Car won't idle - the saga continues

[Steve]

Bugger

[JustGav]

I've got my Ultimate which I'm flogging, which you are welcome to try see if it settles anything down.

[Steve]

Would a dodgy ECU connection cause this sort of problem (not reading a sensor properly).

Is worth doing a refurb on the ECU connections?

 

Ian did a nice write up on this in the Tech FAQ's bit.

[Homer]

It sounds like a sneaky little elec problem in your loom/coil packs/earths or possibly some sort of corruption in the EManage.

 

What's the stock O2 sensor putting out? I know you're in the 10's but is the sensor actually reading rich voltage - just wondering if it's trying to tell the ecu to richen all the time as it's reading lean?

 

What would the stock injector cold idle pulsewidths be, anyone have any ideas?

 

Edit: Does the Emanage blue alter duration basd on battery voltage it's seeing? If so can that be disabled for testing purposes - extra duty could be being added if the emanage isn't seeing 12volts.

 

Hmm, you are probably right about the electrical problem. The coil packs have been tested and appear okay. Likewise the o2 sensor has been replaced (as part of elimination). I will order a brand new one off Nic anyway, just to make certain this is not the cause.

 

Not sure about you other questions though, they're ones for the experts

[Homer]

I've got my Ultimate which I'm flogging, which you are welcome to try see if it settles anything down.

 

Sorry Gav, missed your eariler post. Thanks for the offer mate, I'll be in touch after I've had a chat with TF on Monday. I might well take you up on it (thinking AEM also at the moment!).

[Robert]

One thing that was giving me dodgy signals on my emanage blue was the 02 sensor connectors, where the 02 sensor plugs in to the main wiring loom the connectors has become loose, try tightening them the same way you would with the coil pack wires as per Ian C's method, that helped me, but if im honest the emanage blue just wouldnt work on my single turbo aristo with 650cc injectors, dont ask me why but as soon as i went with the AEM it worked perfectly.

[JustGav]

Sorry Gav, missed your eariler post. Thanks for the offer mate, I'll be in touch after I've had a chat with TF on Monday. I might well take you up on it (thinking AEM also at the moment!).

 

I didn't mean buy it... although you are welcome to, I meant do you want borrow it to see if it works better than the blue? I'm not 100% sure but I believe you can plug it directly into a blue loom, and then switch some jumpers over to enable addition only rather than normal mode.

[Homer]

One thing that was giving me dodgy signals on my emanage blue was the 02 sensor connectors, where the 02 sensor plugs in to the main wiring loom the connectors has become loose, try tightening them the same way you would with the coil pack wires as per Ian C's method, that helped me, but if im honest the emanage blue just wouldnt work on my single turbo aristo with 650cc injectors, dont ask me why but as soon as i went with the AEM it worked perfectly.

 

Thanks Rob, I've not checked the socket connector itself, but will try that tomorrow. Theory says it could be the o2, but i'm not sure it would cause such a massive overfuel. Worth a shot though

 

There may be an issue with the EMB and these injectors, but it would not cause this particular issue. It was running fairly well before (though thats a whole different topic!).

[Robert]

some electrical contact cleaner wouldn't go a miss on the connectors aswell.

[absz]

any news or updates on this ?

[Homer]

any news or updates on this ?

 

None yet mate, I've been too busy at work to do anything on the car (and it'll remain that way for a good few weeks.. off travelling again next week too). Really need to just get it to a garage that offer an end to end solution for these sort of issues... didn't get chance to call Turbofit today but will try tomorrow.

[Kranz]

I've been having a think about this Darryl and it all keeps coming back to the injection system or injectors themselves. I'm thinking that at least one injector has a mechanical problem.

From the vid it seems that its not firing on all cylinders, possibly due to one or more cylinders being too rich to fire. This would be consistent with injector(s) stuck open and hosing fuel into the cylinder(s). This would cause the AFR's to go ballistic.

 

Were the injectors new?

Were they flow tested immediately before installation?

 

Have you tried giving them a light tap while running to see if one will free off? (Not sure if they can be reached?)

[dandan]

Darryl said he is seeing close to 7.0ms pulsewidth and it should be closer to 1.0ms. That suggests an electronics/electrics/earth issue to me.

 

The injectors are actually being fired for too long causing an over rich condition.

 

btw I find a controlled blowtorch application a great way of rejuvinating wetted pulgs...

[rpslaughter1982]

Homer, I have had a very similar problem to you with my recent single upgrade.

The first time I started it up it ran fine, but after about the fourth start up it wouldn't idle, i had backfires, then it wouldn't start at all.

I have gone through a lot of posts looking for the solution and looking over your thread i think it is worth getting new spark plugs again.

 

When you changed the plugs before it started and again and then over a short amount of time stopped working.

 

On my car this was due to overfueling, it stopped the plugs working within 4 starts!!!! I took them out dried/cleaned them and put them back in but still no joy. Then I read that you cannot clean iridium plugs!

 

I put my old plugs back in and it started first time.

 

I know you dont think it is the plugs now but I really think it's worth trying new ones again, but don't run the car until it is mapped. Just see if it starts then wait till you can sort out the overfuelling.

I now have a better map for my car (not overfuelling) and the problem has gone. Over fuelling can kill iridium plugs in a matter of hours!

Good luck

 

Rich

[Ian C]

Well the problem is he's got a base map that should cheerfully run the car off boost as well as running safely rich on boost. It certainly doesn't explain the madness where it runs 6ms duty cycle at idle, and also doesn't explain the previous behaviour where it ran too rich on boost and too lean on closed loop. This schizophrenic behaviour is bizarre, which is why I've suggested going back to a set of 440cc injectors and seeing if the symptoms persist or not...

 

-Ian

[rpslaughter1982]

Well the problem is he's got a base map that should cheerfully run the car off boost as well as running safely rich on boost. It certainly doesn't explain the madness where it runs 6ms duty cycle at idle, and also doesn't explain the previous behaviour where it ran too rich on boost and too lean on closed loop. This schizophrenic behaviour is bizarre, which is why I've suggested going back to a set of 440cc injectors and seeing if the symptoms persist or not...

 

-Ian

 

I know you know what your talking about Ian and I can't say that I can explain the other behaviour. My post was from my experience, my base map was crap and well off what it should be. Just thought it may be worth a try again, its a half hour job to change the plugs, longer for the injectors, and after going through similar start up problems it was the plugs that fixed mine. I hope Homer gets it fixed soon, whatever the solution, there's nothing more frustrating than a poorly supra!

[Homer]

First off, thanks everyone for your replies, it's been a huge help so far in eliminating items.

 

I've been having a think about this Darryl and it all keeps coming back to the injection system or injectors themselves. I'm thinking that at least one injector has a mechanical problem.

From the vid it seems that its not firing on all cylinders, possibly due to one or more cylinders being too rich to fire. This would be consistent with injector(s) stuck open and hosing fuel into the cylinder(s). This would cause the AFR's to go ballistic.

 

Were the injectors new?

Were they flow tested immediately before installation?

 

Have you tried giving them a light tap while running to see if one will free off? (Not sure if they can be reached?)

 

I've just pulled the plugs and every one was absolutely caked in soot (as you'd expect!), as every one is sooted I think this rules out an individual injector sticking. I've heard a car with a stuck open injector and it sounded nothing like this or put out as much smoke.

 

The injectors were 2nd hand (from Terry S), but had never been used. Terry said he had them flow tested previously and were fine. I did not flow test them prior to installation.

 

Homer, I have had a very similar problem to you with my recent single upgrade.

The first time I started it up it ran fine, but after about the fourth start up it wouldn't idle, i had backfires, then it wouldn't start at all.

I have gone through a lot of posts looking for the solution and looking over your thread i think it is worth getting new spark plugs again.

 

I just changed the plugs again but it made no difference. I think the previous time it mirculaously started working was just luck.

 

Well the problem is he's got a base map that should cheerfully run the car off boost as well as running safely rich on boost. It certainly doesn't explain the madness where it runs 6ms duty cycle at idle, and also doesn't explain the previous behaviour where it ran too rich on boost and too lean on closed loop. This schizophrenic behaviour is bizarre, which is why I've suggested going back to a set of 440cc injectors and seeing if the symptoms persist or not...

 

-Ian

 

As Ian says, this particular issue (ultra rich startup) cannot be the map. I don't think there is any way to even make an EMB put this much fuel in on idle!

 

The 440's are back with me know (cheers Mark), I just need to find some time to refit them and try again.

 

Ian/others, can a faulty injector make them all suddendly start putting in too much fuel? That doesn't seem to make sense to me.

 

I'm more certain now its an intermittant electrical problem. I just don't have the skills or knowledge to go any further...

 

Turbofit didn't call me back today either

[Ian C]

I just changed the plugs again but it made no difference. I think the previous time it mirculaously started working was just luck.

 

 

I too think that it was possibly coincidence. Trouble is, once the plugs are sooted up they are a nightmare as they won't run the engine properly in much the same way as an overfuelling problem. So if the intermittent problem clears up, the engine will still run like a bag of nails. Changing the plugs at this point can look like a miracle cure, but it's not. The Ultimate has a fantastic plug cleaning feature which actually works, shame the Blue hasn't got it

 

As Ian says, this particular issue (ultra rich startup) cannot be the map. I don't think there is any way to even make an EMB put this much fuel in on idle!

 

That's actually a good point! I think if I added 50% airflow and 100% duty I might get it from 2ms to 6ms but I'm quite sure I'd have noticed that typo by now

 

The 440's are back with me know (cheers Mark), I just need to find some time to refit them and try again.

 

Ian/others, can a faulty injector make them all suddendly start putting in too much fuel? That doesn't seem to make sense to me.

 

I'm more certain now its an intermittant electrical problem. I just don't have the skills or knowledge to go any further...

 

I don't think a faulty injector can do that. There is no chance all six are bad. It boils down to the stock ECU sending a ridiculously high duration at idle. You know the only time you get durations like that is at cold start, which is scarily similar to what Smarty is experiencing at the moment Perhaps you'd like to investigate your water temp sensor as well? Measure the resistance across pins 65B and 44B of the ECU plug when you've got some heat into the engine, and check the voltage on 44B while it's running if you can (you can knock up a probe with a safety pin )

 

-Ian

[Homer]

I don't think a faulty injector can do that. There is no chance all six are bad. It boils down to the stock ECU sending a ridiculously high duration at idle. You know the only time you get durations like that is at cold start, which is scarily similar to what Smarty is experiencing at the moment Perhaps you'd like to investigate your water temp sensor as well? Measure the resistance across pins 65B and 44B of the ECU plug when you've got some heat into the engine, and check the voltage on 44B while it's running if you can (you can knock up a probe with a safety pin )

 

-Ian

 

I was just ready Smarty's thread and thought it sounded a little similar.

 

I've tested the resistance of both the old and new temp sensors (did I mention I replaced it with the one from the car originally?) and they were both, at cold, showing perfect resistance on the sensor itself. I didn't test it from the ecu plug so will make that check asap tomorrow.

 

I can't get the car warm as I'm not happy running it for any time with this amount of fuel going though (it's over fueling 10x what it should do).

 

Thanks again Ian

[dandan]

Another £0.02...

 

Does anyone know what the pulsewidth is for cranking/starting in the OEM ecu and can that be altered?

 

Often when starting the injectors are not fired sequentially and also the pulsewidths are higher than idle to get the bloody things started. I wonder if a dodgy crank sensor signal (giving rpm in the 200rpm range) is making the ecu think you are cranking and giving you big pulsewidths. DOes the EMB trim back cranking pulsewidths to allow for bigger injectors? If not these two things would definitely not be helping.

 

I can't believe this could be a coolant temp sensor issue - the pulsewidths are massive compared to what they should be but can you subtract injector duty cycle from the warm up period of the map only to verify if this changes anything...

 

Basically I was just wondering if your ecu thinks your cranking (or maybe even running very very cold - I don't like this theory though) and is giving you the pulsewidths it thinks are needed. Maybe the EMB could be an advantage here as you can use it to check different things out.

 

Anyway, I'm still thinking what else could give you symptoms......

[Homer]

Basically I was just wondering if your ecu thinks your cranking (or maybe even running very very cold - I don't like this theory though) and is giving you the pulsewidths it thinks are needed. Maybe the EMB could be an advantage here as you can use it to check different things out.

 

Thats a good point. One thing I noticed during a full cold start is that the initial startup during the first few seconds of startup (when it runs okay) is that injector duration is around the 6-7ms range.

 

Just thinking out loud, could this be a firmware issue on the EMB?

 

Also, and perhaps this is a flawed assumption, but if the EMB is not pressent the car should only run 100% more duration and duty (since the injectors are only slightly less than twice the size of stock ones). If it's not an EMB issue then the car should idle FAR better than it is now. When the injectors were first installed with a zero'd map it would idle nicely, even though it was very rich.

[dandan]

Can you verify what rpm your oem ecu is seeing - presumeably it's identical to what the emb is showing? Is it correct when idling badly or is it down below 400rpm?

 

I would try it without the emb and see if there's any difference - obviously it'll only be a differecne you can hear/smell (rather than see on a screen).

 

Anyone know what temp the ecu is fully out of any warm up enrichment?

 

Also Homer - have you been fiddling anywhere near your crank sensor or even ruled it out with a swap?

[Ian C]

DOes the EMB trim back cranking pulsewidths to allow for bigger injectors? If not these two things would definitely not be helping.

No it doesn't, but starting has never been an issue on any E-Manage equipped car I've been in contact with, from 440's to 850's.

 

I can't believe this could be a coolant temp sensor issue - the pulsewidths are massive compared to what they should be but can you subtract injector duty cycle from the warm up period of the map only to verify if this changes anything...

 

I can believe it is, I in fact see 6+ms of duty on the initial couple of seconds cold start, and that only slowly diminishes over the next minute or so to 3ish ms as the engine warms up. That's why I'm thinking of this as the culprit at the moment, it's the only thing that can give big enough changes to the idle duration via the stock ECU. If it thinks it's at -20degC then ti surely could run up to 8ms duration...

 

One thing I noticed during a full cold start is that the initial startup during the first few seconds of startup (when it runs okay) is that injector duration is around the 6-7ms range.

 

As agreed above

 

Just thinking out loud, could this be a firmware issue on the EMB?

 

Don't think so, I've seen most of the firmware variations and they haven't ever had a noticeable difference.

 

Also, and perhaps this is a flawed assumption, but if the EMB is not pressent the car should only run 100% more duration and duty (since the injectors are only slightly less than twice the size of stock ones). If it's not an EMB issue then the car should idle FAR better than it is now. When the injectors were first installed with a zero'd map it would idle nicely, even though it was very rich.

 

That is flawed, your injectors would see stock durations but deliver near twice the fuel in that time because the holes in them are near twice as big I think that's what you meant though? The EMB isn't clever enough to make up it's own duty cycles either, it can only report on what it sees off the stock ECU.

 

Taking the EMB out of the loop fully would remove all your datalogging abilities, unfortunately, but zeroing the map is a valid test. It the EMB is indeed problematic even with a blank map then trying another one on loan would help, but I think that's the wrong direction to go at the moment. Maybe if the 440's still gave problems then removing it from the loop would be a good next step though.

 

Can you verify what rpm your oem ecu is seeing - presumeably it's identical to what the emb is showing? Is it correct when idling badly or is it down below 400rpm?

 

The EMB taps off the same RPM line as the stock ECU, without cutting it or altering it so it should be spot on.

 

Anyone know what temp the ecu is fully out of any warm up enrichment?

 

According to my oil temp gauge, about 48degC at the filter No idea what water temp that is though.

 

-Ian

[dandan]

No it doesn't, but starting has never been an issue on any E-Manage equipped car I've been in contact with, from 440's to 850's.

 

That's interesting but if (and I'm only throwing ideas around) Darryl's car thinks he is still cranking it could be massively overfuelling as a result.

 

I can believe it is, I in fact see 6+ms of duty on the initial couple of seconds cold start, and that only slowly diminishes over the next minute or so to 3ish ms as the engine warms up.

 

Aha some numbers - I'm assuming that's with (bigger) non 440 injectors as well Ian yes? If so then those figures are definitely in the ballpark Darryl is seeing. Progress?

 

The EMB taps off the same RPM line as the stock ECU, without cutting it or altering it so it should be spot on.

 

Good - makes for an easy test of rpm then although I'm sure Darryl would have spotted this if it was very low so maybe that's cranking pulsewidths ruled out altogether?

 

According to my oil temp gauge, about 48degC at the filter No idea what water temp that is though.

 

Does anyone else know in terms of water temp when the stock ecu is out of warm up?

 

How about comparing Darryl's coolant sensor output compared to a good working car both at the sensor and at the ecu - hopefully checking both the sensor output and the wiring?

 

Darryl - what is the idle like the instant it starts, is it ok for maybe 20secs or is it rough straight away? (This needs to be with good non fouled plugs)

[Homer]

Can you verify what rpm your oem ecu is seeing - presumeably it's identical to what the emb is showing? Is it correct when idling badly or is it down below 400rpm?

 

I'm not sure how to check, but what I see on the rev counter is more or less exactly what I see on the EMB data log.

 

Also Homer - have you been fiddling anywhere near your crank sensor or even ruled it out with a swap?

 

Yes and no, it was fiddled with during the conversion (unavoidable really). Phoenix did find the original one was faulty as it was intermittently reporting a zero value. It was replaced with a known working one. This was thought to be the cause of the last break down which is why I paid a large (but fair) sum of money and took the car away.

 

How about comparing Darryl's coolant sensor output compared to a good working car both at the sensor and at the ecu - hopefully checking both the sensor output and the wiring?

 

Darryl - what is the idle like the instant it starts, is it ok for maybe 20secs or is it rough straight away? (This needs to be with good non fouled plugs)

 

I have two temp sensors, one from my old car which was working fine, the other (which I installed after the conversion) from a scrapper. I have tested the resistance of both sensors on cold and they both reported very similar figures. I swapped my old one back in a couple of days ago anyway, but it didn't make any difference.

 

It takes a little while to fire up (cranking for around 4 seconds), revs rise for a split second then fall to a steady (though low) idle. It performs exactly like stock, except the massive over fuel.

 

 

 

Ian/all, I'm not convinced by my bodged temp sensor wiring after the old harness fell apart so am waiting for some new parts from Steve Manley. Once these arrive and I've test fitted them I'll let you know the result.

 

With the mail delay on the new parts, Supra Pod this Saturday and me heading off to Sweden next Wednesday I probably won't be able to come back with any results for a couple of weeks.