I ran my new engine in, using only actuator pressure (unplugged the solenoid) which created about 0.85bar. The car was incredibly rapid at that boost level bringing comparisons by passengers, of being similar to 350bhp cars they'd also passengered in, with perhaps more low down torque too. Imagine my impatience at the thought of getting some 'real' boost thru it

Unfortunately that has brought nothing but problems The initial attempt was to run about 1.45bar tailing off to 1.2bar at the top end. It seemed quick enough thru the rev range but from 5.5-6,000rpm upwards we got a fairly dramatic missfire. Initially we still had the NGK PFR6 plugs used for running it in with, so we changed them to NGK Racing 8s. It didn't fix the problem.

Later that week we had it on the rollers at PE, technically we had enough fuel and air for 380/400bhp, but only acheived 289b/286t PE diagnosed the problem as Coil packs (which were only 5k miles old anyway) but following another wallet emptying session this failed to solve the problem.

Next task was to check that the fuel was actually getting there, so fuel pressure tests and flow rate tests were performed, again showing that there was know problem there, Pump and FPR presumably working fine.

Running low of possibilities I then regapped the plugs. From a possibly low 0.6mm to around 0.725. Again absolutely no improvement. (Theres a pattern of occuring now!)

So as a last resort this weekend I tested battery performance and created an earthing kit. This has at least made the plug performance more consistent low down/in the mid range (NGK 8s dont like vacuum pressure!) but has made no noticable improvement up at the top where I still get problems.

To protect the car I have re-disconnected the solenoid, to go back to actuator pressure, and although the performance isn't too fantastic (running slightly rich due to map expecting more boost) the top end of the rev range is perfectly clean and smooth again

Thus, I'm looking for any suggestions of any other possible causes of the miss-fire. Is there any chance that the solenoid is duff and causing problems when connected? Any other ideas worth trying are more than welcome.

Otherwise there may soon be a Scoob for sale, that is just one mystery solution away from being a real Supercar!!!!

DIS(Patience running out)CO..

 

have you tried running higher boost with a bleed valve to see if it does it then???

 

I haven't, but its not the daftest idea I ever heard, and would certainly eliminate the solenoid assuming it could make the same boost levels.

 

what ECU and MAP sensor are you using???

 

Motec M48 Pro ECU and Cosworth 2.5bar(I think?) Map sensor .

 

I should mention that I ran 1.4-1.5bar on the old engine, without any probs, which makes it all the more confusing.

 

have you had it mapped for the new engine???

 

Oh yes! I'm daft, but not that daft!

 

LOL

it is a strange one!

 

Have you noticed what the O2 sensor voltage does when the misfire occurs ? (assuming you have one ? ) That could be a BIG clue

Also check the pressure signal line to the FPR, I have had a pipe look ok on the outside braiding but the inner rubber had been damaged during removal. It only showed up at higher boost by leaking the signal air and hence not increasing fuel pressure enough.

Andy

 

TBH I haven't checked O2 voltage. What should I be looking for, and what clues would it give me?
As far as Fuel Pressure is concerned we checked the flow rate at 1.5bar above FPR pressure and nothing was amiss.

 

Well, I've had an interesting day today, the results of which we shall all see on tuesday.

I remapped a car today, then did it again after a fundanental change, from scratch.

One problem we had was a misfire at 5000rpm ish as we hit about 1.5bar. Well, this was the 3rd time we had the misfire, except it was three different misfires, like I said, interesting day.

We only solved the problem in the end by gapping the NGK 7s down to 0.5mm (from 0.7mm) in the end, allowing us to run 'more than 1.5 bar' of boost.

Thing of note, upon misfiring we see a lean reading on the AFR, going from 11.8ish to 13.1 this is due to not all the oxygen being burnt. However you'r first reaction might be to throw more fuel in, but this would make it worse (we didn't do that today though).

During the quest I tried retarding the igntion by about 5 degrees in the areas where it was missing, this made things worse, possibly due to higher cylinder pressures at the time of spark firing.

Disco, have you looked at the motec problems Carlos was describing on 22b? That was put down to interferance on the crank/cam sensors.

Hmm distinct lack of answers, but would try the plugs gapped down to 0.5mm.

 

Couple of things here

Sounds like it might be rich, has it had a proper wide band on it to check it out and what fueling was it running at PE ?

I assume you have uprated fuel pump and reg, that makes for a very "interesting" fuel map slope depending on the pressure you start with and its very easy to get the "heavy end" of the map very over fueled.

Pavlo Pat has been working on Disco's car so I'm sure he has spaced off the sensors.

Boost solenoid ... won't be that.

Gap the plugs down but ... racing 8's are far too cold, you could have well exacerbated the problem by changing to those, stick a set of PFR7B's in, or even better a set of the HKS S35i's (Iridiums), gap at 0.5 and check, if its ok ease the gaps up until you get the misfire then come down by 0.15. You should be able to run 0.7 gaps up to 2.3 bar, thats when mine (HKS) blow out but it is totally dependant on the integrity of the ignition system. I had to run my STi2 wagon with 0.6 gaps on HKS S35i's for this sort of reason.

Ask Pat to check the ign dwell as well, you can help overcome this by adjusting that parameter, I ended up having to replace the coil packs on the wagon when I took the Motec off it, up to that point I had mapped around the problem without knowing it was there.

cheers

bob

 

Thanks for the additional comments etc guys.

I'm running an uprated fuel pump with an RS500 regulator, which has a pressure of 3.5bar. The fueling at PE was OK, up until right at the top where is was actually a little lean, I'll check with Pat if he can remember the actual values.

Bob, do you think its worth gapping the 8's down to about 0.5, or best to wait until some HKS plugs arrive?

Edited to add: Pat adjusted the dwell and we did gain a slight improvement, but it was more of a treatment than a cure.

[Edited by Disco - 5/5/2003 1:31:57 PM]

 

'tis an odd one to be sure! things that we've done so far....

ran engine in with wastegate actuator disconnected, so no boost. swapped oil for a better grade and reconnected actuator, but left solenoid unconnected, resulting in 0.9 bar natural boost pressure. Fuelling at this point went up to 80% IDC on equivalent of 580cc injectors, indicating about 330-350 BHP and it felt like it on the road, it's virtually lagless too This ties up with projected airflow calculations. So far so good, figured we may need a bigger turbo and injectors though!

Ran on the better grade oil for another 1500 miles, then dumped it and went for a fully synthetic oil, and reconnected the solenoid, pushed the boost up somewhat Targetted about 1.4 through the midrange, tailing off toward the top to give the poor little injectors a fighting chance. Engine was misfiring badly, so we swapped the plugs, which seemed to resolve the issue on that night. Lambda readings seemed sensible.

On the rollers the fuelling looked a little lean, and it seemed to misfire around about 6000 RPM, it's clearly visible on the power plot. Initial thoughts were with the coil packs, so we replaced them, but to no avail. Tried adjusting the dwell time from silly short to silly long with little improvement (seems happier at 4.5ms). Tried adjusting fuelling round the misfire region, an extra 10% seems to improve the situation a little but not that much. Figuring that it could be ignition related we attached a set of cans to the car and went out to play with the timing.

Found a little more torque in other areas of the map, but in the misfire region, the situation does not seem to be affected by timing, neither adding nor pulling timing seems to help.

With regard to the Subaru coil packs, we found them to operate satisfactorily up to 2.75 bar without missing a beat on the bench dyno using the NGK Racing 8 plugs. I ran the NGK Racing 8s on this engine previously in my car without a problem... they don't like vacuum that much but that does give lovely blue cones of flame out the exhaust on the over-run even without ALS I ran up to 1.5 bar on my car previously without a problem, however that was on a Pectel not a MoTeC and it was a different car, and the heads have been breathed upon since then.

Our attention was then drawn to the fuel system, perhaps the IDCs were high not because the heads were flowing well, but because the fuel pressure was falling away? Dutifully connected a fuel pressure gauge to the car and took it out for a run... result was that the differential fuel pressure was pretty much constant, so the fuel really is getting in there, and according to the gas analyser it isn't quite enough So the projected airflow and the actual fuel flow all seem to tally up...

At this point there don't seem to be many possibilities left, perhaps the ignitor module is a little duff. Perhaps the heads are affecting the mixture preparation detrimentally (but the combustion space is based on a known-good space). I've noticed odd behaviour with EJ heads in the past, whereby if one uses a 290 degree BTDC injection end angle around 2500-3000 RPM it causes misfires, so naturally I've played about with the end angle to see if that helps, but to no avail.

Another interesting observation is that the previous engine in the same car did not suffer from the misfire problems, but it also seemed to suffer from poor thermal efficiency, it also made less power and less torque than one would have predicted from the air and fuel flow... at least that seems constant between the two engines. Perhaps a poorly ignitor module could be to blame for a weak spark, which isn't lighting the mixture aggressively enough, losing power, and ultimately at the higher boost isn't able to light it up at all. Although the previous engine didn't misfire at the same boost, it was half a ratio lower, so the cylinder pressure at the point of firing would have been lower.

Guess I'll have to attach a DSO to a coil to see what the voltage waveform is implying... ie is it firing just not lighting the fuel, or is it just not firing at all? Swapping the module is easy enough , and it may be worth trying a CDI system to fatten up the spark a bit

Of course, any other ideas most welcome, it may just be a case of not being able to see the wood for the trees....

Cheers,

Pat.

 

One thing I cant get my head around is if the plugs are 'too' cold, I would have thought them to cause more problems at 0.9bar than when trying 1.4-1.2bar?

 

If the plugs start fouling then the effect will be felt more at higher boost as the voltage required is higher, this makes it more likely to track away down the insulator.

 

I read with interest. Different type of engine, but I have also had some baffling ( and still may do when I get the car going again ) missfire problems, that have pickled my brain sumthin shocking.

 

Same here...

...misfire coming onto boost at peak torque. However I think original causes were general electrical goodness and that made it all go away.

Now it seems to be fueling related - or something electrical breaking down over time that the other changes are merely masking.

One culprit could be the igniter block.

Overall a bloody frustrating time!

Rannoch