PorKen

Limited without a SharkTuner, right?

With the SharkTuner EZ-F, you have (5) cells for the air temp correction at cruise and WOT. The stock chip programming is simple, in each state, it retards 3° if the input is <-35°F/disconnected, or >125°F/shorted=octane-loop.

You probably could use the EZ-F coolant temp input too, with programming.

Disconnect the temp sensor, and have the input come from a knock box at a specific resistance. Set the amount of retard to a safe value at that 'temperature' with the ST (6°-10°). With a timed circuit, you could progressively raise the retard if knock persists, with different resistance/retard.

The Knocksense box, with the Megasquirt option, also has a 5V high (low also), output, but I don't think it has the signal conditioning of the Knocklite. (KS wiring, KL wiring)

Then there's the mystery LH2.2 octane loop input. Might be able to richen w/knock using that input.

John Speake

Hi Ken
Yes, lots of possibilities using the SharkTuner. Once you have the remapping for EZK done correctly for the degree of modfiication, then you are just looking for some addtional retard to cope with poor fuel etc.

Louie was presented with a much more difficult situation, a complete reampping of igntion from scratch. With such an expensive motor, I agree with his choice of a top notch knock control system.

Louie928

Yup, sort of. Lot of trouble to go through with still no way to know what each cylinder is doing. Might work to prevent a really bad detonation condition. Maybe as a save until you can get some better gas. I remember Tim Murphy found that one cylinder was more prone to knocking (on an S4) than the other cylinders. You have to remember that each cylinder will have a different level of background noise and knock signal. You need something to address that. Even so, it would be better than nothing. Or, give a false sense of security. The Knocklite seems a better system than the Knockbox. The Knocklite does consider the different background level at different RPMs if only for the engine as a whole. Some engines have so much noise, typically from loose pistons, that the background prevents reliably sensing the knock. This is usually at low loads such as the condition when you calibrate the Knocklite. That's where the gated listening and only allowing knock retard at heavier loads helps. The 928 doesn't have the extra noise from pistons flopping around that some engines have and I doubt the possibility of low load calibration, and knock retard, would be much of an issue.

The situation that convinced me that I had to get really serious about knock control was when I went to Tahoe this June. I think the best fuel there was 89 octane. I filled up again at Lakeview, OR (elev. 5000') and again less than 92 octane. No noticeable problem when I was in the high country as I headed home. As I dropped down closer to sea level maybe 50 miles South of The Dalles, things got worse in a hurry. I couldn't pull much of a hill in 5th without audible pinging, and there were no gas stations to get better fuel. I could carry a gallon of Xylene or Toluene to dump in the tank, but that's not a permanent solution. I fully realize that this engine I have is the only one like it I'm likely to ever have (as long as I'm married). I cannot break it. The cost of the J&S was a lot easier to absorb than fixing the engine. The easy install and setup was a bonus.

BTW, the addition of Xylene, which slows burning, also makes the car harder to start and causes poor idling when cold. Very noticable. Probably makes for less fuel mileage due to more difficult vaporization as well although I haven't checked that aspect. I can't remember if Toulene had that negative effect, but it wasn't as cold as now when I used it this summer. Just for grins, I'm experimenting with adding a little acetone to help that. A small amount of acetone (3oz per 10 gal) is supposed to improve gasoline vaporization and help mileage. I feel a home brew octane boost formula in the works.

PorKen

True, a lot of trouble. Since I'm pretty dumb about electronics, I might try a 5-position rotary switch with different programmed levels of retard, in place of the speaker fader, along with a Knocklite.

The J&S looks great, and if I were spending the big bucks for 85-86 boost, another 5 bills would allow for better advance programming.

You're not a fan of water injection?

Louie928

Water, or water alcohol, injection works for boosted engines especially if they have inadequate charge cooling, but without some way to accurately know when detonation is happening how do you know how much to throw in? I guess if the engine quits, it was too much. If it quits with parts coming out, it wasn't enough. If I were to work up a system to inject something, it would be a mixture of alcohol and Xylene. The alcohol to cool, and the Xylene to add octane and tame the burn. Inject only at high power. No need for it at cruise.

BC

Can I still run the "smart coil" level 5V outputs on the VEMS? I already bought the Cayenne Turbo (and CGT) coils for the car, and that was like 600 bucks for the looms and plugs - the coils are only 12 bucks.

I guess I should actually read all of Louie's post. Lots of info there.

BC

I don't know either. If you run it in sequential, and you have two knock channels, I suppose it would know WHEN the knock happened, and therefore WHICH cylinder the knock happened in - but I don't know.

Lizard928

Brendan, it could be used in tandom with a FULL cop setup running VEMS.
However why dont you ask on the VEMS board if it has the individual cylinder trimming?

IcemanG17

This is a very interesting post....especially since I live in CA with our crappy 91 octane gas.....

I did some calculations using Louies formula.... So 8 psi equals about 12.4CR (like andrew said), which is about the same that the Beast runs with, but with 100 octane fuel.... Now 12 psi would be 13.4CR....or far higher than any 928 out there.....but of course most S4's are not really 10.0CR...more like 9.5CR....which drops the 8psi to 11.8CR (almost like louies) and the 12 psi to 12.8CR

But I think another consideration between NA engines and boosted engines (in terms on detonation) is that when cruising (i.e not under boost) boosted engines will drop back to thier non-boosted CR......vs the high CR NA engine, where every single compression-firing is at the higher CR....& probably more likely to detonate since its running the O2 loop, which also tries to hold a 14.7 AFR during cruise mode......

So a boosted engine that is not run hard, should tolerate poor gas better.....of course if it is run hard, then things would go bad quickly

Mike Simard

Louie, thanks for the report . I think that keeping detonation at bay is the key to having not only an engine that doesn't break but one that you can drive without worrying about. If you worry about your car you wind up either not driving it or detuning it, removing power adding stuff until it's reliable. What you're doing is probably the one of the most important mods to a car like yours although it might not seem important at first.

What's the dynamic CR of your engine? Do you have any recomendations for the static CR on mine? I'm having custom pistons made now have planned on 11:1 static which is 8.5 dynamic in my case. That's considered safe and I designed the pistons to take advantage of all the squish area on the heads but I still worry about detonation and whether the CR should be changed. Here's a handy dynamic CR calculator: http://www.kb-silvolite.com/calc.php?action=comp2

If I ever finish my engine and it does NOT blow up, it will be because of your guidance!

928SS

I'm working around the issue w/a 2nd coding plug. 1 for our 91 crapolo fuel and 1 for 101 yummy stuff... swap plugs and presto!

haven't wrapped up the 101 maps yet, but we did put in 30# injectors since the blackbird made the 24's max out at WOT. hopefully next week... had to start over w/the larger injectors...

any idea how much more HP w/the extra 2.5 or so deg's advance and the diff btween 91 and 101 octane on an 11.3:1 NA motor??

cool thread

blown 87

I have seen that, a motor that will make 12 psi at 6000RPM will make a lot less boost after a port job a some exhaust mods, but it will make more power.

Upper deck pressure (boost) is a bit misleading as a indicator for power output BMEP, is better indicator of power.

Louie928

Hi Mike,
I think you hit it right on regarding thoughts about detonation control. We all think about it when modding a motor, and try to do what we think will be ok. It wasn't until I got out driving some distance in the real world (as opposed to the more tightly controlled dyno world) that I saw how easily a nicely tuned motor could be in trouble really quick. Not wanting to de-tune, I had to do something about it.

That's a great dynamic compression calculator. Mine works out to 12.25:1. The stroke is 3.75", rods 5.85" and the intake valve closes 46 deg ABDC. Static compression is 11.4. This formula along with the other one for figuring boost compression may help a bit. The late intake closing is because I have my cams set 4 degrees retarded, not "straight up". The intake cam lobe center is at 111 deg ATDT and the exhaust center is at 103 BTDC. That setting is a running (hot) engine. Cold setting, you'd add about 2 more degrees retard because the cams advance that much between cold and hot. I have the intake cam closing late to reduce VE at lower RPMs for less low end torque and detonation control, and also to help maintain torque at the higher RPMs. The problem with a big inch motor isn't getting enough torque, it's needing to maintain that torque up hgh in the RPMs. That makes a lot of fun to drive because it keeps pulling the tighter it winds. The ITB intake has throttles right near thevalve. That keeps the reversion flow from the late closing intake valve partly blocked at part throttle. This keeps the mixture in the port under the throttle so it doesn't spit back out and keeps reversion flow from one cylinder from affecting other cylinders since they all share a common (large)plenum.

One thing still missing in the formulas is that volumetric efficiency (VE) figures in too, but it's not in the formula. I use Engine Analyzer Pro to help me determine what I should do for different engine parameters to give me the outcome I want. EAP says that I have 118% VE at the torque peak of 5250 (actual torque peak is around 5100) and it's still 107% at 6500. The lowest VE of 82.5% comes at 3500. It's higher below and above that RPM and the dyno curve does have a little torque dip at 3400. The simulation is pretty close to to the actual engine. The greater than 100% VE is going to give me much higher dynamic compression at those RPMs than the formula would show. In fact, EAP shows an intake pressure of +5.3 psi at 5250 and +7.6 psi at 6500. Those aren't constant pressures like boost, but peak pressure wave value due to the intake tuning. The pressure is higher at 6500 than at the torque peak at 5250 because I purposely have intake runners shorter than they would be for maximum torque. Tuning the runner length for maximum torque would have given me more torque than I wanted at the peak, and the torque would have fallen rapidly above the torque peak. The short (slightly tapered) runners broaden the tuning peak and keep pumping air in quite a bit above the torque peak to give a good top end.

hernanca

Anyone applied a J&S Safeguard system to a pre-87 928 (especially 85-86.5)? The Vampire model seems the appropriate one for the 928.

I am considering such a system but am wondering how they can track which cylinder is knocking and how they can specifically retard timing in that cylinder? It looks like they just count spark events? I understand the 87+ uses the two knock sensors and a Hall sensor to track and address the per cylinder behavior. I have two knock sensors retrofitted to the stock locations on my 86.5, but I don't believe I have a Hall Sensor. Louise seemed to have no problem recommending this for the pre-87 cars, so I am hopeful.

hans14914

I have seen them installed on the S3 motors. I believe the installs were global retard, so did not use the hall sensor for cylinder tracking.