hp18racer

After I got it dialed in I went to just a little more 100 octane than I ran on the short sessions. I got in 2 20 minute sessions with seeing any changes in the hit count rate (50 counts or less per lap).

samluke

hp18 racer,

My car is PCA stock the only mods being opened out cat and K&N filter
I have been running my car for a couple of weeks now, and I get counts on every acceleration run, which I find surprising considering its stock. I am encouraged that others are getting knock counts and can change the behavior based on octane, which is what you would expect. The highest octane pump gas I can get in this area is 91.
Once I get back to the track I plan to do a more detailed analysis and experimentation with higher octane gas. Just as you have done, my plan is to mix regular pump gas with race gas and come up with a combination that results in a consistent low knock count.
It's great to know that Porsche's engineering really works. I have been racing for many years, and never bothered with race gas. I have also run much higher boost and horsepower in the past, again on regular pump gas. I suspect that if I had the counter hooked up then I would have seen some pretty high counts. Despite the abuse I have never had any knock related problems. My motor now has 170K on it. The last 40k were after a minor motor freshening, and since then its had lots of track use and abuse.
More data from others on what their experiences are with knock behavior would be great. It would also be great to know if any of the high horsepower guys have ever looked at knock during tuning, to see what works and what doesn't, and what they end up with as an acceptable count.

Rogue_Ant

Ok, I said I would take a log... I set up my system to capture events on pin #15 from the KLR.
Here are my mods that matter:
MAP conversion
K27DR
3" Exhaust




This was at ~19psi peak falling to ~17psi. This is an absolute value. Gauge reading is +3psi more (because of altitude).

After a few pulls I didn't get a single count of knock. Even after changing the software to simply act as a 'counter', I haven't been able to get any knock... My hardware is fast enough (faster then 1000samples/sec which is what Transaxle was using).
I almost wanted a little knock, so I know its working... I will pull the plugs tomorrow try to confirm the 'no-knock' situation.



Rogue

Rogue_Ant

Pulled plugs. They look great! Almost no speckling.
Is anyone else getting no-knock at all?



Rogue

samluke

I am at a total loss on my car. Its almost totally stock. Light throttle, light boost I get no counts. In 4th and 5th gears, if I stomp on the gas at low RPM I consistently get counts somewhere from 2-12, once the boost comes on. After a couple of seconds it stops counting any higher. Hard runs through the gears I get a couple of counts.

If I am stock and running only 12psi boost, does it make any sense that I am getting a knock count? I have changed everything and the behaviour remains the same. The car feels very strong. I would love someone with a stock car to post a count

hally

looking good Rouge

Rogue_Ant

Thanks Hally! The new chip is great! As a few Camaros and Trans-Ams can attest too.



Rogue

ross255

Is there any way to test the knock sensor to check its working.
Its just that at 18 psi, I have my wideband setup to look for signals on pin 15, its not very fast only about 10 counts per second, but it does have a nice led on the wideband gauge it lights up if it senses more than 1v. Well at 15 its fine but at 18 psi, it was like a damn disco.
This was with running the equivilent of 93 octane.

samluke

There are Rennlist threads that describe a test using a screwdriver and hammer with the engine idling. Supposedly if you get it just right the engine will stumble? Never tried it though.

ross255

I cant get that hammer test to do anything, I was hoping for something a bit more definative. As it is sensing knock, as proved by the signals from pin 15. But I seem to be in the same position as you, in that it seems to sensitive? I mean how do some people run 19 psi (rogue) with no problems, and i get knock with 18psi.

samluke

I think pin 15 is the way to go, as its the KLR determining that knock is occuring. The DME responds based on the signals from the KLR. I like the counter over a light as you can see from the count exactly what is happening per run.

As I am running PCA stock I cannot run a boot controller, otherwise I would like to experiment with what boost level brings on knock.

The thing I am confused at is what a stock car looks like. I don't know the DME logic in response to the "knocking" KLR output, so it may well be that other stock cars also trigger the "knocking" output but its not enough to cause the DME to change the timing.

The other thing I don't understand is how others can be running much higher boost and claiming no knock. I guess the short answer is that there is not enough data yet to know whats right.

hally

Once I'm up and going I will measure a knock count also, wish I had done this before i started modding.

hp18racer

I read about the hammer and screw driver knock test in the same book by Bosch that also says the computer only looks for knock right after the plug fires. I'd be very suprised if you learn anything from this test.

The knock sensor is a crystal that resonates with vibration at a certain frequency. I'd be very suprised if the goal of the computer system is 'no knock events' at all. Another thing to consider is that ANY vibration (valves, exhaust,...) in the motor that resonates at that frequency could be considered to be knock by the computer.

Samluke, your car still not maintaining boost?

samluke

This boost problem is driving my absolutely nuts.

I did some more testing this weekend. We tested HP18's car, 2.8L, 17-18psi, 100octane fuel. The knock counting wasn't so different than mine.

Pulled the cycling valve plug, it max's out at 6psi as expected and I can never get a knock count, which is not too surprising.

In pure stock form the car peaks at 12psi, holds for about 3 seconds and then drops quickly to 9-10psi. It has nothing to do with RPM. Its highly repeatable. I have various gages and they all confirm that its real.

If I use a restricted banjo bolt I can force the peak up, but it still falls back to 9-10psi after 3 seconds. I opened the cycling valve vent line to atmosphere to make sure there was no restriction. No effect.

I have changed everything bar the motor its self. I am so psssssd off.

I have yet to try 100 fuel, but I don't expect it to change anything. I have a manual boost controller on the way. That will at least prove if the Turbo/exhaust plumming is mechanically capable to holding the boost.

I

samluke

I did some more testing today, and found an interesting phenomenon.

5th gear full throttle, the boost bleeds down after 3 seconds. If I do a quick lift from WOT and get straight back on, I get full boost for another 3 seconds. I can keep doing it and the car keeps accelerating, coming back on high boost each time I lift.

I experimented with the lift. Just coming off WOT seems to do it. The lift is as quick as I can do it, so a fraction of a second. Its nice to feel the power again.

Maybe I have a bad WOT contact in the TPS? so next step is to change it to see if it has any effect.