Kangaroo - fuel or ignition?

[RDfella]

Here's one for the carb and ignition experts.
Concerns my 1,000cc vee twin. Runs great, goes like stink but - is a real pain in those 30mph limits. Running at c. 2,700rpm in 3rd it 'lurches'. Hit 3,000 and it's really smooth again.
I initially thought it could be power pulses, given the engine power and the fact it has no cush drive (belt primary) but then thought, how can that be, when the pulses are around 2 - 3 hz and the engine firing is c. 2700 hz?
So started looking at carburetion. It has a pair of 34mm Amal MK2's. Pilot airscrews are around 1 1/4 turns out, so pilot jets must be about right. Needs a few moments running before choke can be dispensed with, so mixture seems OK. A fistful of throttle gives instant, clear response (either on road or in garage) albeit with a puff of black smoke (and occasional flame on overrun). But plugs, if anything, show lean as they come out snow white!
Yesterday tried a CO exhaust reading. Holding engine @ just under 3,000rpm my reading was 4.5 (apparently 2 is spot-on) meaning mixture is rich at around 12:1. Slide cutaways are 3, which is usually normal. Needles set halfway, though at 3,000 don't think they're in play yet as throttle is barely 1/4 open.

Then got to thinking ignition. I swapped the K2F mag (with 60* camring) for a Triumph distributor to ease starting. Despite modifying the advance mechanism to ensure near nil advance to preserve the electric starter, surely the advance is all-in by those revs? If not, I wondered whether the advance was 'fluttering' as the forward push of the flyweights was resisted by the drag of the points coming on the cam lobes. Has such an issue ever been studied?

Would love to solve this annoyance, as the bike goes beautifully - as long as you're not in a 30mph zone when I have to drop to 2nd gear and more revs than should be necessary.
   

[groily]

Despite modifying the advance mechanism to ensure near nil advance to preserve the electric starter, surely the advance is all-in by those revs? If not, I wondered whether the advance was 'fluttering' as the forward push of the flyweights was resisted by the drag of the points coming on the cam lobes. Has such an issue ever been studied?

Not studied in the exact context you describe RD, but the effects of various components set differently / badly or completely absent on ATDs was something we looked at a few years ago just for interest.
From the set-ups we devised, the highest rpm engine we saw before getting to full advance was c. 2800. (That was an ATD that gave 12° movement, note.)

This doesn't actually help with possible fluttering, but when in the past I've strobed engines with ATDs fitted, to see how they perfom dynamically, the advance curves have actually been reasonably steady. They are too when I interpose an ATD between the drive flange of my test rig and the mag. (Not fluttery enough, I think, to create a push-me pull-me effect). I'm not sure that doing similar under load on the road (if one could!) would say anything different as an ATD only really 'knows' how fast it's being turned. I wouldn't have thought the cam etc resistance on a dizzy would make all that much odds compared to a mag. (And congrats if you made the cam!)

I guess, to prove whether the ATD is or isn't the cause of the grief, you could set the thing up temporarily with fixed timing, get some friendly energetic persons to give you a shove to save the starter, and see whether it's any different? Bit of a faff, but on the 'one thing at a time' principle  . . .  Good luck!

[chaterlea25]

Hi RD,
The throttle needle will have an effect at 1/4 throttle, the separate stages overlap by quite a margin
The pilot mixture has an effect throughout most of the range
One aspect you have not told us about is the mounting position of the carbs?
Concentrics will give trouble with fuel height if the downdraught is too great
Instant acceleration and black smoke indicates excess fuel, I wonder if fuel is "puddling" in the inlet and suddenly getting sucked in when the throttle is opened? (subject to fuel height being correct)
Modern fuel does not seem to give good plug readings
I would try leaner slides as a starting point
Make sure the needles and the needle jet are not 2 stroke components and the little brass tube that sticks up around the needle jet is also 4 stroke

John

[RDfella]

Thanks for the replies. This one has certainly exercised my mind.
Not sure I've got enough brute force (human-wise) available among mates to get it started (especially if advanced). The w'shop electric rollers regularly blew a 13A fuse when trying to start it (on retard) before I fitted it with its own starter. Will have to consider that one, although I have to admit it's a bit of a wild card.

Re fuelling - carbs are level and mounted one each side of bike. Rear cylinder could puddle slightly - if left on side stand it will - which is why I fitted a small diameter drain (a la Jaguar XK manifolds) to the lowest point of the rear manifold.
Clearly pilot jets are about right and suit slow running, but had resisted changing slides because a) the black smoke is when you give it a fistfull (about half throttle or a bit more) and b) didn't want to end up going down the same route as I did when I had it fitted with Mikunis - ie buying jets / slides but getting nowhere. Guess I could convert the current 3's to 3 1/2 with a little machining, though.
As stated, the CO exhaust gas reading with engine held just under 3000 is high at around 4.5 indicating a ratio of around 12 instead of 14 ish. Didn't measure amount of throttle opening whilst doing that, but won't be much and so presumably is running on the cutaway.