How reliable is Wire Glue?

[Mooger5]

First, AFAIK Wire Glue isn´t available here yet, I´ll have to import it and I could do without the waiting.

Now this 80017a lost the solder pads for the mounting pins, but I´d like to save it. Looking closer there are vestiges of solderable surface, still in place, enough to carry the signals. I just need to find something that will keep the pins in place.

I have tried with silver and nickel inks, but they´re not solderable.

Could that glue be the cure? There´s got to be a pun there. How reliable is it? Will it hold the pins in place?

If anyone has a better idea, share it please. Thanks!


free image hosting


image uploader

[Rasputin]

I would be more inclined to scrape back the traces and use a more durable (non-conducting) fixative to glue on the pins and then solder the enlarged traces to the pins separately.

Well, I can't exactly (easily) follow the layout based on the pics, but my thought is to first solder some thin kynar to the pins that have no stable pad then tack down the pins with... hot glue? Then solder the kynar to whatever component the corresponding trace leads to or scrape down the trace further up and solder directly to that.

I totally know I don't need to break that idea down for you on a personal level, but just providing clarity for anyone else that might stumble upon this thread, so don't be offended by me stating the obvious.

In my admittedly limited experience, I could never get glue like that to do anything useful, but perhaps I've approached it improperly, had a bad brand/batch, or the like.

[madtheory]

This is probably better:
http://www.maplin.co.uk/p/electrically- ... aint-n36ba

[Mooger5]

Thanks for the replies.
Rasputin, of course no offense taken. I appreciate the detailed reply. I will consider scraping the traces. The problem is the multiple layers. If I manage to scrape one trace without shorting the other beneath it, maybe then epoxy will bond better than hot glue?

madtheory, I´ve tried that before, I called it "ink" in my previous post, sorry... It´s not solderable.

One alternative to this Wire Glue could be epoxy glue little squares of copper foil to the substrate and solder the copper to the traces. Afterwards the pins would solder to the newly formed copper pads easily.
It´s easier said than done. The areas are minuscule.

They call this wire glue "cold soldering" but I have doubts once cured it will be as mechanically robust...

[madtheory]

I think your instinct is correct. The wire glue is not mechanically reliable. I think your epoxy plus copper foil is probably the best thing, short of actually replacing the IC

[Mooger5]

Yeah, an additional problem to applying the copper foil is the epoxy not withstanding the heat of the soldering. Araldite holds the line up to 80ºC. Too low.

Replacing the IC is out of the equation; It´s my last one and I´ve already spent too much on this synth. I´ve learned a lot in the process though. Unless someone sends me an out of spec IC for free or something.

[Rasputin]

I will consider scraping the traces. The problem is the multiple layers. If I manage to scrape one trace without shorting the other beneath it, maybe then epoxy will bond better than hot glue?

One alternative to this Wire Glue could be epoxy glue little squares of copper foil to the substrate and solder the copper to the traces. Afterwards the pins would solder to the newly formed copper pads easily.
It´s easier said than done. The areas are minuscule.

Absolutely. I believe the epoxy/pad method is really the only practical "proper" way to approach it, and yes, epoxy > hot glue, however in my experience hot glue is semi-permanent enough to get the job done but reversible enough to change your plan of attack should you have to regroup.

Not an envious position to be in, at any rate but I'm voting for you to persevere. You can't let those electrons get the best of you!

[madtheory]

Specs on this seem to be saying it's good up to 250C:
http://docs-europe.electrocomponents.co ... 57e335.pdf

Have some here for a SMPSU repair. I can test it for you with solder if you like?

Not an envious position to be in, at any rate but I'm voting for you to persevere. You can't let those electrons get the best of you!

Yesss! Seconded. Good luck with it. REpairing rocks

[Mooger5]

LOL at the electrons. Those mean creatures can be negative at times but they´re so useful Vive la resistance! Or depending on the situation: resistance is futile! Charge! More power to them eh

Well I´ll see what I can do. Cheers!

[Rasputin]

To anyone up against something similar:

While a repair like this might make a lot of people inclined to throw up their hands and say, "impossible!", I think it actually has a reasonable degree of success for a "garage bench" enthusiast.

The "proper" method we've been discussing is something along the lines of cutting out a pad sized piece of copper foil and then using an appropriate heat-resistant epoxy like CircuitWorks CW2500 to tack the foil to the PCB. You'd also want some kapton tape to hold down the fixed pad as the epoxy cures and/or assisting in pad placement and/or masking select areas.

If there's a usable amount of trace left so that the cut out replacement pad will sit over it then that's terrific, if not then use a fiberglass pen (or x-acto knife) to scrape away the coating on the nearby connecting trace so there's enough overlap for the new pad to touch somewhere conductive.

As far as dealing with the tiny scale of the repair, well, that's the trick, right? A microscope, tweezers, and maybe a dental pick will be helpful.

[Mooger5]

Well you´re not gonna believe this. I have just found two more chips, and one is totally refurbished ready for use. My workbench is full of junk from unfinished projects. Damn it I can´t count for s**t.

[Mooger5]

And to prove I´m not chickening out, as you say there, here they are

Rasputin that resumed it completely. I´ve got all of those except the CircuitWorks epoxy.
What can I add? That keeping the fingers off the sticky liquid flux when cleaning the substrate helps a lot! Seriously, I had just finished soldering the ICs back when I found the other 80017As. I was about to cut pieces of copper foil to shape and they just wouldn´t leave my ingers. Rule nr1 keep your working area and yourself thoroughly clean! Not Cleaning Room clean, alright, just enough to handle the tiny, tiny components without a hassle.

I´ll be keeping you guys updated. Of course motivation is not the same (or should I say desperation) since I found that good chip, but I´m not giving up on the others.

Cheers!

[madtheory]

Glad it's working out. LOL I have glued stuff to my hand a couple of times too!

It is a whole other challenge to get stuff organised. Those little plastic storage cabinets and a label maker can get quite expensive too! Here is the guy in Ireland who is the best. He is incredibly well organised in his little shed:
https://www.irishtimes.com/culture/musi ... -1.3076599

[belltones]

I'm anxious to see if this pans out. As I read the earlier stages of this thread I was thinking there was no way this would work. Also, that I suspected that most people replying here were not familiar with the construction of the 80017A because its traces are not made the same way as even a normal SMD board. As the OP probably knows, the traces are most similar to a thin layer of powdered graphite having been pressed onto regular porcelain and then sealed, and the insulation between layers is like some kind of inky paint that I've always been surprised can even function as an insulator-- very different from a solid metal trace coated with scrape-able enamel.
Out of curiosity, I once tried to use several half-failed 80017As to create some new ones, which put me repeatedly in this situation, and eventually I just gave up, deciding that it would always be easier to use the clones (as I use them constantly, I buy them wholesale so the price for me is lower than it is for someone buying just one). But I'm definitely curious to see if this will work. I'll be following this thread!

[Mooger5]

Well the two chips are busted: One broke apart the minute I started to work on it. The other had lost the solderable area for two or three pins. Lose just one of these and the entire chip is wasted.

https://ibb.co/jPDMsF

There´s still hope as the ICs are intact and I could use them and the caps plus the required SMD resistors to make clones on veroboard. The 0603 size seems to fit, the holes will hold L shaped connetors like these providing the mechanical robustness the 80017A have always lacked

and with everything wired with solderable magnet wire or wirewrap, these clones should outlast the synth itself.

One problem is the resistors values and the pin numbers. On the 80017A diagram provided by Roland these pin numbers are only right for the SIL versions of the BA662. The SOIC pins however dont match. It takes a bit of detective work to find them. I did it years ago and noted everything on a piece of paper somewhere. I´ll probably find it after all the work is done

There is also the SSI2144 IC. I´ve been toying with the idea of building "clones" with these wonderfully sounding filters again.