Feature request: progressive fuel cut for overboost

[ArttuH]

Hello,
Would it be complicated to add an option to use progressive fuel cut also for over boost protection, like with rev limiter? So ideally there would be setting for "soft limit" zone where the code would cut out individual injection events progressively and above that hard limit with full fuel cut.

Background for this request is that I have been doing some tuning for a SuperStreetBike class drag bike. Big turbo, street tire, no wheelie bars as short description So precise power control is crucial especially on first two gears. Too little boost will make a slow run naturally and too much boost will ruin the run completely due to wheel slip or wheelying. And due to fast revving nature of these engines and limited testing possibilities tuning the boost control can be slightly challengin sometimes. So my idea is that this progressive over boost protection could provide a nice safeguard for boost control. Progressive cut would prevent excessive boost even if the boost control isn't able to handle it. Current full cut protection is a bit too harsh to be practical for this purpose.

[jsmcortina]

V-twin or 4-cyl ?

I could see 4-cyl working, but on a V-twin it is likely going to be harsh regardless and there are only two cylinders.

James

[ArttuH]

4-cyl, Suzuki Hayabusa in this case. Though I guess this feature could have wider use too. At least many other drag racers would benefit of this and maybe it would be a nice addition for general purpose over boost protection too.

[xrattiracer]

not really directly related to this, but an idea of using a load cell or fork extension sensor on the front to reduce power (via boost control, fuel/ignition cut, etc) popped into my head as a way of dealing with wheelie issues. and typical traction control strategies for wheelspin issues.

[dontz125]

Hmmm ... Spark table 1 normal, Spark table 2 -seriously- retarded; Blend Curve 2 applied according to 'Sensor 1' on the fork? Similar foolishness using Blend curve 6 for boost tables?

[ArttuH]

not really directly related to this, but an idea of using a load cell or fork extension sensor on the front to reduce power (via boost control, fuel/ignition cut, etc) popped into my head as a way of dealing with wheelie issues. and typical traction control strategies for wheelspin issues.

That would be handy indeed Or actually normal traction control with two wheel sensors should handle wheelies as well. However, all kind active traction or wheelie control schemes are strictly forbidden by technical rules. Basically there shouldn't be any feedback from wheel speed or suspension position to power control.

So I'm not looking for ways to implement wheelie or traction control but additional means for boost control Typically you need to control boost from 0.5 bar to over 2 bars during a run on these engines. This alone makes control slightly challenging. The wastegate and its piping must be really well flowing to keep boost low enough at red line. And due to limited space it isn't always easy get the wastegate plumbed optimally. The engine also runs very quickly through the rev range on low gears and in addition the large rotating assembly of the turbo doesn't react too quickly to wastegate control. Due these it's often almost impossible to bring boost back down by normal boost control if it gets over the target on low gears. On the other hand even short interrupts on engine power, like quick throttle dip or over boost fuel cut event, seem to tame down boost build up very effectively. Therefore I think the progressive fuel cut would be pretty good safeguard for boost control glitches.

[tpsretard2]

i have one comment and a suggestion which might lead to a new feature.

Unless the engine in question now or future is running sequential injection, i see this leading to a partially fueled cylinder at some point which would not be ideal.

I see you main problem of being able to control boost from a VERY low amount, 5psi maybe a bit less to your peek boost. We over came this very easy in the world rally cars by running Anti-Phase boost control. You need a dual port wastegate, and a pair of 3 port solonoids. You need to use 2 pressure points for each valve, you can not T them off the same line this is because the pressure is always being vented off, ether in the phase of Anti phase line. All lines should be 1/4 inch or more and the mac valves should all be quarter inch barbs or better an fitting.

The Phase will connect to the top of the wastegate and the Anti-Phase to the bottom.

With this configuration you can run a simple 5psi spring in the wastegate. With this setup we can control boost from 5psi to full load boot as high as 60psi.

for james now. For this feature you have done most of the code already. in 8 our of 10 situation the phase and anti-phase was 180deg out just like on the 3 wire idle control. i have a few cars where it is not but they are VERY odd and extreme cases. We have a table that we can alter the duty on the phase and anti phase, but like i said, i do not "THINK" it would be needed here or for most. This is also becoming very popular with a few drag cars. There are a few 4 port solonoids people have made to try and do this. all though on paper the function is the same, it just does not work as reliably or as well.. I am sure it is to do with having 1 pressure supply and it not being able to keep up with the vented portions of the cycle.

[xrattiracer]

ahh, understandable about racing body restrictions
even with sequential fuel, i could imagine there would still be some issues with underfueling due to wall wetting. i dont know how much real world effect this would have, but it seems worth considering. any reason to use fuel cut instead of spark cut?

[ArttuH]

Naturally this would need sequential injection. Or at least individually controlled injectors with batch firing. I don't think partial fueling would be any real problem if injectors are placed in "normal" position close to the intake valves. And this same functionality is already in use with rev limiter and I haven't seen any reported problems with it.

Progressive ignition cut should be feasible too. I'm just afraid that it wouldn't help over boosting as unburnt fuel may ignite in the exhaust headers and keep the turbo spinning. In addition there is always a small risk of valve damage with ignition cut on turbo engines. Therefore I would prefer fuel cut. But maybe there could be option to use ignition cut too, just like with the rev limiter.

Better waste gate control would help too. Either that anti-phase control or dual solenoid setup with external pressure source (compressed air or CO2). Actually I'm currently investigating how to implement the later one. But this is kind of off-topic in this thread I think the progressive over boost cut would be useful feature even with the most sophisticated waste gate control.

[ArttuH]

James (or other developers), what's your opinion on this? Would this be worth of implementing in the code? While this probably isn't the most urgent addition I guess it won't need huge effort nor cause much conflicts with other functionality. Of course I might be completely wrong. So I hope this could be added on some forthcoming version. It would give a small but nice competetive edge against other ECUs

[dontz125]

The code and hardware will support anti-phase boost control. Select IAC1 (or 2) as your boost control pin; wire IAC1A as the 'phase' pin, and IAC1B as the 'anti-phase' pin. The phase solenoid is plumbed to the top of the wastegate diaphragm; the anti-phase solenoid is plumbed to the underside.

phase_antiphase.jpg

Take a look at this page for comments on anti-phase operation.

[MWPau]

ahh, understandable about racing body restrictions
even with sequential fuel, i could imagine there would still be some issues with underfueling due to wall wetting. i dont know how much real world effect this would have, but it seems worth considering. any reason to use fuel cut instead of spark cut?

Progressive spark cut could lead to backfires which then have an anti-lag/2step effect.
I'm thinking fuel & spark cut would be the best idea.

[ArttuH]

The code and hardware will support anti-phase boost control. Select IAC1 (or 2) as your boost control pin; wire IAC1A as the 'phase' pin, and IAC1B as the 'anti-phase' pin. The phase solenoid is plumbed to the top of the wastegate diaphragm; the anti-phase solenoid is plumbed to the underside.

Thanks for a hint, I need to take a closer look on that. Though I guess that external pressure supply would be the right way to go with this particular application. That's what all the fast guys are using with AMS1000/2000 controllers. But I think the control side can be handled with the MS as well.

Just for clarification, I would like to keep this topic about progressive over boost cut, not different boost control strategies I appreciate all other comments and suggestions too but they are slightly siderailing this discussion.

[arran]

Why would you want to deliberately deliver a lean condition exactly when you need to be doing the opposite?

[dontz125]

I think the idea is not to deliver a lean condition, but rather to cut the fuel entirely - but only to one cyl at a time, rotating.

That said, it does seem to offer the chance of things going wrong. Would progressive spark cut be a better concept? (Not a turbo guy, nor do I play one on tv)

[slow_hemi6]

Why would you want to deliberately deliver a lean condition exactly when you need to be doing the opposite?

Exactly what I thought when I read the thread title.

[ArttuH]

Why would you want to deliberately deliver a lean condition exactly when you need to be doing the opposite?

As dontz125 mentioned this wouldn't mean running the engine lean but cutting fuel completely cycle by cycle. And as mentioned earlier this same scheme is already available for rev limiter and works fine as far as I know. That's also why I think this would be relatively simple addition to the code.

Spark cut would be possible too but I think it wouldn't cut boost rise as effectively.

[jsmcortina]

Progressive fuel cut is a complete cut cylinder by cylinder on a non-random basis. It is designed to NOT give a continual lean condition which would otherwise arise if cylinders were cut and re-applied randomly.

James

[racingmini_mtl]

Progressive fuel cut is a complete cut cylinder by cylinder on a non-random basis. It is designed to NOT give a continual lean condition which would otherwise arise if cylinders were cut and re-applied randomly.

James

Does it (or should it) take into consideration wall wetting, i.e., does it add some amount of fuel after a cut?

Jean

[jsmcortina]

Does it (or should it) take into consideration wall wetting, i.e., does it add some amount of fuel after a cut?

It presently does not add fuel afterwards, but it does take account of wall wetting in the sense that it is a non-rotational fuel cut with a hysteresis to reduce the on/off events per cylinder.

James