Ignitech TCIP4 individual advance correction failure

[piranha-bro2]

Soooo… this may be an oversimplification and a product of my limited knowledge and my misunderstanding of Jon’s explanations but if the system doesn’t allow an “advance” correction, for cyl2, could you (or did you try) retard cyl1 ? Then rotate the backplate to suit?

P.S. if you end up adjusting the physical pickup positions, I was able to correct the same timing issue (on a different machine and not Ignitec) by creating a tiny bit of clearance on the backplate, allowing some sideways movement, then I could move one pickup without moving the other. It was a bit of a pain- microns in it- but it worked, and allowed me to adjust it with the motor running and one eye on the strobe.

Revisiting this very interesting thread now that the SFQ is only hours away from coming back to life. In my endless frustration with this discrepancy issue between advance on cyl 1-2 last night I did precisely what you did to correct a similar issue, Andy - and Eureka! ... it worked. As you said, it's only a bee's whisker, but that's enough to make the diff. Yet to try it on a running engine, but confident it will be good on the strobe at max advance.

I don't know whether i blame myself for not getting Ignitech's printed Guzzi pickup circuit board dead centre on my alu mounting plate, or whether I blame the manufacturer of said board for not getting the sensors exactly 135 deg apart (read half crank speed, so 270 at the crank), but the fact is I now have eq advance left and right and can proceed to the next checkpoint. Whew!

 

[Vince]

You try anything if you're desperate enough. Years ago I had an unexplained lack of power off idle on my Pantah, I spent months playing around with vacuum gauges trying to fix it with no change, someone suggested adjusting the pickup gap so I spent the day doing that with the bloke who suggested it. Using brass feeler gauges, NEVER saw that BEFORE we went to as many changes as we could, it had an effect but didn't improve the off-idle power loss much. What fixed it was manually sincing the carbs, using vacuum had it miles out. great news Q finding your issue so that tiny variation tipped one cylinder into detonating?

 

[Dellortoman]

I missed this thread during my absence from the forum for health reasons. I just read through it. Interesting stuff, although I wouldn't have been any help solving Q's issue. I'd have assumed an advance correction for one cylinder would do what it says on the box.

I'm still confused about John's description of advance actually being a retard from a pre-set advance of (in his example 40°).

The reason I'm confused is that when setting up an Ignitech (I have little experience of other electronic ignitions), I set the pickups such that the trigger event happens close to TDC. It usually ends up being a degree or three BTDC, which is near enough for practical purposes - the software can take it from there. So where does the ignition system get its theoretical 40° datum from?

I don't know anything about the inner workings of electronic ignitions. They're just a mysterious black box to me. But my basic understanding of physics tells me that no ignition system, no matter how clever it is, can read a future event. So the only way it can create a datum (or indeed a spark) before it gets it gets to the pickup at TDC is to work it out from the last time the rotor passed the pickup, which was 320 degrees of crank rotation ago. So a 40° advance is actually a 320° retard. Is there something wrong with this line of thinking?

The typical crank position encoder device used on Laverda engines only has a single-point trigger. By that I mean it only reads the crank position at one point in its 360° of rotation. It's not a missing tooth type encoder wheel that provides a pulse every few degrees of crank rotation, and it can't see what the crankshaft is doing between signal pulses. So the calculation of a 40° "advance" (= 320° retard) must be pretty hit and miss, especially at low revs when the variations in engine speed are all over the place.


Edit: I re-read John's bit about his IIS ignition starting from a datum of 40° advance. I think I now understand that to mean that's where he actually puts the pickup. It's a physical location rather than a software interpolation. That's why he calls it a "hard" data point. Being 40° DTDC, it only has to apply a small delay to the spark to get it to fire at whatever position is called for by the advance map. That seems to me a much better solution (less prone to error) than waiting almost a full revolution of the crank to fire the spark.

 

[iis_iis]

I missed this thread during my absence from the forum for health reasons. I just read through it. Interesting stuff, although I wouldn't have been any help solving Q's issue. I'd have assumed an advance correction for one cylinder would do what it says on the box.

I'm still confused about John's description of advance actually being a retard from a pre-set advance of (in his example 40°).

The reason I'm confused is that when setting up an Ignitech (I have little experience of other electronic ignitions), I set the pickups such that the trigger event happens close to TDC. It usually ends up being a degree or three BTDC, which is near enough for practical purposes - the software can take it from there. So where does the ignition system get its theoretical 40° datum from?

I don't know anything about the inner workings of electronic ignitions. They're just a mysterious black box to me. But my basic understanding of physics tells me that no ignition system, no matter how clever it is, can read a future event. So the only way it can create a datum (or indeed a spark) before it gets it gets to the pickup at TDC is to work it out from the last time the rotor passed the pickup, which was 320 degrees of crank rotation ago. So a 40° advance is actually a 320° retard. Is there something wrong with this line of thinking?

The typical crank position encoder device used on Laverda engines only has a single-point trigger. By that I mean it only reads the crank position at one point in its 360° of rotation. It's not a missing tooth type encoder wheel that provides a pulse every few degrees of crank rotation, and it can't see what the crankshaft is doing between signal pulses. So the calculation of a 40° "advance" (= 320° retard) must be pretty hit and miss, especially at low revs when the variations in engine speed are all over the place.


Edit: I re-read John's bit about his IIS ignition starting from a datum of 40° advance. I think I now understand that to mean that's where he actually puts the pickup. It's a physical location rather than a software interpolation. That's why he calls it a "hard" data point. Being 40° DTDC, it only has to apply a small delay to the spark to get it to fire at whatever position is called for by the advance map. That seems to me a much better solution (less prone to error) than waiting almost a full revolution of the crank to fire the spark.

Your on the path to ignition guru status Cam, especially there is ( Paraphrasing ) no "advance" only large and small degree(s) of Retard. Yes the 40dbtdc is a position on the encoder pcb for the iis, Cyl 1 40degrees before tdc - TDC - Cyl 2 ( 180 degrees opposite Cyl 1 ) for 120 Cyl 1 TDC Cyl 2 Cyl 3 and around you go all Cylx are 40DBTDC now here comes the fly.. the pcb can be moved relative to the crankshaft, so you can add or subtract degrees around TDC AND all 2 or 3 forward reference triggers.... base advance then becomes movable... that fly just landed on your laptop programer... he is a mate of mine... much laughing here... j

 

[piranha-bro2]

You try anything if you're desperate enough. Years ago I had an unexplained lack of power off idle on my Pantah, I spent months playing around with vacuum gauges trying to fix it with no change, someone suggested adjusting the pickup gap so I spent the day doing that with the bloke who suggested it. Using brass feeler gauges, NEVER saw that BEFORE we went to as many changes as we could, it had an effect but didn't improve the off-idle power loss much. What fixed it was manually sincing the carbs, using vacuum had it miles out. great news Q finding your issue so that tiny variation tipped one cylinder into detonating?

Ignition retard on left wasn't the cause of the melt, Vince. but it was an issue for me from day 1 and I wasn't going back on the road until I had a fix of some sort. More problematic at the track day was a. float level a tad low on left and b. ran it out of fuel (so you can imagine it must have been getting leaned out with fuel surge even before it ran dry). Hopefully all good now. Startup tomoz hopefully.