Wednesday, 31 January 2018

Tech 21 XXL

I picked up this 90s Tech 21 XXL as I had read that it was a "sleeper", a nice distortion but not particularly well known. I knew that the design was based on opamp clipping with no diode clipping. I have never built or (to my knowledge) even played anything like that, so I was curious and also pretty confident that this would be simple to repair. This one would power up and would pass a bypass signal but the effected signal was very quiet - volume and gain had to be cranked all the way to hear anything. The tone knob sounded like it was working correctly. A bad opamp or switching FET maybe.


The insides were a bit of a surprise. All the jacks and pots are board-mounted, but it looks like it was assembled using the least amount of solder possible. A lot of the through holes are not completely filled. Re-soldering all these joints didn't help.

Dull joints, very little solder

This construction style is very annoying to work with. The input/output jacks do not butt up flush against the sides of the enclosure like with many other common designs, instead they extend through the sides. This means that the whole pedal has to be basically assembled inside the enclosure, the jacks and pots are installed and the PCB is aligned with all of their pins and then soldered in place. This seems very labour-intensive and makes any kind of repair a real pain.

WTF

After desoldering the 1/4" jacks and DC power connector I managed to get the PCB out and found another surprise on the other side. Tech 21 have used a SMD design (fine) but have covered it with a block of potting compound, presumably to prevent clones of the distortion circuit. Unfortunately this makes repair much more difficult as well.

The switching FETs actually seemed to be working correctly, so I decided to see if I could remove the potting (there are methods for this written up online, often described as "de-gooping"). I guessed that potting material for surface mount parts are probably softer than old-fashioned epoxies to prevent them from cracking solder joints as the compound cures. I hoped that heating with hot air and picking with a tweezers might be hood enough.

Potting removed, plus some parts

De-gooping went fairly well, I started at 100C and the compound cracked and separated from the board in large pieces. I increased the temperature as I went but eventually went too far (I think above 250C) and re-flowed some solder joints - this meant an entire chunk pulled away from the board, holding two opamps and some passives with it. The resistors and caps were easy enough to separate and re-populate, the opamps were replaced with new parts based on a schematic I found.

Restored jumper. Cut trace is visible.

Another surprise, my pedal is a different revision from the one photographed on Diystompboxes. It has a cut trace and a jumper wire embedded inside the epoxy block. Is this a genuine correction, or something designed to make reverse-engineering more difficult?

The pedal still didn't work with new opamps, but now that I could probe them I could see one was not getting any bias voltage on one of the inputs. The Warp control which sets the DC bias of the first opamp was not connected to +9V, I found a trace to the pot that was open circuit, possible from a scratch from a nearby electrolytic cap. A jumper wire fixed this.

Restoring +9V to Warp pot.

Re-assembling was another pain in the ass, I broke the DC jack and had to order a replacement with PCB pins instead of solder lugs. Getting the board into the enclosure and aligned with the jacks so they can be soldered is fiddly, in retrospect I should have tried to replace the jacks with the more common units that have a parts that threads from the outside of the enclosure, like modern EHX stuff.

The XXL sounds quite good - less compressed (and less sustain) than what I would usually expect from a diode-clipping distortion like a RAT. The Warp control doesn't seem to do much throughout a lot of it's range. It also does very little with a weak input signal (especially single coil pickups), probably because the first opamp stage isn't driven into clipping. With a loud signal generator I could different hear distortion flavours, presumably the clipping gets more or less symmetrical through the pot travel. Online reviews and clips confirm this, so I'm confident this is working as expected.

I like this pedal, but I'm a bit soured by the weird construction and the complete lack of giving-a-shit towards repairs. Uncovering the circuit took less than an hour, so it's not much of a deterrent to anyone who wants to make a clone, but it's definitely a hurdle for finding what has wrong. I'm not sure if I think it's worth looking at more Tech 21 stuff.