Jump to content

Juno 106 chip revival


Recommended Posts

  • Replies 132
  • Created
  • Last Reply
  • Members

Thanks guys! :D Flat, they *are* soldered in, not socketed, although I could install strip sockets and just plug the modules in. I'd worry about them working themselves loose again, though.

I placed an order for more desoldering braid last night so I can pull all the rest of the modules and denude them. :D

Link to comment
Share on other sites

  • Members
great photos!


where do you live that would require
ordering
desoldering braid? living in the silicon valley we can generally get desoldering braid at 7-11 and most laundromats.



:lol:

There's probably some electronic supply houses around here (Orange County), but because a) I'm usually too preoccupied to go run errands, and b) I'm lazy, I'd rather just place an order with Mouser and wait for the UPS truck. They have a warehouse close by, so I'll probably have my stuff in hand by tomorrow. Besides, I needed some other stuff, too... I found a schematic for Buchla's sequential voltage source that I want to breadboard. :D

Link to comment
Share on other sites

  • Members

RadioShack has desoldering braid...

 

Nice job on de-resining your bum modules. I wonder if those "black blobs" you refer to are spots where the two circuit layers are connected together.

 

Maybe as a test you could try poking different parts of the board with a toothpick or other non-conductive pointy thing and see if you can make it crackle. It might help isolate some flaky connection that the resin was making worse. I almost wonder if the resin was reacting with the printed resistors or those junctions somehow.

Link to comment
Share on other sites

  • Members

Thanks guys!
:D
Flat, they *are* soldered in, not socketed, although I could install strip sockets and just plug the modules in. I'd worry about them working themselves loose again, though.


 

Doh! :facepalm:

 

Pic #2 looked like the modules just slotted in. Your de-soldering skills are just too good :D

Link to comment
Share on other sites

  • Members

Great, great stuff EP!!

 

Are these chips also covered in resin in the MKS-30? I hardly ever hear about MKS-30s failing, do they run at lower temps maybe? If anything, they should run hotter since the enclosure's much smaller...

 

Now just imagine how many of those chips went to the trash can because people thought they didn't work anymore :cry::cry::cry:

Link to comment
Share on other sites

  • Members
Alrighty, some pics and commentary!








Here it is before I gave it a detailed going-over. Notice how the component

side resin came off as almost a complete sheet; that made life easier. The

back side, though, was being more stubborn.


revival-5.jpg



And now the shot you've all been waiting for, at high res. Notice the cool

multilayer circuit printing on the ceramic. I see a bottom trace layer, an

insulating layer, another trace layer, some printed resistors (
the variously

sized dark gray rectangles
) and the solder mask. Pretty neat stuff. There's

some little black "blobs" here and there, mainly near the resistors; I don't

know what they are, and they haven't softened like the resin. I'm leaving

them there.


revival-8.jpg



On the back side, the resin is still be stubborn. There's no circuitry on the

back, so I'll just leave that resin be.


revival-9.jpg



Here's the module reinstalled and ready to go back into the Juno. Sorry,

Mister Blurry Cam stole the camera from me and took this, not my fault:


revival-10.jpg




Results:


On power up, no magic smoke appeared. I count that as a Big Plus! For

extra bonus points, though, the voice worked again! No crackling, no

popping, no weirdness of any kind! YAY!! I let it run for 20 minutes to get

good and warm and it was solid the whole time.


However, by then voice 4 had had enough and gave up entirely. So, once I

get some more solder wick (I'm out), I'll yank that module and give it the

same treatment!


Questions?



those dark gray square guys printed on the PCB Those are calibrateable sand resistors. They use them in automotive applications to trim the wire heaters on Mass Air Flow sensors to calibrate them. Long story how I know that BTW.

Anyways. Thats why the chips were epoxied. To keep the sand resistors from picking up moisture and swelling. It can drastically change the impednece which could lead to issue like drifiting pull up and down values which ultimately could lead to oscilations in the circut.


Anywyas. I would properly treat those board with a conformal silicone non conductive couting. The ceramic can wick moisture as well.

BTW that fact the acetone disolved that epoxy is proff that the epoxy had thermally fiatuged and cuase the moisture issue which leads to the problem. Being as acetone is a water dispersant soaking the chips in water drys them out and whalla they work.

Oh BTW that looks totally like Hitachi electronics parts and board design. I'd bet you could find the block diagrams for those chips in the obsoleteed and archived hitachi or OKI semiconducter archives.

Link to comment
Share on other sites

  • Members

Excellent info, AluminumNeck, thanks!

Honestly, I'm not comfortable leaving the modules exposed like that, but I didn't know what would be a safe material to protect them with. I'll do some more research for conformal coatings, thanks for the insight.

Link to comment
Share on other sites

  • Members

I just did the acetone soak and for two days so all the coating came off both sides. What was that thing about patience a virtue? anyway i was scraping off some in around the chip pins and the chip just popped off. like it had cold solder joints. could it be the coating is effecting the solder? also i noticed that the coating from the back was a different color slightly then the front. it was more brown instead of black on the side facing the components.
could it be the coating is reacting to the solder or other metals / chemicals to either weaken connections or actually create some resistance in the coating that would effect the operation. after all FET inputs are high impedance and vulnerable to slight variations that could occur.
i experienced this with Bang and Olfson boards that used a glue to hold down wires.
what do you guys think?

Link to comment
Share on other sites

  • Members

It's possible for a chemical reaction to take place, and given time, use, heat, humidity, etc.. you might get the right mix. It could also be the chip was not soldered on well or the board had a defect. Stuff breaks down unfortunately.

 

I think it's easy to understand from a engineering perspective it's engineered for a specification and purpose. Sometimes the bottom line of a product or company outweighs the lifespan past a little over warranty on a product. It could be that those assembly's were subbed out and through the manufacturing process, that company wanted to make a extra 10% and changed the epoxy throughout the run past qualifications with Roland.

 

As all these synths are all old, and there is a uncertainty of which ones have the problem it's assumed as all. Furthermore, Roland isn't probably getting them in for service etc.. so no one is keeping track. What is apparent is that something in the manufacturing process along with the materials used overtime causes the packaged unit to not work. Rather that is failure of the chips or if it's a issue in the epoxy coating that would have to be determined on a case by case basis and shouldn't be assumed that it's only the epoxy.

 

I would say a good chance of a voice don't work, you have a chip/solder failure. If you have drifty voices, then there is probably a chance that the epoxy is playing voodoo with your synths chips.

 

I'd try to re-solder the chip back in provided the pads did not come off with it.

 

Who was it that was making the good quality replacements? My 106 is still trucking just fine with original assembly's but as it seems you just don't know.

Link to comment
Share on other sites

  • Members

Be real, real careful when picking out the remaining goo between the pins on the chips. There's no mechanical connection keeping the components on the board except for the adhesion of the printed traces on the ceramic. If you pry too hard you're liable to to just pop something off. These things weren't meant to be messed with.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...