Jump to content


Photo

Spi Won't Fire Without Throttle Input


  • Please log in to reply
18 replies to this topic

#1 WafuMini

WafuMini

    Learner Driver

  • Noobies
  • PipPip
  • 18 posts
  • Location: Somerset

Posted 02 March 2018 - 11:51 AM

Hi all,

I've recently developed a problem with starting my SPi. When starting from cold it turns over, but won't fire unless I give it some gas on the throttle. I then have to hold the revs up at about 2k for a couple of seconds or it will cut out.

This problem suddenly emerged after I did a bit of work on her to get her MOT ready. Before she would start easily everytime. Work done was:

- Plugs changed (same type as before)
- New HT leads
- Valve clearances reset
- Cat refitted
- Rust cut out and plates welded in between wings and front panel on both sides
- Stepper motor reset

I've fully charged the battery and it's definitely turning over fast. I can hear the fuel pump priming before start. If I run the engine up to temperature then it starts with no throttle required.

Any ideas what could be causing this? Is it a problem I have created with some of the work I've done recently?

#2 Sprocket

Sprocket

    Great on Injection faults

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 7,266 posts
  • Location: Warrington
  • Local Club: Manchester Minis

Posted 02 March 2018 - 12:00 PM

Please explain "Stepper Motor Reset"



#3 WafuMini

WafuMini

    Learner Driver

  • Noobies
  • PipPip
  • 18 posts
  • Location: Somerset

Posted 02 March 2018 - 08:41 PM

Ok I see how that was unclear. What I mean is I have reset the TPS (ign. 2 and pump 10 times). Not that I have used a Crypton ACT or anything to set the stepper motor.

#4 FlyingScot

FlyingScot

    Up Into Fourth

  • TMF Team
  • PipPipPipPipPipPipPipPipPip
  • 2,660 posts
  • Location: Inverclyde Scotland
  • Local Club: Mini Cooper Register

Posted 02 March 2018 - 09:02 PM

That wont reset the ECU or throttle position sensor (TPS) on later ECU.
Is it an early SPi with throttle switch for idle ? Or later with TPS only.

FS

PS I wouldnt include your email address in your profile... theres enough spam around...

Edited by FlyingScot, 02 March 2018 - 09:04 PM.
Spelling


#5 WafuMini

WafuMini

    Learner Driver

  • Noobies
  • PipPip
  • 18 posts
  • Location: Somerset

Posted 03 March 2018 - 08:38 PM

Thanks FS. Now removed, hadn't realised it was in there.

 

It is a later SPi without the throttle switch, just the TPS. I thought that it indexed the stepper motor?



#6 WafuMini

WafuMini

    Learner Driver

  • Noobies
  • PipPip
  • 18 posts
  • Location: Somerset

Posted 11 March 2018 - 10:00 PM

The other day I went out for a drive and found that it started absolutely fine with no throttle input from me. I couldn't believe my luck that it had fixed itself. So after a little bit of driving to warm her up, I pulled over to try again. Once more started perfectly.

It was clearly just a momentary thing as for the last few days it has been back to not starting without a bit of throttle pedal applied.

I am at a bit of a loss with this and not sure where to start digging. Is this likely to be a fuel flow problem, ignition, or maybe something to do with TPS/stepper? As I need to put some pedal in to get it to start I had assumed it was TPS/stepper, but once running it seems to be fine.

Please help, I don't know where to begin.

#7 Sprocket

Sprocket

    Great on Injection faults

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 7,266 posts
  • Location: Warrington
  • Local Club: Manchester Minis

Posted 11 March 2018 - 11:35 PM

Too many variables and not enough information to even start to make an educated guess. My advice would be to work out how to get it connected to a diagnostic of some sort. If you can get hold of the parts required to make the required USB cable, MEMS-ROSCO software program will do everything required to help you get the right diagnosis.

 

http://alum.wpi.edu/...c-protocol.html

 

https://github.com/L...sco/find/master

 

 

PS, please remove your email address from the 'Location' field in your user profile, put your general location in there so people can see roughly where you are in this lovely world of ours. or leave it blank if you want to.



#8 WafuMini

WafuMini

    Learner Driver

  • Noobies
  • PipPip
  • 18 posts
  • Location: Somerset

Posted 03 April 2018 - 06:44 PM

Ok so I have now built myself a diagnostic cable and got MEMS Rosco working (took a little while but as always a quick look through TMF and I found all the information I needed  :D ). I've attached the results from today.

 

As before it wouldn't start on its own, but a few blips of the throttle and I managed to get it to fire. Had to keep a bit of throttle on initially otherwise the revs fall too low and it stalls. You can see where I did this on the log files.

 

I'm beginning to think this is something to do with fuel delivery during the start sequence. I know that the fuel pump is working and it kicks in for the initial 10 sec prime before start. Could it be that it isn't generating enough pressure? Or a blocked filter maybe? I've had a good look and can't see any leaks anywhere that could be causing it to lose pressure.

 

My other thought is that it may be that the ECU isn't switching the main relay over until I move the throttle and so there is no spark. Is this possible? 

Attached Files



#9 genpop

genpop

    Speeding Along Now

  • Members
  • PipPipPipPipPip
  • 472 posts
  • Location: Poppenhausen

Posted 03 April 2018 - 07:27 PM

@wafumini

hi, was this logfile made with mems-gauge or with mems-rosco (github.com/leopoldg)

it holds too less information.



#10 Sprocket

Sprocket

    Great on Injection faults

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 7,266 posts
  • Location: Warrington
  • Local Club: Manchester Minis

Posted 03 April 2018 - 10:01 PM

I can't read the file at this end. I believe you have used MemsGauge to generate the logs. Can you use Mems-Rosco instead?



#11 gdudu

gdudu

    Stage One Kit Fitted

  • Noobies
  • PipPipPip
  • 54 posts
  • Location: Marseille

Posted 04 April 2018 - 06:05 PM

Problem with lambda on this file (always at 1275) or on the car ?



#12 WafuMini

WafuMini

    Learner Driver

  • Noobies
  • PipPip
  • 18 posts
  • Location: Somerset

Posted 04 April 2018 - 06:50 PM

Ok my mistake. Made with MEMSgauge. Now have MEMS Rosco. I'll go and do a run with that. As an aside how do you look at log files from Rosco? It seems to be in a different format than the log files that came with MEMS Analyser.

 

I noticed the constant lamda figure too. I think it is the log file, but we'll know soon when I do another run.

 

Thanks all



#13 genpop

genpop

    Speeding Along Now

  • Members
  • PipPipPipPipPip
  • 472 posts
  • Location: Poppenhausen

Posted 05 April 2018 - 05:55 AM

mems-analyzer can read both formats!



#14 Sprocket

Sprocket

    Great on Injection faults

  • Members
  • PipPipPipPipPipPipPipPipPipPip
  • 7,266 posts
  • Location: Warrington
  • Local Club: Manchester Minis

Posted 05 April 2018 - 11:56 AM

mems-analyzer can read both formats!


Mems analyser also crashes when trying to read memsgauge file. Mems analyser generates a hex file while mems gauge generates a delimited file. The issue I see with memsgauge file is the descriptor names are all different from the mems roscoe file. In effect there are three types on log file for three different software programs. I even went to the trouble and edited the descriptor names and tried to read it with mems rosco, but it still crashed. I didnt try mems analyser though.

#15 WafuMini

WafuMini

    Learner Driver

  • Noobies
  • PipPip
  • 18 posts
  • Location: Somerset

Posted 08 April 2018 - 07:31 PM

Ok so I have made a log file with Mems Rosco this time. Still can't get it to open in Mems Analyser for some reason (I get a type mismatch error), but I have managed to look at some of the values by importing them into Excel. What is showed is two starts in quick succession then a period at idle while I let it warm up. At the end I revved it a bit just to show response. I would have taken it out for a drive, but there is a lot of flooding around here in the SW and I don't think taking it for a swim would help the problem.

 

From what I can see, no error codes and all the sensors appear to be working as expected.

 

What I did notice is that when I turned the engine over to start MEMS Rosco lost comms with the ECU meaning I had to press 'Disconnect' then 'Connect' to get it to reconnect. You can see this as a gap in the timings in the file. Also - assuming that IGN is logeed as 16 and CRK is something else - then there are no data entries from when they key is turned to CRK. Any idea why it would be doing this? It would obviously be very helpful to have the actual start logged so I can look at it. I guess if there is some sort of problem with power going to the ECU while cranking then this could be stopping the ECU from energising the main relay? But why would blipping the throttle fix this?

Attached Files






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users