Smaller injectors than the ECU thinks

Hey knowledgeable people,

Need a hand figuring out what's going on as I'm new to actual tuning but I've done a few courses.

Recently acquired a buggy that has a Haltech elite 500 & Honda blackbird engine. Injector data has proven to be elusive. Previous owner couldn't get it to run.

I've sorted the wiring issues and got it running. Without access to a dyno all I have is a long hill on a farm. I've roughed in fuel and timing and it runs great. Pops a bit on sudden lift off but drives well enough

Recently had the injectors flow tested and they are much smaller than my initial guess. When I changed the flow data in the ECU the engine over fuels until it stalls with wet plugs.

Flow test was done at 3 bar and I'm running 3.5 bar. They averaged at 219cc/min, my guess was 380cc/min.

Do I not have enough timing? Is my base fuel pressure too high?

If it's working fine do I leave it alone?

Need some advice from smarter people than me

You need to scale the VE map back down by the difference in flow rates you entered. 219/380= 0.576 multiplier on the VE table.

Or just change everything back as it was if you don't intend on chopping and changing injector setups. Basically if it's running well and you aren't running a torque demand engine model/gearbox control it doesn't matter if injector capacity/VE/fuel pressure differences are baked into any particular table. The only one that will cause significant problems is if dead times are significantly off as that will still effect AFRs under different ambient conditions at idle and low load.

Technically at 3.5 bar static flow will be closer to 219cc × SQRT(3.5/3). ~ 236.5cc. But unless you are going to swap injectors down the track it's irrelevant.

Some high performance motorcycle engines run 2 injectors per cylinder, one closer to the head in the intake manifold(s) for low rpm running, and another set closer to the air-box, even inside it, for high rpm power. If you've just been handed the project, at some point one rail, or set of injectors, might have been overlooked and not actually been connected correctly by someone previously?

You should be able to find manuals on-line, with a little searching, which should clarify this.

I'm so confused

Not planning on changing the injectors as I doubt the engine will outlast them in racing conditions. I've just put everything back to how I had it and it runs fine again.

I was hoping to gain some understanding as to why it over fuelled when I decreased the injector flow rate in the ECU as it seems counter intuitive.

Please correct me if I'm wrong cos I wanna understand it.

Because I've scaled the VE map to run with larger injectors it is applying fuel to suit those injectors, using smaller injectors with the same map it over fuels because the VE numbers aren't the "true" VE of the engine and are actually significantly smaller?

At no point have the VE numbers in my table exceeded 90 and applying the correction that @slides gave the VE to the table the largest numbers of see are just over 50.

This engine can't be so inefficient at filling it's cylinders with air. Why does it run so well with my map? Could it in theory run MUCH better?

The simple answer is whatever combination of physical behaviour of components (including fuel pressure fall off or line hatmonics) and nominal injector flow, VE map, AFR target map, dead times was delivering you a running engine, if you change a number in that set by 40% without physically changing anything on the engine, the fuelling will change by that amount.

Are you using a wideband?

What is your target AFR table/values? You haven't mentioned target table values at all yet?

Can you screen shot and post VE, AFR target and injector output/setup pages?

There are several tables effectively multiplied together.

Things that impact final fuelling: injector dead time error, injector nominal flow, base fuel pressure and fuel pressure instability across demand range, VE, AFR target table, air and coolant temperature corrections, injector phasing and overlap can have an impact too.

Can you give a detailed description of the full mechanical and wiring setup?

Definitely only has one furl rail and set of injectors? Not running staged injection?

Is it running sequential semi-sequential or staged injection (don't expect staged without 8 injectors)?

Have you traced all the injection wiring?

Was the flow data you got off a bench that matches closely to data from Bosch or ID on their injectors? Or just a random squirt in a tube test?

Has anyone on a make specific forum posted factory injector flow data for that engine?

Do you have any weird stuff in air temperature or coolant temperature compensation tables?

High revving engines can be effected proportionally more, and waste spark/semi-sequial more again at full power by injector dead time errors as it starts to make up a fair portion of the available injection window but I wouldn't have thought that would be a significant problem in your case.

I would assume something like that would be reasonably happy around 0.85 lambda on load (probably idle too), maybe a bit richer if getting flogged in a heavy buggy?

The answer to most of the questions is the same.

I don't know......

It's my first attempt at tuning and I'm just happy it runs.

The questions I can answer are:

HALTECH wideband

Single set of injectors

Sequential ignition and injection

I've traced all the wiring, it's the main reason why it runs.

Flow data was a squirt in a tube with details attached.

The amount of different opinions on the flow rate is identical to the amount of make specific forums discussing the exact topic. Hundreds!

I'm aiming for 0.85 lambda under load, 0.95 at idle and somewhere in between those numbers for partial throttle percentages.

I can't get the internal data logging to work so I haven't added much timing so everything is done by trial and error and what I feel in the seat.

I've been right through the compensations and nothing looks like it should bother it.

The buggy lives at a mates place that has more space 150 KMs away so I can't screen shot anything until I'm near it again.

I'm not aiming to get maximum power until I work out what I don't understand.

The only thing I don't understand is why it gets more fuel when I decreased the injector flow rate to what it actually is.

Because the VE model compensates for injector size, you hit your target AFRs with the same physical size injectors when the software "sees" the bigger number, you then inputted a small injector size so the ecu added injector open time to compensate for the "smaller" injectors that were actually the same size as before. That's the point of those fuel models, if you tune it with accurate injector data you can swap injectors without remapping, again with correct data and it should compensate open time to suit.

If you tune with incorrect data the map has built in errors.

Possibly stupid question, it's not actually a ~750cc block is it? That would explain close VE with injectors that much smaller than your initial guess.

That makes so much sense!

This is the piece I was missing on understanding, now that I know why I feel a bit daft that I didn't figure it out.

Definitely a blackbird engine, the whole bike with ECU, cluster and key barrel came with it in case the aftermarket stuff wouldn't work.

It says 1068cv or something on the block.

I'd say it works with the map I've put in because I had to pull a tonne of fuel across the whole map to stop it spluttering everytime I touched the throttle.

I'm only feeding it large fuel numbers north of 5000 RPM and over 20% throttle.

I've been very conservative with i enjoying timing because I haven't been able to read the onboard data log to tell me what it's doing. I've only been adjusting based on what I'm feeling when I drive it. I have a 300 metre long hill that climbs quite steeply in a straight line so I can watch the lambda numbers through most of the rev range before I rapidly run out of real estate. Holding it on the brakes isn't really an option as the single disc rear can't handle much heat and it's working hard to stop the torque. If I have the brake bias any further forward if lights up the tyres and screws up my test.

It's not ideal but I think I've got my head around it.

Thanks heaps for helping me understand what I was missing