I'm very curious about the frequency and correlates involved in the failures that are being reported. I'm currently running 3 tri spark units (commando, TR7, T120) and have had no issues, but am not betting on luck of the draw. Maybe we could ask Jerry to start a thread for this issue? I think if we could pool data from different users regarding their individual set ups maybe we could identify what's going on, at the very least it may help Tri Spark identify problem set ups.
I'm running all my units with dual output coils (2 dyna's and one accel), charging systems are up to snuff, all three are directly wired (bypassing all kill switches etc.), all related wiring is well insulated, I have a small pieces of foam under the points covers to keep the vibes down, and I avoid spark testing (plug on head)
Good idea to pool all available info on these failures. My Trispark systems have all been the later type with the self-test facility. I went through a very lengthy correspondence with Steve Kelley via email, with at least one email per day over a 3 week period, during which Steve had me carry out test after test and answer question after question, all of which intended to rule out other possible faults (eg: with my bike's electrics). Fair enough to do that, of course. But I did find it frustrating to have to repeatedly respond to the same questions (I told him at the start that there is no kill switch in the picture on mine, yet I was asked to check the kill switch wiring several times).
My bike has its original main harness which is in surprisingly good condition. The front end has been changed due to using different handlebar switches. This work was done by a professional auto electrician and is top quality. It has a Lucas 3 phase alternator with a Podtronics type regulator/rectifier. It has Trispark's own coils. The wiring to/from the coils / Trispark module is duplicated, ie: in order to rule out any issues within the original wiring, I ran wires direct from the coils to the module.
All the wiring is exactly as per Trispark's instructions. All of Steve Kelley's tests were carried out (repeatedly) and Steve eventually conceded that the fault must be with his module. He sent me a free replacement and I bought another (at half price) to carry at all times, having lost confidence in them.
Steve asked me to send the failed module back and I said I would if he'd cover the postage. He didn't, and I still have it. There is a small blister on the outside face of the encapsulating material at about 7 o'clock. It seems pretty clear that something inside got far too hot. This is the one that failed first, with virtually no warning (the engine spluttered and stopped, just as if it had run out of petrol). There was an occasional, very weak spark after that and with repeated kicking, the engine might occasionally fire just once.
That's about it. The second one didn't actually pack up completely, it just began malfunctioning as if the timing had slipped (it hadn't, not in any mechanical sense anyway - I think it did electronically). I recognised the symptoms from the first failure (it had sounded and felt a bit retarded for maybe 10 miles before it suddenly failed completely). I fitted the third one at that point and stuck the malfunctioning one behind the side panel as an emergency spare (I was again on my way to southern France so couldn't come up with anything better).
I have seen on other forums, Triumph twin owners who have had similar repeated failures and who have tried various things such as placing a mica heat insulating disc between the module and the timing cover housing, and making air vents in points covers - all to no avail, apparently.
I know nothing about electronics but I can;t help suspecting that miniaturising whatever normally goes in a separate 'black box' to the point where it all fits into a points plate size disc, is asking for trouble.
I'd be interested to know whether anyone has been successfully using one of these 'Classic Twin' Trispark systems for 20000 miles or more.