Subaru Impreza GC8 & RS Forum & Community banner

help: detonation

2K views 10 replies 7 participants last post by  Graham 
#1 ·
1st post,

I have already posted this on the i-club, but I cannot get much help, I have already checked all of the normal things and I have everything setup how it should be. These are all of the posts that I had from the i-club thread:

2nd time around, I have all of the right stuff, but I am stumped.
I will start by giving the list of what I have:
Legacy turbo setup with only around 3-4 psi (lower than before)
SVX fpr(34 psi at idle) and vortech 6:1 FMU(around 60 psi at boost)
Walbro pump
Magnecor wires with NGK step 7s
J&S safeguard (kinda helps but does not fix anything
EGT on #3 around 1250-1275 at cruising, around 1500-1550 while WOT (This does not really help because it is pinging on the driver's side)
----->Should'nt this be enough to keep me from running lean???
mcgyver and I spent some time on trying to figure it out and it is probably either the #4 running lean due to an injector or maybe a problem with spark like Larry Ganz had. The only weird thing is that when I bought this car with 17000 miles, it had a burned electrode on one of the Denso plugs (and I cannot remember if it came from the driver's side where I hear the pinging) Reguardless there must have been something wrong for it to have burned up a plug. This is why I am thinking it might be a problem like Larry Ganz, even though it is not really a misfire. It just rattles while WOT. The J&S picks up something, but I do not want to avoid the problem. I can have the J&S help but it would retard too much to get it to stop. I am going to try to put some race gas in tommorrow (mix 50/50 100+91) but even that will not be a true fix, and I cant afford that for too long. SOMETHING has to be haywire. (ALREADY TRIED DID NOT FIX IT)

So, now I am entirely frustrated with this thing. The car stopped pinging only for a while after doing the ECU reset and messing with the plug wires. Now it pings, but the J&S does pick it up. But I have to have it so sensitive to where it hurts the performance. I really need some serious help. If somebody could please help me try to figure this one out I woulld really appreciate it. Like I said before I have about 64 psi of fuel at boost and I think that would be enough to stop it from pinging. The EGT probe is on the #3 cyl, but I hear the pinging from the driver's side. #3 is about 1250-1275 at 65 cruising and the times I have gone faster it would stay around 1500, which is good as far as I know. I have done a compression test and all cyl are at 185. Plugs all look good with tan residue. The only thing I can think of is that maybe the SAAB intercooler is not efficient enough, even though I never had a problem on the other car. And it is low boost. I had a guy scan the data and everything looked good except he thought the timing was too advanced (even though he is not a subaru guy). It said the timing was at 40 degrees or something. Please help, my turbo is useless at full throttle.........................

I have a Vortech 6:1 and am only running around 3-4 psi. I cant see why I am running into problems. The AFR guage is after the two cats and reads green(rich) during boost. So you think I need more fuel pressure? That means I would have to have a 8:1 or an adjustable one. A lot of people run 4:1 with no problems. I even swapped the SVX fpr for the stock one and it gave me around 70 psi at boost. I think I am going to buy another probe and have a switch for both #3 and #4 to use the same guage.

This is pretty much my whole problem, maybe you guys here can elaborate.

Thanks,
scootr
 

Attachments

See less See more
1
#4 ·
You may have hit the proverbial nail on the head with the fouled injector. Try swapping injectors from one side of the engine to the other, and see if the ping follows the injectors. It would also be a good time to inspect the injector inlets for dirt/debris. The 65 psi could have damaged the fuel filter, enough to cause the filter element to start to come apart. Just a thought. Have you ever had the timing belt off? The driver's side cam can be tricky to time correctly, and a cam that's one tooth off will cause some pretty bad spark knocking, but won't be far enough off to bend valves.
 
#5 ·
RS25- I have a 6:1 and when I had the car scanned the timing was at 40 degrees in the data.

yama rocket-I thought maybe an injector was the problem, and I may try to swap them around. Yes I have had the timing belt off, but I thought it was either hit or miss. There are three marks on the belt, and the cam gears both point straight up, the middle one points up to its mark and then as long as you line the three up exactly you are ok. The belt would not go on correctly if you did noy get the marks to line up. So is there a different way I should have done this? I did it on my other RS and did not have any problems. It makes sense that it might give me a spark knock but I thought it was pretty self explanatory. If there is a better way to do it I will take the damn thing apart again and do it better.

Also today I bought a vacuum/fuel pump tester and t'd into the vacuum line to find that when my autometer guage says 3 psi I am really getting near 6 :eek: . So could I be getting close to maxing the injectors? Or maybe need more fuel? I'll be sending that guage back :mad: .
 
#6 ·
I know you said you messed with the plug wires, but have you pulled the plugs out and looked at them? They should tell you a lot.

Have you done the parallel fuel rail mod? Many people have had a problem with the number 4 injector from running high fuel pressures.
 
#8 ·
Scootr, I was blessed with having a friend of mine who is a Scooby Tech do my timing belt, so I wouldn't be one to ask in this department.

I'll let you in on a little secret, I found a new way to do the parallel fuel line mod. Well, its not exactly parrallel, but it gets rid of the #3 cylinder running lean. Let me know if your thinking about it.


Graham
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top