When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Hello to the world of the 928 on line, it’s been a while since I was lastly here. Long story made short, the orange car now has ITBs and I’m at the stage of discovering the shark tuners functionality. I’ve only scratched the surface and The depth of work that has gone into this from John Speake and co is amazing, more so that a complete non initiate can read the manual and start tuning. These are the bmw throttles from an e39 M5 with spacer plate courtesy of Alex Popov whom I have to thank for supplying technical support through the project and installation I’ve also had valuable advice from other UK 928 people, Adrian Clark’s advice on initial throttle plate setting helped enormously.
For various reasons I’ve built a hood scoop and the ITBs breathe through a bonnet opening, data logging shows it to be a true CAI. We’ve been having beautiful cool and clear weather here in NZ for the last several days and it’s been fun to drive the car again. The engine is a dry sumped GTS lump, Elgin grind S3 cams and headers with a dual exhaust and an X pipe just after the merge collectors.
Ive got a handle on the fuel requirements, mapping to a lambda value is relatively straightforward. When it comes to the EZK however I’m slightly baffled. Part throttle and off idle is nice and crisp throughout the rev range so far, but what should my max load/Wot targets be is my question? The Elgin cams have definitely pushed peak torque higher in the rev range and the intake without flappy dosent seem to give the split torque peak. In fact it feels a lot like a single longer torque curve with the peak nearer 5k rpm.
Does anyone has advice on WOT ignition advance in a modified 928 engine?
I will be going to a dyno eventually but I’m keen to map as closely to ideal as i can before that.
Really all you can do is to keep advancing by 1 degree until the engine starts knocking (as indicated by the sharktuner)
IIRC you can adjust advance by cell, albeit at a low resolution like the LH fueling. There really is no strict way to get the best results for your setup as it is different than what anyone here has on their 928.
Well I might be getting closer to finding peak torque although sideways at 4K rpm in third it’s a struggle to watch the road and computer screen. I think I’ll stick to data logging runs and scrolling back through them going forward
Blake, yes this is the standard advice of pushing the targets just a couple of degrees at a time and I guess that’s what I’ll be doing. No such thing as nock with 27 degrees of timing so far.
drnick, welcome back, the base circle grind intake cams I purchased from you with 6 psi pulled 425 rwhp,,,,,,it was a blast, glad you are back !!!! I took my stroker to a tuner on the dyno and after 3 sessions had it dialed in pretty well.....
Andy, I’m glad you had some use from the reground GTS cams. Looks like you’ve found a few other ways to get power now it’s fun to have the car back on the road here in NZ.
I know adding ITBs isn’t the most common thing to do but I’m sure others will do this in the future. Using the shark tuner and getting the car to run is at first quite daunting. There were a few surprises in getting things set up including initial setting of the throttle plates - trying to get them as even as possible and guesstimating how open they should be at idle. The TPS needs attention during installation as well to ensure it’s giving you readings that correspond to idle and cruise as a lot goes on with even slight throttle openings. The other surprise is how much fuel the engine wants off idle with the ITBs, it wants to see a wall of fuel practically. Setting the ICV to a fixed number is also the way forward to begin with, I’m sure to go on and experiment with the JDS idle controller as the stock program won’t work correctly. You also need to pay attention to the amount of advance just off idle for smooth operation, it wants a little more than the stock maps. The acceleration enrichment maps benefit from adjustments too. It also helps if you know your injector size and set that correctly to begin with
As for the ignition mapping.. setting the scaling is important. I’m at 50% and probably heading towards 55 or 60 as I get closer to some kind of ideal. First impressions are that the 32 valve engines can cope with quite a wide range of values without complaint.
Trying to do sharktuning solo on a stock motor is difficult enough but on a monster like you are building I would say it is nigh on impossible unless you are on a dyno [or loopy!].
One point I would advise is that things like the exhaust system one is running make a difference yet alone the ITB's, cam profile etc etc. Folks like Louie and Alex may well be able to give you some guidance but given the custom element factor I rather suspect it is more a "suck it and see" type of thing.
Even my modest efforts on a lightly modified system gave results that surprised me a bit but whether they are optimal remains to be seen as I have no access to a dyno. What I found running the local 95 RON fuel [much to my initial dismay] was that ST2 suggested that I needed to pull some ignition advance from the top end 27 degrees when I had hoped to be able to add some. I was later wisely advised that a fast exhaust requires less advance so maybe this was normal. Presumably pushing a lot more mass airflow your situation will be controlled by your exhaust and its ability to cope with the increase noting the lengths Tuomo has gone to with his tubo monster's industrial chimney.
Trust life is treating you well down under. Look forward to reading about what you manage to get out of the thing.
Fred, life is good here and I’m having fun taking the car out for runs. Mind you I think I’m really stroking a ***** cat with this engine compared to some of the beasts here on the board. I’ve switched to data logging to get my info rather than trying to wedge the laptop in my field of view and drive - not recommended I here you about the effects of cam timing and exhaust on air flow, i have a growing appreciation for how complex a system it is that we are trying to optimise and I’m re adjusting my goals simply to have a safe enough tune with good part throttle response - until I can arrange some dyno time.
Ake, I’m assuming 36 degrees advance is at maximum rpm and you have something closer to 30 at 3k rpm? The only tables I’ve had to look at for the 928 relate to the 86.5 and a couple of tables that are in the shark tuner. One of the many things I don’t understand is why some of the tables have a ramp at low load, particularly in the upper rpm range. The 86.5 tables seem not to have this ramp and launch straight into maximum advance.
...Ive got a handle on the fuel requirements, mapping to a lambda value is relatively straightforward. When it comes to the EZK however I’m slightly baffled. Part throttle and off idle is nice and crisp throughout the rev range so far, but what should my max load/Wot targets be is my question? The Elgin cams have definitely pushed peak torque higher in the rev range and the intake without flappy dosent seem to give the split torque peak. In fact it feels a lot like a single longer torque curve with the peak nearer 5k rpm.
Does anyone has advice on WOT ignition advance in a modified 928 engine?
Nick, each engine is different and you need to tread lightly with respect to advance. The whole point of ignition advance is to give the flame an advance start, before the piston reaches TDC in order to generate max pressure for the downstroke. If you light the fire too late then the pressure reaches a peak as the exhaust valve opens and is wasted. But if you light the fire too early then it detonates as combustion meets compression.
The speed of fuel combustion depends on air pressure and compression ratio, efficiency of the intake design, temperature, and of course octane. Our early (lower-compression) '88 loves advance, and does best with 33 deg of advance at the top end (WOT at 5000+). Our GT, not so much: higher compression means less advance, around 25 deg in stock trim. And in its new incarnation (5.9L, around 10.5:1 compression) it is happiest at around 22 deg. That's stock intake (ported), yours would be more air which means more pressure and faster combustion and therefore less advance.
And then there is fuel. I am not sure what you have available, but the best here is 92 USA rating (R+M/2) which is 96-98 RON. I am sure that Ake gets better fuel than we do, but it is all relative-- you've got to run what you've got.
So what I would suggest is to set the last few rows of the advance map conservatively, log data with Sharktuner and look for where the retards are occurring. A couple of cylinders for a couple of degrees is fine, more than that and maybe back off a bit. And if you see all 8 cyls for 9 deg then that is trouble for sure-- 9 degrees is max retard so if you hit max retard and it is still knocking then you have no idea how bad it really is.
Jim, thanks for the excellent and practical advice. So far the very few occurrences of knock have been under 2k rpm and I’m running rich almost everywhere. I might be close as I’m going to get without a rolling road.
Andy, the foam filter lays across the top of the trumpets - when installed. Not ideal but it will be better k for limited road use. I’ll post a pic once I’ve got it in place.
Fred, life is good here and I’m having fun taking the car out for runs. Mind you I think I’m really stroking a ***** cat with this engine compared to some of the beasts here on the board. I’ve switched to data logging to get my info rather than trying to wedge the laptop in my field of view and drive - not recommended I here you about the effects of cam timing and exhaust on air flow, i have a growing appreciation for how complex a system it is that we are trying to optimise and I’m re adjusting my goals simply to have a safe enough tune with good part throttle response - until I can arrange some dyno time.
Ake, I’m assuming 36 degrees advance is at maximum rpm and you have something closer to 30 at 3k rpm? The only tables I’ve had to look at for the 928 relate to the 86.5 and a couple of tables that are in the shark tuner. One of the many things I don’t understand is why some of the tables have a ramp at low load, particularly in the upper rpm range. The 86.5 tables seem not to have this ramp and launch straight into maximum advance.
Nick,
Having ITB's creates its own problems as well as well as potential rewards and getting them synched correctly must be no small task but clearly doable. The down side of exhaust gas analysis is that with a common WBO2 port for all 8 cylinders one does not know what the readings are really saying detail wise. Assuming your are using Jim's sharkplotter programme I started with fuel analysis tuning and it seemed to me you can see a clear pattern as to what is part throttle and what is full throttle- they seemed to form two distinct zones on the map when plotted. for the part load I tried to target stoich and occasionally, a tad leaner. For the high load region I reasoned that an AFR of 13.0 seemed to have the motor happiest [stock tuning gave low 12's].
The problem with the stock manifold tuning is that air flows have a significant [but not huge] variance runner to runner and rev band to rev ban [it is how it is designed] so batch firing means cylinders to some extent are all over the place AFR wise. For whatever reason at high rpm's I found cylinders 6 and then 2 were controlling advance. I also found that in our hot summers I had to back off the timing a bit- ideally I should have hot weather and cooler weather timing maps. Eventually I found I was able to tune out knocks running our 95 RON octane. Although somewhat empirical, I found that RON octane rating was worth about 1 degree of advance per unit of octane. After lobbying some of my "friends in the know" a limited supply of 98 RON has been introduced into a few service stations [at a price]- one of these days I will get round to tuning on it but the local boys in blue in getting very efficient at finding speed merchants in the capital area these days!
I also found that AFR played little part in knock control and advance was the controlling element in the stock motor. I suspect you will not have any problem getting hold of RON98 and that is what you should run [that or higher if available]. Once I had knock under control I then did as Jim suggested and dialled in bit more advance trying to ensure there were no "huge jumps" between adjacent cells. Once I started to see occasional knocks events occurring up to 3 degrees on two cylinders max I called it a day. When doing this set the knock step to 1 degree so you can see what is going on. Once I had the desired settings I then put the step width back to 3 degrees. The knock retard system has a very quick response and seems to correct by the time the next firing stroke arrives. I have tried to see if we could get consensus as to whether we can use the knock control as a dynamic control system -i.e. deliberately set the timing 3 degrees more than the threshold point and let the system dynamically control but that received a mixed response.
To help this process Ken [Porken of the PKT tensioner fame] kindly wrote a neat little patch for me that enables one to retard selected cylinders across the board. It can be accessed from one of his tuning threads in the search engine. Using this I pulled 3 degrees of advance on cylinders 6 and 2 so that I could introduce more advance on the other cylinders and sure enough I managed to induce knock on most of the other cylinders thus concluded that with the stock manifold one was leaving something on the table top end as it were. With ITB's you should not see this particular phenomena but you may well see other similar patterns.
Bottom line line - what one is trying to do is tune the motor so that maximum cylinder pressure is achieved at something like 14 degrees after TDC as this is what creates the optimal push on the piston. Much as I respect Ake's excellent work, I would love to know how he manages to achieve 35 degrees of advance and prevent detonation. If my fuel theory holds linearly, he would need something like 110 RON. Theoretically this is possible if one could get the blending cut from a refinery Platformer unit. Alcohol produces an octane in this range albeit with much lower energy output- great for boosted motors less so for n/a motors or so I understand.
Do keep us updated how you get on and try to "keep it between the hedges"!
Awesome advice guys, really helpful. It pays to be scientific about the process and not get carried along too far with preconceptions before reviewing them. I’ve only just realised that the fuel I’m using, 98 octane, is a blend with ethanol. I think this is why I haven’t seen much in the way of knock.
Its probably also also the reason the car is drinking it like a fish.. there is less actual petrol in each tank. My understanding is that it’s quite knock resistant and that the ethanol burns slowly - needing more advance.
06-14-2018, 06:56 PM
Ignition targets
I’ve also had valuable advice from other UK 928 people, Adrian Clark’s advice on initial throttle plate setting helped enormously. 
