[Bullet Whisperer]

Over the last two years, much work has been done to try and squeeze even more power out of our R.E. 350 Clipper racer. It has had full, self generating, twin spark electronic ignition since I built it nearly ten years ago, but it was upgraded from analogue to digital CDI a couple of years back, a reworked [short conrod version] of one of the two engines was also installed in the rolling chassis and testing on the dyno showed promise and around the same power as the other, previous engine which was now out and being upgraded.
 Despite good dyno readings, on the track, under full load in top gear, it would feel like the back brake was coming on very briefly and at random.
 Ignition timing adjustments helped a little and I found that where the power was best on the dyno was where the pre-ignition was most violent. Advancing the ignition timing by around 8 degrees greatly reduced the pre-ignition [so I thought at the time], but also cost us about 2 BHP.
 Over the two years, experiments were made on the dyno and on the track, with timing, jetting and even compression ratios and cams, all to no effect.
 Then one day, while discussing this problem with the very helpful guy we buy our ignitions from, he asked me how close the CDI box was in relation to the coils - 'more or less on top of them', I replied. I was then informed that the CDI box must be at least 6 inches away from the coils and testing for resistance also showed our plug caps were not the resistor ones we had believed them to be 
 The CDI box was re-housed well away from the coils and resistor plug caps fitted, as recommended, then testing followed. Then racing. All sorted, no more pre-ignition, even with the ignition timing retarded back to the 'sweet spot' 
 So, that problem is behind us, but here's a thing - I started to wonder why advancing the ignition timing seemed to 'fix' the problem, while also reducing the power? Well, one stripdown of the top end revealed a dark ring around the liner, which corresponded with the top ring when close to BDC., this turned out to be a sort of 'photograph' from the blast that was occurring just after BDC when the piston was trying to come up on the compression stroke. What had felt like the back brake grabbing was in fact the occasional pre-ignition firing against the rising piston, trying to send it back the other way! Advancing the ignition must have moved the timing of this event closer to,or even to BDC exactly, giving the impression that it was fixed, when in fact it was still occurring and costing us the occasional firing stroke at the correct time as well, also the optimum setting for power was about 8 degrees later.
 All sorted now, but food for thought 
 B.W.

[High On Octane]

That's pretty crazy Paul.  I have heard of this happening before.  In fact, I believe there was mention of this somewhere in the instructions for my Thorspark ignition.  I have heard/read that this is also possible if you mount an ECU too close to the coil/s.   The ECU will actually pick up on the electro-interfereance and cause the readings of different circuits to go bezerk causing false readings of sensors and even cause rich and lean conditions.

Glad you figured it out and congrats on the finishes again. 

Scottie J

[ace.cafe]

Very informative story!

This is the kind of lesson that really sticks with people when they read it, because it has such an impact.

I really enjoyed reading it.

Thanks Paul!

[ajb235]

I experienced exactly the same thing on my standard 500 bullet, on full power it would miss a beat and it just felt like the brake had been applied. No amount of adjustment of the ignition timing would cure this, and it was only after a holed piston that I realised what it was. The exhaust valve seat was sitting proud of the combustion chamber by 2mm. The exposed edge of the seat was sharp, because the valve seat had been cut right to the exposed edge. This sharp edge was probably glowing red hot and causing the pre ignition.
I had the seat  machined back, with a 3 angle seat ,fitted a slightly oversized exhaust valve, and a new piston,and the
problem was gone.

[Bullet Whisperer]

I have just found a picture showing the dark impression of the top ring near the bottom of the stroke - it is quite plain to see in the bore, although no actual damage was done and it cleaned right away. I would say that is evidence of a random ignition event, occurring at the same degrees ABDC at the start of the compression stroke, as and when.
 B.W.