Forum

[PeteF]

OK RB, new to me though.

[Rattlebattle]

FWIW I think it is most likely that the UCE has an ignition advance curve along the lines of something like a Boyer. I feel that the engine is too big not to have some form of advance curve. What I am more skeptical about, based on empirical evidence on my own bike and others who have fitted the later Hitchcocks carb conversion with no ill effects, is that the ignition advance is more sophisticated than that eg that it is affected by the TPS, or inlet vacuum. If this were so then either the engine would run permanently retarded, in which case it wouldn’t be quicker than it was with EFI as mine is, or it would run fully advanced, in which case it would probably pink. The only other option that I can see is that, if the ECU senses that there is no input from one or more sensors, it either defaults to a fixed advance curve ( like a simple Boyer) or sets a mid-point limp home mode. I think that a default advance curve, uninfluenced by TPS etc input is the most likely. After all, a lot of us used to put Boyer and similar systems on our twin cylinder bikes back in the day and they worked perfectly well based on engine speed alone, not needing to know how far open the throttle is. That said, it’s all speculation, but I do know that disconnecting the sensors and running with a carb produces a good-running engine. I’d love to know the truth of how it operates....

[Wheaters]

I know the title of this thread says 500 UCE but my 350 Electra (iron engine) has TCI ignition. The TCI box actually has “9-32” cast onto the case. I’m sure this corresponds to the built in advance curve.

The only “Closed loop” information this box receives is rpm, just like the old style mechanical advance mechanism.

[Cranky]

I dont seem to be getting my question accros very well.

If the spark is to be advaced and there is no question that it would be better if it did how does it advance from say 8 to 32 when it hasnt happened yet. The tab is fixed to the flywheel and it clears the trigger at 10 BTDC. SO- How can it move the command forward in time.

Even the ECU with its 28 tabs and one TDC tab must count 27 tabs and then 26 tab to decide where it is. Many call this the crankshaft position sensor.

However how does a TCI or CDI know when to advance earlier than it has the information and command to do so.

A capacitor delays an electonic charge like a battery so in theory it could delay but we are asking for an advance.

Unless it starts to read the leading edge of the tab
????

[Wheaters]

The TCI sensor on my bike only has one “trigger” tab on the flywheel to look at and it’s quite a narrow strip of metal. The system is therefore very basic. It can only cater for rpm; effectively it’s just the same as fitting an electronic module in a distributor, except being crank driven it is a “wasted spark” system in that it fires every time the engine approaches TDC. As the engine revs higher, the advance angle increases. How quickly it reacts to rpm changes is open to speculation but I think it might be just a couple of engine revolutions.

The engine in my trials car uses a “36 minus 1” setup. When I installed the “new” engine with a fully programmable ECU I had the outer perimeter of the flywheel machined to form the “lumps” that the crankshaft sensor reads; the trigger point is the “minus 1”, ie. the missing lump. The ECU also needs a TPS, a manifold pressure sensor, an inlet air temperature sensor and a coolant temperature sensor. It fires all three injectors together; ie batch firing. I think the ECU can be set to read either the leading edge or the trailing edge of the trigger “lumps”; I don’t think it changes from one to the other by itself.

[Haggis]

Cranky, I would imagine that the ecu runs full advance until it can work out what revs the engine is doing and retard the timing to suit.
It will delay the spark at low revs.
So its not preempting when to spark.
As you say it can't do something that hasn't happened.

[Adrian]

Cranky,

I've never worked out HOW the electronics in a suitably equipped TCI (Or CDI) box manage to vary the timing with nothing more than a button or strip of something whizzing round on the outer edge of an alternator rotor to tell them what's going on (ignore ECUs and TPS for now), I'm just grateful that they DO, even if it's all techno-voodoo to me.

For my (sort of) Electra-X based project I've decided to adopt a slightly different approach, not sure how you could fit one on the UCE engine.

A.

[Rattlebattle]

What Haggis says. The ECU will take its signal from the fixed position of the pickup. It will simply be programmed to count the revolutions per minute and retard and advance the ignition accordingly, exactly like the early electronic ignitions that replaced points and auto advance/retard setups. I don’t see how else it could work. In reality once off idle it’ll be at full advance most of the time anyway, going by memory of strobing Triumph twins with this set-up. They never were more sophisticated. Cars of the day had vacuum advance to cater for load on the engine when driving up gradients at low revs.

[Haggis]

Rattlebattle, thats what you loose with a carb conversion on an efi.
You still have a reasonably good base advance curve but you loose out on the dynamic stuff that the map and tps sensors would give on light throttle openings.
Not enough that you would notice though.

[Wheaters]

In my not too distant past I’ve rebuilt Lucas distributors and modified their mechanical advance mechanism for my tuned engine. I needed to do a fair amount of research on advance settings.

It’s usual to aim for full ignition advance by about 3,000 rpm. The range of the curve can vary quite a lot, depending on a number of things.