Noob with LeCroy 9400A DSO

[Seg]

So, looking for my $50 scope I ran across this:

http://www.ebay.com/itm/LeCroy-9400A-Dual-175Mhz-Oscilloscope-/310726635524?&orig_cvip=true

A LeCroy 9400A 175mhz DSO from 1988 with a bad channel. I figured what the hell, it has at least one working channel and Dave makes fixing anything look easy, so I threw down the minimum bid, fully expecting the opening bidder to come back and outbid me, and promptly forgot about it. To my surprise, I woke up one morning to my wife telling me I won an oscilloscope. Paid $50.99 + $49.94 shipping.

A week or so later it shows up on my doorstep. I plug it in on the floor, it makes some crackly noises, eventually the CRT fires up and for a few glorious moments advertizes that it does in fact have the FFT option, then fizzles out.



I pick it up and put it on the table, power it up, and it runs long enough for me to fiddle with the knobs for a few minutes. Eventually I turn it off, and turn it back on. It crackles and fizzles and hums, but the screen remains black. All the crackly fizzly noises make me suspect the CRT, but I notice the front panel is not responding to input, so there must be a problem with the embedded CPU. Having dealt with shipping computers before, I open it up and re-seat all the cards and cables. No luck. No amount of fiddling with connectors, percussive maintenance, and flicking it on and off will bring it back.

I really need to get at the CPU board, but it's impossible to get at from the top. I open up the bottom, and start removing the board, but have trouble with the corner with the BNCs, eventually finding myself half tearing out one of the HVV200 amplifier chips because they're stuck to a strange heatsink arrangement that bolts to the board and the case. (Fortunately they turn out to be in low-profile sockets, ) I decide to set it aside, before I destroy it further, and do some research.

After finally downloading and reading the service manual, (the "kurt" site seems to be down most of the time.) I manage to finish removing the motherboard:



There's what I'm looking for. Do we have any MC68000 fans in the house? O0



Looks like legit FFT roms.

Apparently you're not really meant to remove the motherboard, rather you remove everything else from the unit first. The service manual also reveals that if all the cards are not plugged in, it will refuse to boot up. There's not much more diagnosis my noobness can do with it in pieces, without another oscilloscope and/or LeCroy's test fixtures. I air-dusted it all out, went over all the boards re-seating every socketed chip I could find and did not see any obvious problems. With a shrug I put it all back together and what do you know, it boots up again!



That's one bad-ass piece of late-80's technology.  Eats 200W powered up.

So far it has been booting up reliably since then. The crackly noises seem to be coming from one of the power supplies. Bad caps? Going to have to take a look at that eventually.

Now I need some probes. It was originally supplied with 350mhz probes, but that seems to be the point where even the cheapie probes start getting pricey. I ordered a set of 200mhz probes, that ought to be good enough for anything I'm likely to use this thing for: http://www.ebay.com/itm/320847586124

Eventually I need to take a shot at fixing the first channel, but for now I'm just happy to have it running at all. I'm not looking forward to taking it all apart again. The problem must be in the front-end circuitry and that's the hardest part to get to. Channel 1 works at 50ohms, but at 1Mohm in AC or DC it just displays double arrows on the bottom of the screen, indicating the trace is offscreen and no amount of adjustment will bring it on screen. Set to either ground mode you get a seemingly normal flat trace.



Some random related links:

http://japonoff.blogspot.com/2012/08/disassembling-lecroy-9400-125mhz.html

https://www.eevblog.com/forum/testgear/lecroy-9400a-options/

http://www.antiqueradios.com/forums/viewtopic.php?f=8&t=158426

http://www.ebay.com/itm/Maytek-Instruments-Power-Supply-Extender-Cable-for-LeCroy-9400-9400A/290976407191

[israel]

Mmmm ... a nude very sexy

[marshallh]

Careful running it without heatsinks and cooling... the frontend hybrids will nuke themselves.
Look up the input stage circuitry. In 93xx the 1Mohm path goes through a high impedance buffer. Check this first. Then it merges with the 50ohm path to the S&H fanout buffer.

[AlfBaz]

I find it interesting that you have those set of coaxes going off to 2 similar if not exact same boards. I'd be tempted to swap them to see if the problem transfers to the other channel. If it does it means the problem isn't the front end on the mother board but on one of the "daughter" boards

[Sigmoid]

Don't forget that bad power can really muck up the operation of otherwise perfect digital circuits. Before suspecting the CPU, suspect the PSU!

[Seg]

Well I got my probes and... it's dead again.

Looks like the +15v power supply quit, the LED on the back went out. But the -15v supply is the one making the arcing noises. So now I have two power supplies to fix...

Two blown fuses inside the +15v supply. Hmmm.

Turns out the arcing in the -15v supply was one of the fuse contacts. It and the fuse were all corroded where they touch. Filed off the corrosion and now it doesn't crackle. So I guess that one is fixed...

[Seg]

Okay, $4 worth of overpriced Rat Shack fuses later, I determine the fault is within the power supply. It's an OEM unit from ELBA Electric GMBH, made in West Germany. Are these units some kind of standard? If so I can't find the name of whatever standard it is.



Finding a replacement doesn't seem to be likely, so I guess I have to fix it...



Caps all look fine, I don't see anything obviously fried. Now what? I googled up this for some guidance: http://www.mikesarcade.com/cgi-bin/spies.cgi?action=url&type=info&page=psrepair.txt

Output rectifier looks okay, MOSFET looks okay.... Bingo! Found a shorted diode in the input rectifier.



Yes, that's how it came. One sloppy soldering job later, it lives again!

Now maybe I can get back to fixing the 465B...

[scopeman]

Hello Seg,

You might want to consider joining my independent LeCroy_Owners_Group on Yahoo Groups!.

You can find a lot of information there plus there are many users of the 9400 series scopes on the group.

Many manuals can be found there and on the www.ko4bb.com site in the LeCroy section.

Sam
Moderator and Owner - LeCroy_Owners_Group

[Seg]

And... it's dead again. 

The +5v supply this time. What a surprise, it's the input rectifier again.

Yes, I did find my way onto the yahoo group. I think it's how I ultimately found the service manual.

[poorchava]

I suppose old scopes, no matter Tek, HP or LeCroy have a thing with bad bridge rectifiers...

I you do not have equipment to measure ESR and inductances, I'd replace ALL of the capacitors and bridge rectifiers. They are not very expensive, especially that you live in US.

[PA4TIM]

Rectifiers go short if the voltage becomes to high or it gets to much current. To high a voltage is not likely because there are mainly two choises, 110 and 230V and in the states you have 110 (we have 230 and if we buy something from the states we first have to set it at 230)
To much current can be caused by dead caps. And as Sigmoid wrote, first be sure the powersupply is 100%. So all voltage exact (within the specs of the manual)  what they should be. And ripple also within specs (you will need a scope for that)
If you do not have the gear (a very accurate Voltmeter and a scope (a 10 MHz will  do)) just replace all psu caps.

[Seg]

So, this time we have a Schroff, with a repair sticker:



Judging from the flux on the back, this isn't the first time the rectifier has been replaced. Originally a cylindrical button thing, replaced with some 1N4005H:



And it lives once again:



Also a bunch of capacitors destined for the 465B...

[GK]

Good job, but whatever you do, don't blowup a vertical channel as those hybrid input modules are unobtainable. A colleague at work blew up CH1 on ours by directly probing a 1400V neutron tube supply. The hybrid module went poof and I couldn't fix it. I absolutely hated that bloody scope anyway, due to the way the traces would shoot right off screen if you so much as breathed on the vertical positioning knobs.
 
EDIT: BTW- under those vertical input shields you will find some protection circuitry which includes low leakage-type clamping diodes (part# PAD5, IIRC) from the rails to the hybrid module input. It's possible that one of those diodes has been zapped and is now excessively leaky, causing your trace to deflect off screen when terminated by 1M rather than 50 ohms (any leakage current would cause a 1000000/50 times larger DC offset error in the 1M mode).