PeterG

  • Frequent Contributor
  • **
  • Posts: 813
  • Country: au
EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« on: November 18, 2011, 09:39:27 am »
YouTube is saying it has been removed?

Did Dave remove it or is YouTube on the fritz?

Regards
Testing one two three...
 

Offline Kiriakos-GR

  • Super Contributor
  • ***
  • Banned!
  • Posts: 3525
  • Country: gr
  • User is banned.
    • Honda AX-1 rebuild
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #1 on: November 18, 2011, 09:52:20 am »
Working link here :    
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 29656
  • Country: au
    • EEVblog
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #2 on: November 18, 2011, 10:04:45 am »
YouTube is saying it has been removed?

Did Dave remove it or is YouTube on the fritz?

Both.
Youtube screwed the original upload. It turned out 4:3 with a corrupted start.
Deleting and re-uploading fixed the problem.
Also, I learned the hard way that Youtube does not allow you to upload the same video twice (it does a checkum), so you can't leave the old one up until the new one uploads, you have to delete first.

Dave.
 

HLA-27b

  • Guest
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #3 on: November 18, 2011, 10:24:52 am »
How about a small competition?

Who can guess correctly what the problem is before #218 is uploaded?

The winner gets special recognition for remotely diagnosing LeCroy 9384C Oscilloscopes.
Also my special bonus "Pi to the precision of the Feynman Point" for free.
 

Offline phil_jp1

  • Regular Contributor
  • *
  • Posts: 103
  • Country: ua
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #4 on: November 18, 2011, 11:37:20 am »
How about a small competition?

Who can guess correctly what the problem is before #218 is uploaded?

The winner gets special recognition for remotely diagnosing LeCroy 9384C Oscilloscopes.
Also my special bonus "Pi to the precision of the Feynman Point" for free.

My guess is video #218 would be not about LeCroy scope repair.
http://JumperOne.com - Electronic projects, tutorials, hacks, etc.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 29656
  • Country: au
    • EEVblog
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #5 on: November 18, 2011, 11:50:52 am »
My guess is video #218 would be not about LeCroy scope repair.

You are likely correct!
I have the service manual now, but it takes time to debug digest it and debug these things.
If anyone knows what #218 will be, please let me know, because I have no idea!

Well, actually, I do have 3 videos to edit from the HAM radio meeting I went to the other night. I don't really like giving those types of videos a blog number, but I guess it doesn't really matter and help identify them?  :-\

Dave.
« Last Edit: November 18, 2011, 11:52:34 am by EEVblog »
 

HLA-27b

  • Guest
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #6 on: November 18, 2011, 11:59:50 am »
My guess is video #218 would be not about LeCroy scope repair.
You are likely correct!....

Engineers for ya  ::)
 

Offline samgab

  • Frequent Contributor
  • **
  • Posts: 423
  • Country: nz
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #7 on: November 18, 2011, 12:51:43 pm »
My guess is video #218 would be not about LeCroy scope repair.

You are likely correct!
I have the service manual now, but it takes time to debug digest it and debug these things.
If anyone knows what #218 will be, please let me know, because I have no idea!

Well, actually, I do have 3 videos to edit from the HAM radio meeting I went to the other night. I don't really like giving those types of videos a blog number, but I guess it doesn't really matter and help identify them?  :-\

Dave.

Excluding the ham radio stuff, maybe 218 can be your solution to the Agilent upload problem and analysis of the second 1000 resistors? ;)
 

Offline lindworm

  • Contributor
  • Posts: 26
  • Country: de
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #8 on: November 18, 2011, 01:19:50 pm »
Seems like 217 isn't in the actual blog but on youtube only?
 

Offline McMonster

  • Frequent Contributor
  • **
  • Posts: 417
  • Country: pl
    • McMonster's blog
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #9 on: November 18, 2011, 01:30:52 pm »
Sometimes there's a delay of up to a day between video appearing on Youtube and eevblog.com.
 

Offline FreeThinker

  • Frequent Contributor
  • **
  • Posts: 791
  • Country: england
  • Truth through Thought
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #10 on: November 18, 2011, 01:39:27 pm »
How about a small competition?

Who can guess correctly what the problem is before #218 is uploaded?

The winner gets special recognition for remotely diagnosing LeCroy 9384C Oscilloscopes.
Also my special bonus "Pi to the precision of the Feynman Point" for free.

After watching the video my first best guess would be corrupt display ram. The Dacs (4 off) all appear to be working as normal and the corruption is Static in the waveform and consistent across channels. This would indicate a common link, either power supply or display? More twiddleing would be required but a possible test would be to zoom into the wave form and see if the corruption stayed in the same place (on the screen) or panned with data stream. This may help point to the problem. Check the memory stick is seated/ clean seen similar problems on video cards in a pc before. :)
Machines were mice and Men were lions once upon a time, but now that it's the opposite it's twice upon a time.
MOONDOG
 

Offline Kiriakos-GR

  • Super Contributor
  • ***
  • Banned!
  • Posts: 3525
  • Country: gr
  • User is banned.
    • Honda AX-1 rebuild
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #11 on: November 18, 2011, 03:30:59 pm »

After watching the video my first best guess would be corrupt display ram.

Looks that we are playing the wild guess game.  :)

In this case I would re-flow the display CPU,  memory issues are permanently and cause digital garbage on the screen.
Some hot air on the display CPU pins would show if it is responsible or not, before actually trying to re-flow it.

 

Offline firewalker

  • Super Contributor
  • ***
  • Posts: 2328
  • Country: gr
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #12 on: November 18, 2011, 03:47:33 pm »
Dave, does this scope use a standard "tv" CRT (easy obtainable)? Ghost images annoy me very much! I had an old monitor with windows 3.1 logo burned...

Alexander.
Become a realist, stay a dreamer.

 

Offline david77

  • Frequent Contributor
  • **
  • Posts: 793
  • Country: de
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #13 on: November 18, 2011, 04:22:04 pm »
One can safely assume it's not a standard TV tube. The amber colour gives it away, it is a tube more like they were used in old monochrome computer monitors back in the 80's.
 

Offline nixxon

  • Frequent Contributor
  • **
  • Posts: 313
  • Country: no
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #14 on: November 18, 2011, 06:49:49 pm »
Next time the scope is powered up it will probably work flawlessly. Dis- and reassembly have fixed stuff before...
 

Offline FreeThinker

  • Frequent Contributor
  • **
  • Posts: 791
  • Country: england
  • Truth through Thought
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #15 on: November 18, 2011, 07:34:17 pm »

After watching the video my first best guess would be corrupt display ram.

Looks that we are playing the wild guess game.  :)

Nice to see you post a sensible comment  ::).
Acquisition seems to be ok, triggering ok, time base ok. This would leave power supply problems, display, and vert amps. Looks likely that the vert amp or associated circuitry needs a look at. This however doesn't explain the display corruption and my suggestion of checking the ram chip is a quick and easy test, (provided of course that it holds the vid output) based on experience and available facts. To suggest reflowing the board is a reach to far given the Facts.
Machines were mice and Men were lions once upon a time, but now that it's the opposite it's twice upon a time.
MOONDOG
 

Offline phil_jp1

  • Regular Contributor
  • *
  • Posts: 103
  • Country: ua
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #16 on: November 18, 2011, 07:36:35 pm »
How about a small competition?

Who can guess correctly what the problem is before #218 is uploaded?

The winner gets special recognition for remotely diagnosing LeCroy 9384C Oscilloscopes.
Also my special bonus "Pi to the precision of the Feynman Point" for free.

After watching the video my first best guess would be corrupt display ram. The Dacs (4 off) all appear to be working as normal and the corruption is Static in the waveform and consistent across channels. This would indicate a common link, either power supply or display? More twiddleing would be required but a possible test would be to zoom into the wave form and see if the corruption stayed in the same place (on the screen) or panned with data stream. This may help point to the problem. Check the memory stick is seated/ clean seen similar problems on video cards in a pc before. :)

I don't think that there's any problem with corrupted display ram or display board at all. It's because vertical scale was all over the shop. It was outputting right input signal shape, but with the completely wrong vertical scale.
As Dave said, first of all the power supply output should be checked. It was 12 years in(presumably daily) operation. Even if caps looks ok, they can have a very substantial ESR.

Next time the scope is powered up it will probably work flawlessly. Dis- and reassembly have fixed stuff before...
Yeah!

http://JumperOne.com - Electronic projects, tutorials, hacks, etc.
 

Offline McMonster

  • Frequent Contributor
  • **
  • Posts: 417
  • Country: pl
    • McMonster's blog
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #17 on: November 18, 2011, 07:50:44 pm »

Next time the scope is powered up it will probably work flawlessly. Dis- and reassembly have fixed stuff before...
Yeah!



But beware of the Leftover Screws Axiom!

Quote from: Leftover Screws Axiom
After each complete and successful cycle of disassembly and reassembly of any sufficiently complex electronic device number of leftover screws from this process is proportional to the complexity of the device and no lower than one.

Given finitely large number of cycles of the above algorithm you run into danger of being killed by excess of screws filling your lab as given from the axiom.

There's also a positive side effect of possibility to research a new free energy source by deriving the axiom and turning the screws into energy using the famous E=mc^2 equation.
 

Offline tesla500

  • Regular Contributor
  • *
  • Posts: 145
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #18 on: November 18, 2011, 09:09:53 pm »
The dangling cable coming out of the power supply is probably for the internal printer option.

I wonder if the maximum sample memory is only available at lower sample rates. The RAM cards have 192 pin connectors, so the data bus on each card is probably 128 bit. The ram chips are rated at 66.6MHz, giving only about 1GB/sec bandwidth to each card for 2GB/sec total. Running all 4 channels at 1GSa/sec requires 4GB/sec.
 

Online don.r

  • Frequent Contributor
  • **
  • Posts: 657
  • Country: ca
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #19 on: November 19, 2011, 02:14:31 am »
Next time the scope is powered up it will probably work flawlessly. Dis- and reassembly have fixed stuff before...

Yeah. Quite common on Tek scopes with the connectors.

I love the Sagan background "soundtrack" on the video, btw. I haven't heard crying like that since I dinged my new car at a car-park.   :'(

Definitely looks like a memory reading problem with all those glitches and then the time base being screwy right at the same point. Maybe some of those push-pin connectors were loose.

Finally, hearing the 68030 being called "old school" makes me feel like a grand-dad!  :-[
Shippers: for the love of Pete and all that is holy, STOP USING UPS INTERNATIONAL!
 

Offline BubbaMc

  • Contributor
  • Posts: 10
  • Country: au
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #20 on: November 19, 2011, 04:42:50 pm »
Dis- and reassembly have fixed stuff before...

And have broken stuff  :P
 

Offline hacklordsniper

  • Frequent Contributor
  • **
  • Posts: 597
  • Country: hr
  • Don't turn it on, take it apart!
    • HackLordSniper
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #21 on: November 20, 2011, 04:22:57 am »
Im interested howmany $ did this scope set you back Dave?
Oh, the joy of sending various electronics to silicon heaven
 

Offline scopeman

  • Regular Contributor
  • *
  • Posts: 160
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #22 on: November 21, 2011, 04:19:56 am »
Hello Dave,

As an introduction, I am the moderator and owner of the independent LeCroy_Owners_Group on Yahoo! Groups. You can find a lot of information on the group site as well as overflow manual and firmware files,etc. on www.ko4bb.com in the LeCroy Manuals section.

I can tell that you like poking fun at the circuitry and design as you go through things but remember in 1999 no manufacturer had an equal to these scopes in terms of sample rate, memory depth and math measurement capabilities. If you add in all of the possible options the deep memory (2Mpt/channel simultaneous 1Gs) version of this scope went for something like 40K, maybe more. I have several LeCroy scopes from this era that I still use on a regular basis and a more modern 104MXi that I use almost daily.

It took years for HP and Agilent to finally begin to understand that deep memory is important in a digital scope, and in some cases I still don't think they get it. And how you manage that memory is even more important. LeCroy currently makes some of the real "killer" scopes on the market today. Their new 12bit one is a must see.

Maybe you should contact your local LeCroy office and see if they would agree to have you review one of the more modern ones. The math and performance is outstanding. I think their fastest scope is 45GHz at 100+GHz sample rate!

To address some things in your video.

Originally this series of scopes was manufactured in the LeCroy plant in Switzerland (93XX and 93XXA). The C model was manufactured in Chestnut Ridge NY as all of the high line series (making LeCroy the only remaining U.S. manufacturer of oscilloscopes.

These scopes were never really intended to be serviced in the field like a Tek 475A. The units were calibrated and verified by specialized calibration and troubleshooting equipment developed by LeCroy. This is not to say that you can't fix the scope, indeed I have repaired many but there are certain things that make the repair difficult.

For one thing there are many, as you have noted, custom IC's that LeCroy designed that are simply not available on the open market or from LeCroy since the units are long out of production. Specifications for the parts are also unobtainium. As you also noted there are many GALS in the scope that are of course protected so replacing those would be difficult as well. I would recommend finding a similar one one ebay for spare parts.

One particular problem with the 9384A/C is that the input in the 50 ohm mode is susceptible to ESD damage from a charged 50 ohm cable. This is a common fault. The HEF series part was a custom and the only fix is to find another scope that you can use for parts. I currently have a 93X4C scope that I junked because it had three wrecked inputs.

As for your scope problem since it is common to all channels I would suspect a data bus problem, maybe even system ram. There are 72pin SIMMS on the CPU3 MC68EC030 card that you can swap to see if the problem goes away but I don't think the ADC cal errors are caused by that but swapping them is easy. There is a note that I wrote up on the group site about the memory. Maybe just reseating the CPU card may work as these get jostled in shipping.

As an aid to trouble shooting there is a maintenance mode that you can enter from the from panel with the scope stopped. See the attached excel file for hints.

Also don't forget that we have a fairly large base of users on the group so feel free to join and post your problems there. Many of the group members have experience fixing these scopes.

While you have the unit apart I would also recommend that you measure and scope the power supplies. 

I believe that the Coax you see is for distributing timing to various places so that the sampling is truly synchronous. One of the features if these scope is the ability to combine 2+2 channels to double the sampling rate and the memory depth. By using an external PP adapter you can double that again (but you are restricted to 50 ohm measurements with the adapter).

Fortunately I have not had to work on the main acquisition board on these scopes except to fix an input relay (sometime a rap with a pencil will un-stick them) or
to make a rare calibration adjustment. If these scopes are not abused they will hole their calibration for years. Usually all that is needed is to occasionally blow out the dust and every 5 years or so replace the rechargeable lithium battery on the CPU card. There is a note on the site on the battery replacement as well.

Take care and good luck and let me know how you make out.

Sam Reaves

(Scopeman)

P.S. I do not work for LeCroy or have any official affiliation, I just like their stuff!
P.P.S - The real LeCroy scopes are the ones that are above the entry level WaveAce and WaveJet models.
P.P.P.S -- You need to add rar as one your acceptable file upload types!
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3751
  • Country: au
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #23 on: November 21, 2011, 06:32:03 am »
Sam,

Nice website! Lots of great pages there and thanks for the manuals sections. I like the GPIB stuff as I have to get some GPIB working at home somehow.

The HP1740A manual is that lousy version floating around with no schematics. Arghhh!

Do you want a copy of the service manual with the schematics - I can upload it, if you like.

Richard
 

Offline hacklordsniper

  • Frequent Contributor
  • **
  • Posts: 597
  • Country: hr
  • Don't turn it on, take it apart!
    • HackLordSniper
Re: EEVblog #217 - Lecroy 9384C Oscilloscope Teardown
« Reply #24 on: November 21, 2011, 11:03:19 am »
Lets keep the discussion about this great scope Dave has gotten then instead various personal wars
Oh, the joy of sending various electronics to silicon heaven
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf