Author Topic: Inlet Backfire When Restarting After Stall  (Read 17129 times)

Offline Sprocket

  • Hero Member
  • *****
  • Posts: 867
  • BHP: 29
Inlet Backfire When Restarting After Stall
« on: August 25, 2007, 11:32:59 pm »
I think i may have found a bug in the software? If the engine has stalled and you try and restart without resetting, in some instances, the engine backfires and blows the air filter clean off the injection manifold, LOL, its funny but its not if you know what i mean :P. I believe this is because as the engine stalls, it can counter rotate, the primary trigger is counting teeth in both directions and gets VERY lost. Is it not a simple matter to reset the primary trigger algorythms when zero RPM is seen? Yes granted you can switch the ignition off and back on, but surely this is not a complicated thing to work in? Something to think about if its not already been addressed?

Offline rob@vems.co.uk

  • Hero Member
  • *****
  • Posts: 3115
  • BHP: 49
    • VEMS Forum
Re: Inlet Backfire When Restarting After Stall
« Reply #1 on: August 26, 2007, 08:09:49 am »
The triggering algorithm resets in all manner of circumstances once RPM drops below a certain value it will consider the gaps to be too large and will then try to synchronise with one clean rotation.

What is your cranking advance figure?  And what sort of values have you got for Crank Warm PW, and afterstart enrichment scaling at 77decC?

Offline Sprocket

  • Hero Member
  • *****
  • Posts: 867
  • BHP: 29
Re: Inlet Backfire When Restarting After Stall
« Reply #2 on: August 26, 2007, 08:14:22 pm »
Crank advance is 10 degrees.

I dont think there is a problem with crank and afterstart pulse widths unless they change after stalling??. The backfire, only occurs after a stall without resetting the ECU. I do also get trigger errors reported some times but not all the time, only in this situation.

I did find this some times when i had ( still have) the stalling issue on over run, when jump starting the engine by bringing the clutch back in while still moving. Trigger errors reported then, but it is the same thing, just didnt use the starter motor.

All other times the engine starts first time every time cold or warm, and never has it backfired.

I still feel that something is not right with the software??


Firmware version 1.0.71

Offline rob@vems.co.uk

  • Hero Member
  • *****
  • Posts: 3115
  • BHP: 49
    • VEMS Forum
Re: Inlet Backfire When Restarting After Stall
« Reply #3 on: August 26, 2007, 10:48:40 pm »
Trigger errors are really really really bad news.

Thats not a software issue, its a trigger hardware issue, noise, trigger hardware issues, VR sensor polarity...

IIRC you modified a trigger system yourself?

The only difference between stalling and resetting is the power-on priming pulse.

Without datalogs, crank traces and tables.txt and config.txt we're still flailing in the dark as to what could be wrong.

Offline Jorgen

  • Global Moderator
  • Jr. Member
  • *****
  • Posts: 59
  • BHP: 500
Re: Inlet Backfire When Restarting After Stall
« Reply #4 on: August 26, 2007, 11:21:21 pm »
The cranking injector pulsewidth demand is different after a reset if you use a priming pulse.

With the priming pulse you can often start the engine even if the cranking pw is not sufficient. A restart after a stall would not have the benefit of the priming pulse and it could cause a lean start condition. As you know a lean condition can cause backfires. That is why Rob asked for that information.

But if you publish msq, tables.txt and config.txt we will take a look at it. My bet's is of course on a marginal VR trigger.

Jörgen

Offline Sprocket

  • Hero Member
  • *****
  • Posts: 867
  • BHP: 29
Re: Inlet Backfire When Restarting After Stall
« Reply #5 on: August 27, 2007, 12:34:03 am »
I'll sort all those out in the morning.

Prime pulse is switched off.

I hear where you are coming from , however, why does this only happen in this situation. if there is no real difference in PW when starting after a reset compared without a reset, and the primary trigger algorythm resets in each case, why is it back firing.

If there is a marginal VR, then the problem would exist all the time, not specificaly after a stall? Trigger errors are only reported after the backfire since VEMS has counted either not enough or too many teeth (counter rotation giving a large count marginal VR a minimal count) for which it has been configured for. If the software resets then there isnt a problem, it would just be like a hard reset.

The problem is only there when the system has NOT had a hard reset.

Engine stalls, Restart imediately without hard reset, backfire, trigger error reported, reset, starts first time, no trigger error. Stall, hard reset, starts every time first time. That is the chain of events.

Not being funny but I carry out a lot of diagnostics on control systems at work, My title is Technical advisor, and I work for the a world Leading Manufacturer. The control systems our machines are not that different from VEMS. I feel i have narrowed down the area of the problem too look at fairly acurately.

However, i am human, and do make mistakes, lol, lots of them. i will tomorrow check everything and RE RE check the air gap on the VR. The trigger kit I am using is a tried and tested system on engines seeing 8500rpm regularly, the sensor is off the Ford fiesta with a narrow body to take better advantage of the narrow trigger wheel. I have used the same sensor and trigger wheel on the OEM system without issue, where the OEM trigger wheel is a completely different as is the sensor.

Also while im at it ill sort out the rest of the info for the stalling issue where VEMS goes into idle control on overrun ::)

Thanks for the help so far, it is apreciated

Colin

Offline rob@vems.co.uk

  • Hero Member
  • *****
  • Posts: 3115
  • BHP: 49
    • VEMS Forum
Re: Inlet Backfire When Restarting After Stall
« Reply #6 on: August 27, 2007, 08:54:43 am »
Cool, I'll be able to run the thing up on my test bench, and also convert your setup to 1.0.73 so that you can run a release version of code rather than an intermediate, which may be buggy.

I dont think anyone is being funny about any of the suggestions and replies, and there is no doubting your abilities as you've got a running system, the problems come from the huge number of variables involved, not just with VEMS, but with the hardware, the engine etc.

We still dont know of all the issues involved with a TBi system, and from where we are the only way we can see the issues are by trying to replicate the problems is with a datalog of the events described and the system's configuration.

Offline Jorgen

  • Global Moderator
  • Jr. Member
  • *****
  • Posts: 59
  • BHP: 500
Re: Inlet Backfire When Restarting After Stall
« Reply #7 on: August 27, 2007, 03:09:46 pm »
It's an interesting problem. Does this occur only when you restart immediately or also when you restart after 5-10 seconds?

If I recall correctly there is a short delay before the engine goes from "running" to "stopped" status.



Offline Sprocket

  • Hero Member
  • *****
  • Posts: 867
  • BHP: 29
Re: Inlet Backfire When Restarting After Stall
« Reply #9 on: August 27, 2007, 05:11:23 pm »
Data Log

http://freespace.virgin.net/colin.milne/datalog200708271316.xls






All from the same log at different times

Clean starts











Backfire while cranking No reset












Stalling

I generated this situation simply by blipping the throttle, you can see TPS drop to 2% some times 1%, where idle requires 4 to 5%. You can also see that Lambda is on the rich side, ignition advance increases in respect to how the lambda and spark tables have been set to try and overcome this situation.

This is all due to the system going into idle control at about 2000rpm with tps below idle threshold, this is not correct, there is not enough air in the engine, fuel and spark is there, just not the air.










Trigger error rported first at this point

These errors are reset with a hard reset, but i do not know where they happen again as Vems Logviewer keeps the warning there even if you scroll back before the error, a bug in itself??

« Last Edit: August 27, 2007, 05:26:42 pm by Sprocket »

Offline rob@vems.co.uk

  • Hero Member
  • *****
  • Posts: 3115
  • BHP: 49
    • VEMS Forum
Re: Inlet Backfire When Restarting After Stall
« Reply #10 on: August 29, 2007, 01:25:02 pm »
Just a quick question, why is the TPS threshold for IAC up at 10%?  Normally I'm used to seeing 2 to 3%.

This would explain why the IAC is coming on at higher than wanted RPM.

I'll keep digging...

Offline Sprocket

  • Hero Member
  • *****
  • Posts: 867
  • BHP: 29
Re: Inlet Backfire When Restarting After Stall
« Reply #11 on: August 29, 2007, 11:12:53 pm »
It is there because with all load on the engine, lights, heater, and wipers, TPS can reach 6% at idle, normal idle TPS 4 to 5%

On cold start and warm up TPS is about 8% If i set idle threshold to 8% there is no leeway and idle control is not in effect, resulting in a frozen stepper above the threshold, this is with follow TPS enabled. If you dissable follow TPS, on starting the engine the stepper freezes above threshold and engine sits at around 2500rpm.

No matter where the threshold is (As long as its higher than idle TPS)TPS always drops to well below idle TPS when on over run out of over run fuel cut, not above.

If I were to set Idle threshold to 2 or 3 % as you suggest, I would have no idle control and the stepper would freeze, considering that idle TPS is on average 4%. Its confusing, the PID refference table has an effect

Normal cruse TPS is around 12 to 14% TPS


Think about an over run situation without worrying about whatthe idle threshold is set to for a minuit. You lift off the throttle, throttle closes, TPS is below idle threshold, so idle control resumes, yet engine speed is above idle set point. Idle control reduces steps, thus TPS, to try and bring engine speed down to idle set point. Problem is the engine is in over run, not idle. Throttle is totaly closed, deep vacuum is registered, 5kpa ( bad for the 50 year old designe A series oil consumption), engine speed aproaches idle set point, overshoots, yet stepper does not respond quick enough, not enough air in the engine and the engine stalls.

Stepper and idle control works well, espetialy with a high ignition advance in the very low rpm load sites. You can keep your foot on the brake, bring the clutch in and watch the stepper open up the throttle, advance increase, and feel the torque of the engine build up, I could not stall the engine easily this way.

If I recall correctly how rover prevented this situation. As the throttle is released on over run, the stepper would catch the throttle and return to a predetermined set point of roughly 35 steps as the engine speed aproaches idle set point, then if required at this point would adjust from there. With this OEM system you tune the throttle stop to achieve idle speed at this 35 steps, so in theory the stepper should return to idle set point every time. This is one reaon i set the minimum PID refference bin to 40 steps, but unable to see what steps is current, its not easy.

I have tried all sorts of figures in the last bin of the PID stepper reference table, which i would have thought would have been the idle steps set point, yet this situation still exists. I have looked around but cannot find anywhere that reports back the number of steps, again making it difficult to see what is realy going on. A Megatune Gauge would be nice, showing the max number of configured steps on the scale and the current stepper position in relation, % would confuse things. What also would be nice is a warning light at the bottom of Megatune to show when idle control is actualy on, much like the one for EGO and Stepper. though you could assume that if stepper is on that you are in idle control? While Im at it ( again) whats wrong with having a status light for the misc outputs to show its current state, whether its on or off, cant be that demanding on software

There needs to be a configurable point in idle control for steps at idle set point and an idle enrty speed thus preventing the stepper clossing off the throttle at high over run engine speeds

Rover originaly used a switch on the throttle pedal to determine idle condition. The problem was the switch would fail. I recently fixed one of these after A very well known and renowned Mini specialist couldnt fix the problem, the car would backfire, hesitate and run badly, so much so it melted the cat. The cat was replaced for it to melt again within two days!! a complete injection manifold was fitted which included all sensors, the problem still existed. I looked at it and found the problem within 10 minuits, now im not being big headed, but it really does show the skill level of these so called experts. when you moved the throttle the stepper moved to zero and the advance retarded to the point of inlet backfire, incomplete combustion in the exhaust melting the two cats. The owner had spent over £500 at the garage for what, in the end was only a £20 switch. I ripped out the old ECU and installed and rewired for a later version that did away with the switch, and it cost him no more than a pint afterwards. Point im getting at, i can see this  as very similar to the problem im having, though i havent managed to get it to go pop yet, lol. Well i can when restarting it, ooh, now i wonder. :-\

Sorry i do tend to ramble on a bit ::)

Offline rob@vems.co.uk

  • Hero Member
  • *****
  • Posts: 3115
  • BHP: 49
    • VEMS Forum
Re: Inlet Backfire When Restarting After Stall
« Reply #12 on: August 30, 2007, 11:47:56 am »
I'm not sure why TPS would change based on loads and cold starts?

The IACV should be opening up to raise the RPM.

Offline rob@vems.co.uk

  • Hero Member
  • *****
  • Posts: 3115
  • BHP: 49
    • VEMS Forum
Re: Inlet Backfire When Restarting After Stall
« Reply #13 on: August 30, 2007, 12:59:07 pm »
I've put this problem over to the Wiki as more of the development team look there than here, as I'm not afraid to admit, I am stumped on this one.
Maybe its the TBi and wet wall fuelling, maybe its the IACV
http://www.vems.hu/wiki/index.php?page=MembersPage%2FColinSprockets

Offline Sprocket

  • Hero Member
  • *****
  • Posts: 867
  • BHP: 29
Re: Inlet Backfire When Restarting After Stall
« Reply #14 on: August 30, 2007, 07:16:43 pm »
Ah, right, i now know where you are coming from Rob. However i now believe that you are unaware that the stepper is NOT a valve, but a throttle kicker, hence the TPS changing.

One other possible improvemnet to the ALS control is having the stepper as the throttle kicker and configure steps for ALS????


i dont think that the valve would work well on the TBI as the fuel is squirted ontop of the throttle, with the throttle closed and a bypass valve, what would carry the fuel over into the engine???

The Bosch MonoJetronic system uses a throttle kicker stepper also, its almost the same thing as the Rover SPi, just a little more compact.

Hope that helps a little
« Last Edit: August 30, 2007, 07:18:25 pm by Sprocket »