Twin Plug 1883 Advance Curve & Ignition System Advice Needed

[Bill Sargent]

I am in the process of building a twin plug 1883 motor that will be 10.5/1 compression LN P&C with an Elgin 7208 cam. Carbs will be Solex with new shafts and butterfly plates - about 40.5 mm bore. Heads have stock exhaust and 42mm intake valves with valves unshrouded etc. I have not yet landed on an ignition system. One consideration is what advance curve to use. Can any of you out there with similar motors advise on the details of the advance curve you use? Static advance, full advance, what RPM advance starts, what RPM full advance is available etc? Other non flaming advice welcome.

Any thoughts on twin plug ignition systems also welcome. I am using one of Ibrahim Kuzu's 4 cam cooling units on the motor, so distributors much larger than stock diameter will not fit due to shroud to distributor clearance (i.e. nissan twin plug cap at 104 mm dia is too large). Precision Matters side by side distributor set also will not fit. I have looked into the Compu Tronix electronic coil pack unit that triggers with a distributor replacement unit, as well as a Clewett crank fire set up. Placement of the coil packs will be an issue with either, but are workable. Short of making an outboard V drive, al la Al Lager's Abarth clone set up, I am looking for other ideas.

[Eric Cherneff]

Bill, I'm guessing there are not a lot of these engines out there, so it may be hard to get much data. Although my engine is not a twin plug, it is similar in most respects, and I have been experimenting with advance curves.

The engine is an 1883, 10.0:1 compression, Elgin 7010-18 cam, 40mm intakes, stock exhausts, Solex 40's bored to 42, stock muffler.

All results are "seat-of-the-pants" based, I have not dyno'ed it yet. All distributors are modified VJR 4 BR 18. Here are some observations:
-maximum advance of about 32 degrees gives the best results
-stock distributor settings work, although static timing is closer to zero than seems good
-one distributor is modified for about 24* total advance, so I use static timing of about 8* BTDC, the curve is similar to stock but advances earlier, with it all in at around 2600 rpm. Occasional pinging can be heard on 92 octane, not on 94. I would like to delay the advance by 400 rpm or so to be safe, but I don't have the needed springs. Advance starts around 1000 rpm. All around good performer - I have lots of miles on this one, and am happy with the results, but want to experiment a little more.
-I am currently testing a distributor with about 18* total advance, so starting with about 14 degrees static. Very limited testing so far, but doesn't seem happy off idle or under 2000 rpm. Advance starts around 1200 rpm, all in around 2800. I think the initial timing may be excessive. Again, stronger springs to defer the timing are needed.

One factor that is not discussed often with respect to timing is the "quench area". I have minimized this by using about .033" clearance. Larger clearances here will likely have an impact on the amount of advance needed.

If anyone knows of a source for stronger springs or various springs that will fit a BR 18, I'd be interested in knowing.

[Bill Sargent]

Hi Eric,

Discussions with others a few months ago indicated that twin plug motors usually run best with 24 to 26 degrees total advance, as opposed to the 32 to 36 degrees on single plug motors. The Compu tronix unit is aimed more at the VW market and so has 20 degrees advance, however with the twin plug set up this may be about right if I were to use 4 to 6 degrees static. The curve is linear and fixed, starting at 1000 and all in by 3000. Compu Tronix said several months ago that they were working on a version that would have the advance curve adjustable by PC, but they seem to have made zero progress. I have a note in to another registry member who posted in Oct 09 that he was using a Compu Tronix set up on an 1883 twin plug with good results. I want to check long term results. Will relay that on if I receive a reply.

[Glenn Ring]

Does it have to be a distributor? Have you thought about a crankfire system with coil packs?

[Jacques Lefriant]

Hi i'm all for the crank fire since you can get a rock solid advance curve with all the bells and whistles. but you purists you can retain the distributors with the advance springs/weights to keep the rotor timed and fire the coil/s with the crank fire.
jacques

[Glenn Ring]

You can do a simple system with a crankfire trigger only, but the real gain is in controlling the curve of all 8 spark plugs individually. For example you can have the upper and lower plugs fire a degree or two differently (if that provides better performance).

[Jacques Lefriant]

Hi Glenn
Even if you just fire the plugs simultaneously a good crank fire system is far superior to any distributors since you can control the advance as a funtion of air density(temp/vacum), engine temperature, rpm, throttle position and retard after advancing and compensate for incipient knocking if you use the more advanced systems. you can retard for starting and if you add a blower or turbo you can even have more fun.
jacques

[Ron LaDow]

Bill,
From what I've learned, it has to do with flame travel, BIEP, rod angle and chamber shape. And the dyno results pretty much confirm what you can visualize even in 2-D.
Ideally, we want zero instant pressure on the piston before the crank reaches TDC. Given the chemistry of the reaction, and the limitations of dirty, old mechanical objects, that's not possible.
The limiting factors are CR and ignition lead; we want the highest CR to maximize mean pressure after TDC and the least lead to keep the instant pressure low early in the reaction. Where those curves intersect gives us the most power for that design.
Adding a second plug obviously shortens the travel, but doesn't won't cut the required lead in half, since the flame still has to travel out to the 'sides' of the chamber. And a bit of scribbling and sums shows the flame travel distance with two plugs to be ~.7 of the distance with one plug. Meaning for 86mm bore, the all-in lead drops from ~34* down to ~26*, right at what the dyno told us.
What was also interesting is power output with two plugs was tolerant of 3* change either way with practically zero effect. And asymmetrical timing (within that ~3*) had no effect.
As far as the curve, the Pre Mat parts are straight line and all in by 2,250. Fiddling with the curve within any reasonable amount didn't change the power curve measurably; again, the second plug seems to make the engine tolerant of timing changes.
For this reason, a crank fire system probably won't add power, but a programmable ecu might tell us of some advantages in really changing the curve.

[Ron LaDow]

Bump.
Finished my comments by edit feature.

[Glenn Ring]

Thanks for sharing actual experience Ron.

[Bill Sargent]

Talked to Lew at Compu-Tronix today. Their system replaces the entire distributor with a US made unit that is basically a mechanical advance distributor that triggers an electronic system to drive coil packs. Advance curve is stock 009 which is ~ 20 degrees advance, linear starting at 1500 RPM and all in by 3100 RPM. As supplied it is probably not optimum for a 10.5 compression twin plug 356 motor. 20 degrees advance is probably about right if static is set at 6 degrees, but it needs to come in starting at about 1000 RPM (or lower?) and be all in by about 2500 RPM. Given that it is mechanical advance, the curve could be changed, so it is a possibility. But $700 ($800 with rev limiter) seems kind of steep for what is essentially a 009 with an electronic cap and coil packs that needs to be recurved.

I also spoke with Clewett Engineering and for $1000 (XDi-1 module) you can get a true crank fire Electromotive set up, however the advance is adjustable only at static, 3000 and 8000 RPM, linear in between. For $1300 (XDi-2 module) you get a set up that the advance curve can be programed in with a PC (plus the possibility of many other sensors like knock etc.) The trigger plate mounts on the rear of the crank pulley. Alan at the Stable has built a couple of fuel injected 356s using the Electromotive crank fire system that he told me run very well. Again, with the 4 cam shroud issue is where to mount the coil packs.

[Ron LaDow]

Bill,
You have no need of the "Invasive Index" constraints that drove the Pre Mat twin-plug design.
'If it were me'...
I'd (do all the other mods and purchases) weld the heads and put three 12mm or 10mm plugs in each chamber.
If the 2D geometry model holds, you'd now cut the all-in lead to <15*, and I'd bet the CR/lead curves intersection would allow >11:1.
Disregarding economy/emissions, good carbs would do just fine for fueling.
Hang the damn coils on the compartment side walls.

[Brad Ripley]

When it comes to crank fire/injection (plus twin turbo) 356 engines, talk to Reiner Cooney in NH.

[Bill Sargent]

Ron,

Thanks for a great post on the 25th. I had assumed max advance was something always determined on a dyno and did not know it could also be calculated. Makes perfect sense though. Helps me in my research on what ignition system to use.

Interesting that Rainer Cooney is also a gas heater expert, having helped me with the one for my 64C.

[Jacques Lefriant]

Hi Bill
To do it right you must do it on a dyno. Max advance is not a linear funtion of RPM depending on the induction and exhaust resonaces your optimun advance curve can be convuluted. 4-Cams mount their coils on the firewall. you could also consider using coil over plug araigement. The Electromotive units are optimised for their wastefire coils but i have adapted them to single coil and dummy distributor. John Wilhoit has done extensive testing with similar motors and has published his results.
KTF
jacques