David_Wallis

bob,

Some decent timing lights that allow you to adjust the offset in degrees cant cope with running the timing light at 0 degrees if you are doing something similar to scattered spark / multiple sparks.. (gems does scattered spark for idle apparently)

The snap on light I have shows my timing jumping about all over the shop, however add 1 degree eitherway offset and its fine.. (theres an option on the msd for setting timing up)

Also its supposedly fine at 0 degrees with a normal cheapo timing light.

Also

Later cars have the ignitor in the coil (99+) hence the coil having four connections as you state.. Ignitor +, coil ground, Pair 1 and Pair 2, 98 cars have three wires, coil ground, Pair 1 and Pair 2.

92 - 96 have 4 seperate coils, a 4 channel ignitor, never paid much attention to the driver side of the oem ecu.. so in theory this could be true on a 92 -96 but not the latter, the only possible theory is andy's idea.

David (getting out of his depth, if he goes any further)

The Fixer

Thats all very well gents but my point was that the GEMS for the new age was not using wasted spark but was using no 1 cam sensor in combination with the crank sensor to start the car. I watched on the screen of the laptop as the cam sensor repeatadly came up with numbers other than the 51 teeth it was supposed to see. When the car was extremely cold or very warm the cam sensors 1,2,3 & 4 werent seeing anything, this is why the car failed to start when cold or warm and was seen to be throwing sparks in that the wrong time. GEMS had assumed wrongly that the newage cars were using these cam position sensors on the OEM ECU to start the car but it turned out that they wasnt. It was also noted that once the car was running the cam sensors worked perfectly and the GEMS switched into sync mode. I hope this clarifies things.

Conrad

NB: My car is a coil on plug model also.

Andy.F

My Microtech ECU used to throw a fault indication - cam sensor -sometimes when hot. Didn't affect the running or starting though

Bob

Is it possible that what you were seeing was the timing gun switching frequency ? Depending on manufacturer, they flash every pulse up to a certain rpm then drop to every 2nd pulse. This does not change the actual timing though.

Andy

The Fixer

Sorry David just seen your comment, what pictures?

Conrad

DuncanG

Mmm a lot of interesting but confusing stuff here.

Cams are running at half crank speed so the variable reluctance cam sensor signals are going to be weaker than the crank signals. So maybe its more reliable for the ecu to start by ignoring the cam and that implies all 4 cylinders spark on every stroke (double-wasted?). However maybe the GEMS sensor interface isn't up to the job and they've come up with this as a work-around to get it started. In other words theres no reason why GEMS and oem ecu have to use the same starting strategy. My car wont start if the cam sensor is pulled so I assume that it uses the cam sensor for starting, but its a different model etc and maybe it does a resistance check on the sensor as part of power-up self-test.

Conrad,
You saying its got 4 cam sensors each with 51 teeth? I thought the AVCS engines used 1 cam sensor, same as older engines, with the weird Subaru 1,2 or 3 pulse timing marks plus 2 cam angle sensors for AVCS feedback (51 teeth?). Does anyone have detailed info on the AVCS sensors and control mechanism? Does GEMS require its own timing wheels to be added?

David,
"gems does scattered spark for idle apparently" Can you tell me more about that, it sounds interesting.

The Fixer

Duncan, as previously mentioned the GEMS was previously using the crank pulley and the no 1 cam "position" sensor to indicate no 1 at TDC to throw the spark in. Without using the cam sensors and just using the crank sensor you will never know whether its no 1 at TDC or the other cylinder. (No 1 comes to the top every two revs of crank pulley.) Doesnt fire a spark on every cylinder only no 1 or its opposite cylinder(cant remember which one).

I believe the cam position sensors and AVCS sensors are different sensors in different positions.

GEMS have now altered the software (eprom) to fire a spark every time the gap on the crank pulley is found up untill the point the engine goes into "run mode" which is taken from a given RPM at which point the GEMS switches to sync mode using the cam sensors.

GEMS uses the standard crank wheel and hall effect sensors.

Can't speak for your model only the STi 7 & 8's we have been messing with.


Conrad


[Edited by RSVR Racer - 12/17/2003 1:31:19 PM]

The Fixer

Right, drove the car tonight after Steve has extensively remapped / fine tuned it in the time he had it, it definiteky starts ok now and seems alot more responsive and quicker to spool up than it was before, he did say it was running very rich before he adjusted it

Headers are here, going to wrap them tonight. Tommorow hopefully, I'm going to take the car to Well Lane before I fit the headers to get a power run. I can then compare it to the two identical HP G-Force runs and it will also give me a base HP/Torque figure before the headers are fitted. I will then take it back to Well Lane after they are fitted and the map has been adjusted to see how much they actually made on there own.

Little concerned about the flange on headers where the up pipe connects, it seems slightly rounded???? I will have to check how flat it is later this evening.

Conrad

David_Wallis

I have a surface plate.. and depnding on size I have access to a surface grinder.

David

RB5SCOTT

Conrad

Mine were ok at the up-pipe joint, Its probaly better anyway if its slightly bowed in the middle because the bolts on the outside of the flange will pull up and keep the middle tighter

Scott

The Fixer

Good point Scott, plus there is gasket in between, did you use 3 new gaskets or the old ones.

Regards

Conrad

RB5SCOTT

New ones

The Fixer

How did I know you were going to say that? Did yours come with them? Or where did you get them from?

RB5SCOTT

I got the up-pipe gasket with the headers but XtremeScoobies kindly supplied the other two

btw, got my credit card bill the other day £307.11, not a bad price for a set of headers

The Fixer

Not sure what mine ended up costing yet, duty & import tax was £56 though ontop of the $300 so still a good price I think.

Still sat on the floor looking at me, haven't wrapped them yet

Conrad

The Fixer

OK...UPDATE time.

I have been to Well Lane Turbo Centre this morning to put the car on the rollers before fitting the GRUPPE-S headers. I didnt tell them what I'd got anywhere else so as not to allow figure nudging. Car was ran up a few times and........

HOLY SH*T


Exactly the same figure as G-Force Motorsport - 362 BHP @ 6600 RPM 305 ft/lb @ 5500 RPM


Air temp outside 10 degrees, roughly the same as the last time we ran at G-Force. Graph to follow. How spooky is that, two identical runs on G-Force Rollers followed by the same at Well Lane. They also checked everything over to ensure no boost leaks while it was being run up.

My car seems to like this 362 HP figure

Next Up....Headers to fit over Christmas and remap to suit.

Conrad


[Edited by RSVR Racer - 12/23/2003 1:53:59 PM]

RB5SCOTT

I thought you was expecting 500bhp at well lane

T-uk

I think I read a post somewhere by andyf , explaining why well lane can vary about between private runs and shootout days

The Fixer

Scott, someday

T-UK Would be interested if you or Andy can point me to it.

Conrad

The Fixer

Bodystyling, the car cutting out is a mapping issue I had the same problem and Steve Simpson has since corrected this. GEMS have also generated a new eprom which needs fitting into your ECU which helps sort out the starting problems. Your best bet is to contact Steve and get him to sort these two issues out for you.

Conrad

The Fixer

Just had it power run at Well Lane before I fit the headers, then I will go back after the headers have been fitted and the car has been remapped so i have a before and after run so we can find out how much difference the headers actually make on there own.

After that I'm going to remap the car, maybe try the big turbo off the 2.5 project for a laugh and help set-up the physical position of everything ready for the 2.5 in early Feb

Drop me a mail and I'll fill you in re: some other stuff.

Conrad

[Edited by RSVR Racer - 12/27/2003 9:32:18 PM]