PING Louis Rossmann, you're needed in surgery! Repairing an Agilent DSO8104A

[Jwalling]

I've got a customers DSO8104A on my bench that won't self calibrate. CH1 has a bad/blown pre-amp - it's input impedance from the attenuator is way too low (The calibration routine doesn't even detect that the coax cable is connected from CH1 to the AUX OUT connector, and "common" calibration fails). The cal log is attached.

Luckily I have a few scrap 54831B/32B acquisition boards around as they use the exact same preamp. Once again though, I don't have the needed equipment or expertise to do the repair as there's a large GND pad underneath the chip. The chip is entirely made of ceramic; whether this requires more delicate handling and heating to keep it from cracking is unknown. It looks like the "cap" of the chip is bonded to the bottom ceramic plate where the die lies; I'm hoping the cap doesn't fall off as I have no idea what material holds it together. Hopefully not glue!

I'm going to have Louis Rossmann do the repair work again like he did in this thread:
https://www.eevblog.com/forum/repair/looking-for-someone-who-can-replaceswap-tek-tds694c-trigger-chips-for-$/msg1186021/#msg1186021

Does he read this forum? Let's find out!

Pictures of the good chip on the scrap donor board, and the bad chip on the board to be repaired are attached as well.

[vze1lryy]

Sure. Might use the zm-r6200 if more precise heating is required for the ceramic. Send it in.

[Jwalling]

Excellent. Will get it shipped out to you tomorrow. Thanks!

[tooki]

[...]I'm going to have Louis Rossmann do the repair work again like he did in this thread:
https://www.eevblog.com/forum/repair/looking-for-someone-who-can-replaceswap-tek-tds694c-trigger-chips-for-$/msg1186021/#msg1186021

Does he read this forum? Let's find out!

Why on earth is this a public post? You could have sent him a PM, or an email, or called his office... nothing about this needed to be public.

[Toasty]

Because someone else might have this problem and need assistance.  Just a thought.

T

[KE5FX]

Why on earth is this a public post? You could have sent him a PM, or an email, or called his office... nothing about this needed to be public.

Much the same could be said of your post.

One difference is that the result of Louis's repair attempt will make for an interesting follow-up. 

[RGB255_0_0]

Would be cool if he live streams it 

Also would be interesting to know if it's the most expensive piece of equipment he'd have worked on - besides priceless data recovery.

[Jwalling]

[...]I'm going to have Louis Rossmann do the repair work again like he did in this thread:
https://www.eevblog.com/forum/repair/looking-for-someone-who-can-replaceswap-tek-tds694c-trigger-chips-for-$/msg1186021/#msg1186021

Does he read this forum? Let's find out!

Why on earth is this a public post? You could have sent him a PM, or an email, or called his office... nothing about this needed to be public.

This is a forum about repairing equipment.

This particular thread is about documenting a repair on a piece of equipment, specifically an Agilent DSO8104A. Of particular interest to some (not you) may be the difficulty of replacing a chip entirely made out of ceramic with an unknown bonding agent that may or may not survive the process.

With a little bit of luck, I'm hoping to see a Youtube video made of the repair process; this may interest some people (not you) as well.

With a little more luck, I'm hoping for a successful repair.

If you don't want to read this thread, then don't. I'm pretty sure no one is forcing you to read this (I'm not), but correct me if I'm wrong. Or not.

[Cyberdragon]

Louis has his own forum, but I think it's a paid subscription. He probably stays over there rather than coming here and having "4chan" find him.

EDIT: WAIT A MINUTE...is vze1lryy actually Rossman? If it is, I need to slap myself. (employees of his don't count, my point still stands)

[Rasz]

Why on earth is this a public post?

marketing

[retiredcaps]

EDIT: WAIT A MINUTE...is vze1lryy actually Rossman?

He is.

[Armadillo]

What's Up with this work? Any updates?

[bigeblis]

I've got a customers DSO8104A on my bench that won't self calibrate. CH1 has a bad/blown pre-amp - it's input impedance from the attenuator is way too low (The calibration routine doesn't even detect that the coax cable is connected from CH1 to the AUX OUT connector, and "common" calibration fails). The cal log is attached.

Luckily I have a few scrap 54831B/32B acquisition boards around as they use the exact same preamp. Once again though, I don't have the needed equipment or expertise to do the repair as there's a large GND pad underneath the chip. The chip is entirely made of ceramic; whether this requires more delicate handling and heating to keep it from cracking is unknown. It looks like the "cap" of the chip is bonded to the bottom ceramic plate where the die lies; I'm hoping the cap doesn't fall off as I have no idea what material holds it together. Hopefully not glue!

I'm going to have Louis Rossmann do the repair work again like he did in this thread:
https://www.eevblog.com/forum/repair/looking-for-someone-who-can-replaceswap-tek-tds694c-trigger-chips-for-$/msg1186021/#msg1186021

Does he read this forum? Let's find out!

Pictures of the good chip on the scrap donor board, and the bad chip on the board to be repaired are attached as well.

Sorry, English may not be very good.
I've done this work. You need a far infrared auxiliary heating platform to heat the bottom of the PCB board. I'll heat it at 180 degrees, and then use the hot air gun over the top to heat it up to 260 degrees to remove the chip.
Note: never hold the upper part of the film to move it, especially when the temperature is very high. Otherwise, the chip will be scrapped immediately. Only the bottom ceramic substrate can be clamped to move the chip.
It is impossible to use only a hot air gun to disassemble the chip, and I have damaged the two chips.

[Jwalling]

What's Up with this work? Any updates?

The boards were received on 8/21/17.
Last update on 8/30/17 is "It's still in progress. It's proving to be difficult. We will keep you posted once we have any news."

So nothing yet...
I'll probably shoot them an email this coming week.

[KE5FX]

What's Up with this work? Any updates?

The boards were received on 8/21/17.
Last update on 8/30/17 is "It's still in progress. It's proving to be difficult. We will keep you posted once we have any news."

So nothing yet...
I'll probably shoot them an email this coming week.

Watching the TDS 694C repair, I was surprised that his idea of preheating was "stick the board in an oven for a while before putting it on the workbench."  I wouldn't be surprised at all if he's discovering the limitations of that particular process.

[Jwalling]

Watching the TDS 694C repair, I was surprised that his idea of preheating was "stick the board in an oven for a while before putting it on the workbench."  I wouldn't be surprised at all if he's discovering the limitations of that particular process.

I'm hoping you're wrong!
I let him know that one of the pre-amp hybrids on the donor board was known bad (put an X on it with a sharpie pen as well), and could be used to experiment with / get a heating profile.

[Jwalling]

I just got an update.

"Apologies, that email was not intended for you.

We've replaced the chip you marked, but we don't really have a
way to test it."

Well, that's a relief! 

So it looks like it's all done. Just have to pay and hopefully it'll be back next week for further testing/repair.
Disappointed that there's no video.

[Jwalling]

Update.
Got the board back today. It now will complete a calibration (no more cannot detect cable so you can't proceed) But it fails CH1 trigger and vertical. Cannot get any kind of a decent trace on the screen...

Further testing shows that there's a solder short on pin 20 to the large GND pad, see pictures.
Checked it in four wire REL mode on my Keithley 2000, it's 60uOhms, so it's definitely between the pin 20 pad and the GND plane (no way can it be the chip or at the other end of the trace).
The pad can be seen on the donor board.
It looks like it's going to be heading back to NYC

Question: Will performing the operation twice increase the likelihood of killing the chip? In other words are the heating effects cumulative?

[PA0PBZ]

Looks like you also have some excess solder between 20 and 21? What about sucking the pin(s) clean with litz wire? With a bit of luck the short will be removed I'd guess.

[TheSteve]

Preheat with a heat gun and use some really good solder wick possibly?

[PA0PBZ]

Preheat with a heat gun and use some really good solder wick possibly?

Solder wick, that was the word I was looking for

Also I was thinking if you suck it clean and then still have the short put a piece of kapton tape in between and a wire from the pin to the track...

[Jwalling]

Looks like you also have some excess solder between 20 and 21? What about sucking the pin(s) clean with litz wire? With a bit of luck the short will be removed I'd guess.

I just measured pin 21 to GND and it's 1MOhm. Good spot though, I couldn't see it with my optical visors and never thought to look closer at the picture. A little flick with the end of an x-acto blade removed it. Not entirely sure it wasn't flux; there quite a bit of it remaining even after cleaning with some IPA.

[wraper]

Preheat with a heat gun and use some really good solder wick possibly?

I'd say preheat the board if possible (borrow hot air from your wife), apply decent amount of tacky (gel) flux and solder the pad. Short will likely go away even without a solder wick.

[PA0PBZ]

Looks like you also have some excess solder between 20 and 21? What about sucking the pin(s) clean with litz wire? With a bit of luck the short will be removed I'd guess.

I just measured pin 21 to GND and it's 1MOhm. Good spot though, I couldn't see it with my optical visors and never thought to look closer at the picture. A little flick with the end of an x-acto blade removed it. Not entirely sure it wasn't flux; there quite a bit of it remaining even after cleaning with some IPA.

So one non-existing problem solved If I compare the amount of solder with the IC above it he didn't really hold back...

Pins 30 and 31 are also looking very suspicious on the picture.

[Jwalling]

Preheat with a heat gun and use some really good solder wick possibly?

Another good idea! I've got a nice tiny nozzle for my hot air station. Thanks Steve, I'll give that a try tomorrow.