scoobyboothy

iTrader: (1)

here's the graphs showing sudden drop in Ignition timing

scoobyboothy

iTrader: (1)

here's another one, ignition timing is 2nd one down and throttle position is 5th



you can just make out the serie of dips with constant throttle

dynamix

iTrader: (3)

what are you trying to show ?

dynamix

iTrader: (3)

seems to me the problem is with the maf

scoobyboothy

iTrader: (1)

i was trying show that i sometimes get random drops in ignition timing, it was splitpin that suggested to compare it against throttle position,
so you would say that the maf is suspect, would this cause the random drops.

off track slightly, my exhaust seems to soot up quite a bit, could this be a sign of over fueling, if it is, what O2 voltage should a averaging on the stock lamda?

dynamix

iTrader: (3)

if the maf voltage is jumping around - which it looks like it is from those logs then it will be flitting between load cells and timing set in those cells may be on the extreme ends of the results you are seeing.

try swapping in a known good maf sensor.

scoobyboothy

iTrader: (1)

so it shouldn't flit about like that, even if it's not by alot.
cheers for the advice mate

dynamix

iTrader: (3)

no it shouldnt. It is flitting a lot - 0.5v jumps are a lot.

Splitpin

Boothy - in your original post on this subject on the last page you said the timing was "dropping to zero or as near as dammit". In the log snippet above, you're on continuous low throttle (0.82 volts equates to somewhere between 8-10% opening angle) and the timing is dropping one degree - it's going from 31 to 30 and back up again. You are talking about a variance there of one degree so this isn't a big drop at all in the great scheme of things. If this is what you were worried about in the first place, there's no need to be.

What does look a tiny bit odd in that first one is the behaviour of the engine speed readout which is a bit more up and down than one might expect from the constant throttle position. Did you have a slipping clutch or something like that? It doesn't look obviously dodgy enough to worry about, just a bit strange. Also if you look carefully up the graph, each of the 1 degree timing pulls coincides with a small trough in indicated RPM. It looks as though, as Duncan has already suggested, it's flicking between adjacent cells of the map and making a small timing adjustment as a result.

If I were you I would stop worrying about this ignition timing thing in and of itself as, as Duncan suggests, the timing changes look like a symptom of another problem:



The MAF voltage there looks more than a little "spiky", in particular at around 3.5 seconds where it drops from about 3.6 to around 3.25. Once you turn that into a scaled value it represents a drop of around 40% airflow as far as the ECU and the load calculation is concerned. The timing maps are load indexed so any airflow measurement error will knock into the timing lookup.

I see from your posting history that your car had a front-end shunt over the winter and you've also bought a secondhand airflow meter in the last few months. If you're running either at the moment, if I were you I'd fit a brand new one and see what difference you notice in your logs.

Also, (and apologies if we've already covered this) would it be correct to assume you're running the standard airbox and induction tract inc resonator?

Splitpin

Oh - further question Boothy. Do you have a boost pressure reading for that second log you screenshotted from RomRaider, the one I was talking about earlier?

Would be handy to see whether there was any oscillation in boost that corresponds with the airflow peaks and troughs.

scoobyboothy

iTrader: (1)

cheers splitpin, you know what its like, once that seed of doubt has been planted. not had a chance to try the other maf as yet, my car has a fmic witch was on when i bought it, so it has a cone filter fitted. the graph with the funny engine speed must have been a one off cos all the others are smooth

i'm going on a power run tommorrow at area 52 with mid scoobies, so they should tell me if anything looks untoward

cheers again for the info guys

scoobyboothy

iTrader: (1)

the boost seems ok in all the logs that i have done, not spikey in any way just smooth

Splitpin

Has it been mapped for these mods? It's difficult to make out for certain due to the small pic and highly compressed y-axis, but the target manifold relative pressure line in the DD log you've posted looks exactly where you would expect on a standard MY99 UK car.

If so then the front mount and cone filter are unnecessary at this level of tune and aren't doing the car any favours. Indeed the presence of the induction kit raises all the usual problems for the airflow sensor. What make is the cone, and is it bolted solid into the wing (or connected to the engine via a hard pipe)?

Again it's difficult to tell for certain given the small pic and compressed scale but there is what looks like a little low frequency oscillation in your DD log from the 8 to around 9.5 second mark. Looks like it goes up a bit before dropping a bit, and then drops consistently back to the slightly lower level at around 11 seconds. Can you screenshot that specific area of the boost graph with an expanded y-axis so we can get a better idea of what the numbers are?

Splitpin

Oh - one more thing. That DD log. What's with the huge downward MAF voltage spike at around 18.2 seconds? It comes about half a second after you start to lift the throttle and the voltage then climbs straight back up to a steadily reducing trend. Have you got a dumpvalve delete, or an aftermarket VTA DV with a very stiff spring or something?

scoobyboothy

iTrader: (1)

just a bit of an update from the RR day at area 52 yesterday, and it's not good, it's running only 175bhp. Apparently it's massively over fueling, witch is robbing the power! it's been suggested that the maf is at fault aswell as an air leak somewhere. I have changed the maf to a second hand one i got off here, and checked the O2 readings (albiet on the stock sensor) and they are more around the .9v, 1v reading idling, and drop a bit on boost to around the .5v area. but the maf voltage readings are a bit similar and idles similar to the old one. the boost is on the low side at about 11psi. so now checking pipes on the pressure side for tightness and may loose the hks bov it as it seems a bit loose in the mount.

scoobyboothy

iTrader: (1)

bump

Splitpin

That's not bad for a Justy mind. It is a Justy you have, yes?

When you say "massively", what sort of AFRs are you talking about? Give us a number.

Faulty MAF sensors almost invariably under-read under boost, meaning the mixture goes lean. An air leak before the turbo will also lean the mixture off by allowing unmetered air in, so in all of the above circumstances, you'd expect it to be lean.

About the only way an air leak will richen the mixture up is if it happens after the turbo - so look at your compressor to intercooler y-pipe (especially if you've got the sh*tty plastic one), and the dumpvalve.

Secondhand airflow meters are best avoided, purely on the grounds that you can never be certain that they're 100%.

Just so we're absolutely clear on this, are you saying that your O2 sensor is reading around 0.5 volts on boost? Are you absolutely sure about that? If so, and assuming your oxygen sensor isn't knackered (which is possible), that suggests your engine's running lean under boost, not rich.

Also, you say it gives an O2 voltage of 0.9-1v at idle. Does it sit there constantly at around that value or does it oscillate between that high point and around zero volts?

That earlier DeltaDash log looked okay - albeit with the caveat that the Y-axis is so compressed it's difficult to make out the exact numbers it's hitting. Also see the questions I asked in the last couple of posts. Answers to them would help us help you.

Re the HKS, if you have a standard recirculating dumpvalve available, try it.

Saint AAI

Have you got an aftermarket FPR on? If so, what type is it and what's the base pressure set to?

scoobyboothy

iTrader: (1)

I'm not sure about the fpr, there's nothing that looks different, i can't give you afr's from the rr, cos they were going by the high speed lambda in the exhaust and romraider only logs correction, which gets as low as -25% (i think it's %). I've got more logs floating around than seven trent! Lol so i'll do some fresh short ones and try and and get them up

Splitpin

A/F correction (or learning) doesn't help you at all under boost as it switches off once the ECU goes open loop. Does your software package monitor the O2 sensor voltage? I presume so as you were telling us about the idle and boost voltages earlier.

And whatever was being used on the rolling road, if a wideband lambda sensor was in use, and whoever was looking at the readout told you it was rich, if they didn't actually tell you what the AFRs were at the time, phone them it was up and ask them if they have either a printout, or remember what soft of ballpark it was in.

Before you do that can you answer the questions I asked earlier (posts 43,44, 47) re. mods and mapping please. It's difficult to help you when we're not in possession of all the relevant info. Before you go getting more data, it would be worthwhile clearing up outstanding issues on the last lot!

ZEN Performance

This might not be a helpful time to point out, but someone in my position should be able to get to the bottom of this, and solve it, in about an hour.

I know the OP is trying to do it all DIY, but sometimes it's really not worth it, and instead your wasting the time of the likes of Splitpin by not giving all the info.

Based on the logs you're getting some compressor surge, or a sporadic leak and your timing is probably normal based on the current circumstances, which themselves are not normal.

If you go to an experienced mapper, they will know how much maf voltage you should be hitting more or less at a given boost level, check the AFR is within the bounds of normallity and check the boost level matches up with the sort of boost control the ECU is trying to carry out.

Splitpin

Agree with Paul's comments entirely. If you want to diagnose it yourself, then as above, we'll need the relevant info to help you - in this context your trip to the rolling road appears to be a waste of time without coming away with the specifics.

If you just want it fixed, this is something that anyone who knows how would likely be able to pinpoint and address very quickly.

scoobyboothy

iTrader: (1)

sorry guys, I thought i had mention mods, it only has a hks ssqv, fmic induction cone and centre cat removed.

I have found small split in the pipe from the compressor housing which i have managed to move to before the jubilee clip (temp measure) and i think the o ring in the dump valve adaptor is not sealing quite right (is there a repair kit for this?) so I will be borrowing a standard one from somebody soon.

I like to faff about and learn whats happening with things, and exhaust all my ability B4 going to the experts. with your expertise and suggestions I am learning alot

I don't know if it's been remapped, and is probably unlikely as the boost is still standard. if it had been remapped with full decat and the standard cat put back on, would this cause rich issues

I have done some fresh logging tonight so at some point in the next few days i will post up some, hopefully the 2 probs i have discovered may cure it all thumbs crossed lol


once again, cheers for yoour input

scoobyboothy

iTrader: (1)

bump

Saint AAI

I'm the same with liking to have a go myself and learn along the way. I know it's better to go to an expert, but I enjoy having a go first and with the Impreza being the "toy" car, it doensn't matter if I mess up and the car has to be off road for a while.

I would see if Area 52 saved the dyno run and ask if you can get hold of a a graph of AFR against boost pressure. Also, are there any bends in the induction pipework just before the MAF housing?

scoobyboothy

iTrader: (1)

i'll give them a bell tommorrow. where would the FPR be located. there are no bends just b4 the maf housing

joz8968

iTrader: (13)

FPR is on bulkhead end of fuel rail, on the off/driver side part of the inlet manifold.

Look down by the turbo into the abyss lol and you should see it (looks a bit like a dump valve on its side! :0)

scoobyboothy

iTrader: (1)

found it!

joz8968

iTrader: (13)

Good man! lol

scoobyboothy

iTrader: (1)

I have taken slight change in direction in my investigation into my running issue, After replacing the cone airfilter to a K&N one i had given me (other one had seen better days) and replaced plugs and leads it ran alot better, seems to use less fuel, although the lambda reading is still over 1.0v , but would suffer random powerloss which happens in relation to A/F correction dropping to -25%! Someone suggested disconnect the lambda sensor and see what happens, and low and behold it runs sweet as a nut. would a dodgy lambda sensor cause this issue, and would running without one cause any dmage to the engine