GT250R won't start (but cranks)

[CookMySock]

OK. So injectors don't have degrees of open?

They do, but not the way you are describing it.

The ECU changes the ratio of the amount of time they are open. So, for small amounts of fuel admitted, they open them for 1% of the time and shut them for 99% of the time. Contrast this to a large throttle opening and high revs, where they will be open for more like 75% of the time.

There are a wide range of fuel and air control strategies, but they measure the incoming air charge by either sampling the manifold pressure or by using a mass-aiir-flow sensor in the inlet. They use the throttle position sensor to augment their readings, and for additional tricks like acceleration enrichment (the old throttle pump.)

Consider also, that these days the ECU is going to be loaded with lots of modern software tricks that either just make life easier for the engine, or else run it leaner but safer.

There is lots and lots of home-made EFI stuff around now, and quite a few forums dedicated to learning about it, so there is no shortage of reading material for those so inclined. Search for megasquirt.

Steve

[MSTRS:Me]

Search for megasquirt.

I will never be able to clear that image...

[CookMySock]

See, here is a 3-D fuel table (or "fuel map") which has Intake Manifold Pressure (Y-axis), RPM (X-axis), and Fuel Values instead of spark Advance (Z-axis).

Don't panic when you first look at it. The colour is "how much fuel to add at this advance, manifold pressure, and revs. Red is more, blue is less."



Steve

[MSTRS:Me]

So...with a faulty temperature sensor, the injectors would not open for the 'correct' length of time? Therefore, letting fuel in, but insufficient for a cold engine?
Pray tell then...why did crash starting get a result?
I'm sorry, but I'm still convinced that there was some sort of physical blockage that cleared as a result of the greater pressures from a rolling start. The engine would be turning over quicker than a starter motor was capable of.

[bogan]

So...with a faulty temperature sensor, the injectors would not open for the 'correct' length of time? Therefore, letting fuel in, but insufficient for a cold engine?
Pray tell then...why did crash starting get a result?
I'm sorry, but I'm still convinced that there was some sort of physical blockage that cleared as a result of the greater pressures from a rolling start. The engine would be turning over quicker than a starter motor was capable of.

Yeh that about sums it up. My guess is crash starting results in a higher rpm, which went to a part on the map which made the mix a lot more palatable. Or as you say it unblocked something somewhere.

[MSTRS:Me]

In some ways, I'm sorry that crash starting worked. Because now all we're left with is guesses. Perhaps we'll never know what the problem was?

[CookMySock]

So...with a faulty temperature sensor, the injectors would not open for the 'correct' length of time? Therefore, letting fuel in, but insufficient for a cold engine?

Correct, but it is not clear whether a faulty sensor would falsely indicate "high" or "low" - that could only be determined by experience (sensors usually all fail the same way) or by the technician at the time.

Pray tell then...why did crash starting get a result?
I'm sorry, but I'm still convinced that there was some sort of physical blockage that cleared as a result of the greater pressures from a rolling start. The engine would be turning over quicker than a starter motor was capable of.

All of that is only predictable from a statistical point of view - what is more likely or not. I submit, it is unlikely that anything was physically blocked, as this area is very very well filtered indeed. If there was a non-flammable liquid in there it would have acted like it was blocked.

In the end, inside tiny little spaces with no windows no one really knows what happened, and we roll around opinions until we find one that doesn't easily fall over when subject to logic, and that answer's good enough for now. Sooner or later another fault will come along which either adds weight to the hypothesis or blows it out of the water.

Steve

[MSTRS:Me]

Sooner or later another fault will come along ...

Statistical evidence, plus physical evidence of it's being a Hyosung, suggests you are right.

[CookMySock]

Statistical evidence, plus physical evidence of it's being a Hyosung, suggests you are right.

I was meaning, if there really was something broken here, it will resurface shortly.

Steve

[MSTRS:Me]

I knew what you meant. The chance of a dig at Hyosung's reputation was just too good...
That is what worries me when a reason for failure has not been identified. Next time it happens may be somewhere distinctly inconvenient (not in his own driveway) or possibly even dangerous.

[neels]

That is what worries me when a reason for failure has not been identified. Next time it happens may be somewhere distinctly inconvenient (not in his own driveway) or possibly even dangerous.

That's what always worries me when things 'fix' themselves.

It sounds like this could just be a case of a flooded engine though, crash starting shifts a much larger volume of air through the cylinders than turning on the starter motor, and no throttle response for a short time is not unusual until everything sorts itself out.

[MSTRS:Me]

Yea, but this happened over a period of 2 days. You can't tell me that with no further tries, and new plugs, that a flooded engine is still flooded 24 hours later.

[neels]

Maybe?

The injectors will still be pumping fuel into the inlet manifold, if it's cold starting then more than usual. If that extra fuel hasn't cleared trying to start it just adds more to it, even with new plugs.

Anyway, I don't really know, just adding another theory

[bogan]

first step would be to see exactly how the cold starting circuit works, which sensors it uses, how much extra fuel is added etc, hyosungs have been around a while now, should be on the net somewhere.

[Racey Rider]

.... , and the tech needle was going a little crazy, which was strange.

We have had a problem with our carb'ed GT250r race bike like that.
Turned out there is a problem in the Dash unit, which the wiring diagram shows links directly to the front coil.
So the problem for us was not coils. We removed the electronic dash and ran the bike on the open road for 20 minutes without any fault re-ocurring.
Will test at the track in a couple of weeks to be sure.

Racey