0
\$\begingroup\$

I'm not sure if this is the place to ask or it would be better suited for retrocomputing.

I'm salvaging some old components from old boards and I came across a very hard blue thermal compound that a heatsink was mounted with.

blue resin on an IC

I'm wondering how can I remove it. I tried cleaning it up with isopropyl alcohol and acetone, but it didn't seem to work. I'd rather not use physical force as the ICs it's on are mostly ceramic and I don't want to break them.

What is this? Can I dissolve it somehow? What would be the best way of removing it?

\$\endgroup\$
3
  • \$\begingroup\$ I'd use a dremel. \$\endgroup\$
    – Andy aka
    Commented Mar 30 at 12:41
  • \$\begingroup\$ I'd suspect this is a modified 8255 device. Pinout seem about right. Ceramic seems about right. Could be totally wrong. RFNA is always a good choice for dissolving crap like that. Nasty stuff, but it works like magic. \$\endgroup\$ Commented Mar 30 at 12:55
  • \$\begingroup\$ @periblepsis It's Intel C82586, the 10Mb Ethernet LAN coprocessor. But this is only an example where the epoxy is easily visible because the heatsink itself is missing. I have loads of different chips with this stuff on. \$\endgroup\$
    – daneos
    Commented Mar 30 at 13:04

2 Answers 2

2
\$\begingroup\$

Given that you have a ceramic IC, excessive force, heating, and acids are out of the question. That leaves you with organic solvents.

According to this source, dichloromethane is most effective. If you use DCM, apply it only locally to the epoxy and don't let it come into contact with the rest of the IC. Don't let it run off the sides of the IC as it may seep into it through tiny cracks in its seals. DCM will also dissolve the markings as well. Alternatively, toluene or methyl-ethyl-ketone might work.

Note that dichloromethane is quite harmful, only do this outside.

In general, read the safety datasheets for any chemicals you want to use before ordering / using them. Always wear proper protective equipment, which in this case is at least a pair of well-sealing safety goggles. While it might seem counter-intuitive, do not use gloves (at least not random supermarket ones). DCM will attack most glove materials, which makes a splash onto your hands worse than if you just didn't wear any.

You should of course avoid getting solvents onto your skin, so I'd recommend applying small drops of solvent to the epoxy with a pipette and then just letting it sit.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks for the ideas and all the safety tips. I'd definitely fall into trap of doing this in gloves. I'll try to get some DCM and see how it goes. \$\endgroup\$
    – daneos
    Commented Mar 30 at 13:14
0
\$\begingroup\$

If it is epoxy, your options are more or less methylene chloride (and other harsh and toxic solvents), or heat and scraping. Solvents will only swell and soften epoxy, making it easier to remove; it's a cross-linked substance so it can't be dissolved into a liquid form. It's also fairly inert, so needs quite aggressive chemicals to decompose it (usually, fuming nitric acid; which would most likely dissolve metal parts of your package).

The ceramic package at least should handle heat well, be immune to solvents, and be resistant to scraping. Be careful around the seal and metallization, which can be damaged by metal tools, and depending on which type of solder holds the seal in place, it can melt.

Use only enough heat to soften the material; I doubt going over 200°C would be recommended.

\$\endgroup\$
1
  • \$\begingroup\$ Most epoxies go chalky at around 100C and can be scraped fairly easily. \$\endgroup\$
    – Kartman
    Commented Mar 31 at 5:20

Not the answer you're looking for? Browse other questions tagged or ask your own question.