GPZ7000 repair?

[sinclairuser]

i noticed someone mention the white paint the ML used before, soda blasting removes that, there was a video on YT about it.
newer models don't use it but for repairing older stuff you might find it useful.

[ivconic]

i noticed someone mention the white paint the ML used before, soda blasting removes that, there was a video on YT about it.
newer models don't use it but for repairing older stuff you might find it useful.

Once upon a time I serviced some Minelab, SD series, if I'm not mistaken.
I had to replace the IRF transistor and a couple of semiconductors.
I removed the white paint from that part of the surface; by first coating it several times with nitro thinner, it softened it,
then I gently removed it with a scalpel piece by piece.
But here we have a drastically different case. Here the pcb and components are well submerged and fused in thick layers of epoxy compound.
And this is about "cured" epoxy. Hard nut!
No amount of "freezing" and "overheating" will help. It is a matter of thick layers and not a thin coating.
Dichloromethane is the only solution. Have no doubt about it.
The module must be completely immersed in dichloromethane, soaked and left to stand for hours.
And then check with a sharp object how much it has softened.
It won't soften all the way, that's for sure. It will soften layer by layer.
That's why it has to be removed from the dichloromethane every few hours, rinsed with cold water (to remove the dichloromethane) and
then removed part by part of the softened epoxy with a sharp blade.
And so repeat until the epoxy is completely removed. It's the only way. There is no other.
So... patience and care when handling dichloromethane.
Success is guaranteed if these tips are followed.
It's no use talking about any other methods, there's nothing to it, it's all empty talk.
This time there is only one way to skin the cat...
You can take my advice and succeed... or you can not take my advice and screw up the module.
There is no third option.
In short; my way or the highway!
I am so tired...
I will retire for a certain time....

[gef12]

Surely if using this chemical method and if you succeed .. will part numbers if left on components be dissolved also ..
otherwise interesting subject ..

[bklein]

I guess I’ll find out. Most fets/diodes will have like a 3 letter code I’ll have to try to identify.

[bklein]

I did just receive the solvent. So I’ll put it in a jar with the board - but what should I expect? Will the epoxy just soften and I will have to pick off pieces or does it come off the board on its own? Wondering how I know when it’s done all it will do.

[ivconic]

Surely if using this chemical method and if you succeed .. will part numbers if left on components be dissolved also ..
otherwise interesting subject ..

Not gonna happen if monitored during the process and stop it when the epoxy gets soft and easy to remove.

[ivconic]

I did just receive the solvent. So I?ll put it in a jar with the board - but what should I expect? Will the epoxy just soften and I will have to pick off pieces or does it come off the board on its own? Wondering how I know when it?s done all it will do.

I know my posts are sometimes too long and tedious to read, but read the first post on the first page again.
I explained that it was a "monitoring" process over several hours.
With a needle or a sharp blade, occasionally (every 20 minutes to half an hour) you check how far the process has come, by poking the mass.
When you notice that the needle enters the mass very easily, or you can easily separate part of the mass with a knife;
then it's time to take the module out of the container and transfer it to another container with water.
Then with a knife, a scalpel, a sharp chisel; you remove all the softened mass that can be removed, but nothing by force!
When you can no longer remove layers of mass so easily; put the module back in the dichloromethane container and let it stand for the next half hour.
And then repeat the same again. And so you repeat the same thing until you remove all the epoxy from the module. It usually happens in a couple of attempts.
So, at the start you have two containers, one with dichloromethane and the other with water. And they remain there "in the game" until the end of the job.
The first "poking" check is done after 20 minutes, and the next one is done even longer, if you notice that the process is running slower.
And vice versa, if the process goes faster, then you do the checks in shorter intervals.
Epoxy will not fall off on its own even at its softest. You have to do it with a blade or similar tool.
The epoxy will turn into a gelatinous mass that easily separates from the module.

[sinclairuser]

i don't find your posts too long ivica, since it is always relevant to the topic at hand the length of the text is neither here nor there.
as for this task the only way to shift it is with chemicals as you suggest, i only mentioned soda blasting just in case some numpty thought thought about using dichloromethane on the white paint.

[bklein]

12 hour report: the areas that became brittle from heat gun 150 degrees C and later scraped exhibited some of the brittle material coming off the board and settling on the container bottom. The areas of the board not previously heated show no change with the soaking - still a hard surface. Bummer.

[ivconic]

12 hour report: the areas that became brittle from heat gun 150 degrees C and later scraped exhibited some of the brittle material coming off the board and settling on the container bottom. The areas of the board not previously heated show no change with the soaking - still a hard surface. Bummer.

Two possible explanations;
either it is "diluted" and "weakened" dichloromethane, has no "strength" to process faster,
either the epoxy is really hard "cured" and will need much more time to soak.
In both cases now we know it will not go fast, so no danger to keep it soaking over the night, maybe few days.
With periodically checking and testing it. By poking it with the needle.
You have nothing to lose, you must be patient. No need to rush.
Even if it takes 10 days, so what? Finally you will do the job.

[ivconic]

i don't find your posts too long ivica, since it is always relevant to the topic at hand the length of the text is neither here nor there.
as for this task the only way to shift it is with chemicals as you suggest, i only mentioned soda blasting just in case some numpty thought thought about using dichloromethane on the white paint.

[bklein]

So it?s been a few days and I decided to check it. Again some of the areas that were affected before would freely come off a bit more but nothing major. I decided to try heating it with the heat gun some more. The epoxy would pop off like popcorn! But it took a part with it - looks like a diode perhaps. I?ll have to get my microscope out to see if I see any markings. It?s from the area that got hot with power. I checked with my meter and it is shorted. So if a diode, how does the replacement go? The pad inside does ohm out to several of the pins on the sides. The outer pad doesn?t ohm out to any of the pins. 50-50 shot at it I guess!

[brs]

Perhaps gasoline, soaking it in it for a whole night, will help

[bklein]

Nah, I think it’s time to give up. I tried a replacement diode both directions. One at it was very poor sensitivity, the other way none at all. I need a replacement black RX module. Maybe if you hear of one of these getting caught in a fire?

[sinclairuser]

is there anyway you can get this block xrayed?, its a long shot but you have low sensitivity so its now working at least, an xray of it might just show which other parts need changing.
if another part looks distorted you can just focus on that, taking just enough epoxy to get to that bit, but my guess would be a cap/caps connected to the diode you changed.
if it was a cap the circuit would work but not properly, most other parts would stop the circuit from running, if you know what i mean.