I just did the repair on my two 3S-GTE ECUs today. Took me a bit longer without a solder sucker but overall wasn’t too bad. I replaced with Panasonic FC series — these capacitors are just bulk decoupling caps for the 12V rail inside the ECU and the regulated supplies for the MCUs/etc, so I wasn’t too worried about ESR. I dug up the original datasheets for all of the stock capacitors and verified the new caps met or exceeded performance of the original parts.

For a 3S-GTE (I’ll look up the part numbers of ECUs later here, but from a 1995 SW20 MR2 and a 1997 ST205 GT-Four).

• C105
  • Original: 220 uF, 10 V, Nichicon KM.
  • New: 270 uF, 16 V, Panasonic EEU-FC1C221
• C002
  • Original: 47 uF, 63 V, UCC SXE
  • New: 100 uF, 63 V, Panasonic EEU-FC1J101B
• C004
  • Original: 33 uF, 25 V, Panasonic NHE (aka GE)
  • New: 56 uF, 35 V, Panasonic EEU-FC1V560
• C101, C894
  • Original: 100 uF, 10 V, Panasonic NHE (aka GE series)
  • New: 100 uF, 16 V, Panasonic EEU-FC1C101
• C816
  • Original: 10 uF, 50 V, Nichicon PF
  • New: 15 uF, 50 V, Panasonic EEU-FC1H150
• C823
  • Original: 15 uF, 35 V, Panasonic NHE (aka GE series)
  • New: 27 uF, 35 V, Panasonic EEA-FC1V270
• C895
  • Original: 10 uF, 50 V, Nichicon PF
  • New: 15 uF, 50 V, Panasonic EEU-FC1H150

Minor diffs in lead diameter and height, but no fitment issues observed. All should be available Digi-Key, Mouser, etc.

 

I didn't know what era of 3S-GTE has capacitor problems.

 

I think any Denso ECU from the 90s could be affected as similar core design / similar vintage of caps that were QATS based. S, JZ, UZ, etc ECUs seem to all be the same basic design, differing only in numbers of cylinders / etc and then the ROMs with specific fuel maps / logic.

I have a JZZ30 as well that I need to examine. Most of these are so low mileage that I’d guess the relatively low time spent biased may have prevented any leakage from occurring. One or two of the caps did spill as I desoldered, I blame the heat more as I wasn’t gentle, but perhaps they were close to it.

 

Caps shouldn't puke their guts just from soldering. I don't attempt to solder suck the board I cut the leads off then pull them out, after that clean up with MG Chemicals solder wick. Safest method at least for me.

 

That's what I tried to do, but had a few holes where I couldn't quite wick all the solder out. Judging from some of the caps, I don't think they thermally relieved the. negative/GND pads from the planes / polygons, which made things annoying.

Brought the boards into work today to scrub / wash off the flux so I'll take some close-out pics and post them here.

 

[QUOTE=Yamae;11795784]

1, LScowboyLS had gone about a decade ago and never came back.

I wanted to reach out on here to let people know I've made contact with him via Facebook (Japanese ECU ECM PCU Repair by Lscowboyls) and had my ECU repaired successfully in my 97 Celica just recently. Response time was really good and turn around on the repair upon receiving ECU was good.

 

Reporting another repaired ECU. I had Tanin Auto Electronix do the repair (along with the gauge cluster). I have opened up a couple of laptops/phones but am not doing that and/or soldering consistently enough to be confident in doing the repair correctly.

I wasn't experiencing any ECU-specific problems, but figured I would have it done preventatively since my speedometer and tachometer were intermittently working.
The car has around 171k miles, with about 3k of them under my ownership over the past 10 months or so.

Thankfully my ECU still had factory seals, so the warranty given by Tanin was 1 year (if you don't break the new seals they install).

Here's the video I made about it -