mark kibort

gee, thanks for reminding me.

However, i dont know if you know how i do my comparisons at the track, but i take many factors in to accout. I was overdriving at T-hill, running door to door with a few guys that i always race with and also do those 60-100mph runs to and from on the highway. when they usually end up in the low 6s (seconds) and now are in near 7 range, i know something is up, especially when i do this trick several hundred times a year! I also have marks at the track that i always hit, and when i go past those shift points, its pretty obvious im down on power. last time the engine felt flat, it was down 20hp and was able to fix this ith my cam setting changes and new plug wires. (only changes) This time, who knows.
sure, tires , air temps all are huge factors, but i have ways of knowing when thats an issue as well.

we do know many of the components of the holbert engine. we know the cams are 86's, we know ive set them like i found them from the factory set at 2mm each for cam timing, we know the ECU boxes are stock 928 S4 and we know i run Amzoil

The help im looking for, is what is coming through this thread. i really never knew about the effect of the knock sensors and the cam hall sensor. if i have a short and its retarding timing, it could be an easy fix. That darn flappy was driving me crasy, unitl i just pegged it open! now, im thinking of how i remove the knock sensors from the equation!

mk

[QUOTE=michaelathome]Mark,

I know that I am not the first to say it but you are a wierdo. That being said I am a wierdo, we are all wierdo's with 928's. Your 928 may just may be a little different than most.

IIRC air pressure changes in your tires can rob horsepower. Personally I think you might have had an off day. 1 second over a lap on a track is tiny. Did you buy a new pair of shoes or have you gained a lb or two? Did the tires change or did you have a new/old worn bushing.

Sorry to throw it out there but NONE of us can really help you to say that, "Your valves are not timed correctly" or "your car is to heavy or light" till someone pulls it apart.

I love what you have been able to do w/the car but untill the rest of us know what is in there it is a little pointless IMO to ask us what is wrong.

lorenolson888

sounds significant... but couldn't it just be the car wearing a bit from racing... It seems reasonable that the car will lose some power as she ages...

though if it starts to lose more power I would be alarmed.

You recently posted leakdown or compression results... were the numbers good???

Good luck on trouble shooting.

Loren

mark kibort

thanks Loren!

yes, compression was pretty good, equal to what the car had when i bought it 6 years ago, so im wondering, where is the wearing part?
however, i have never done a leak down. Ive heard there could be an issue that would go undetected as you use so many cranks to do a static compresssion that could mask an issue, right?

oil showed no bearing wear, and filter is always very clean. (no bearing binding)

I guess it cant be fuel pump or filter, as fuel mixtures look pretty good. so, im thinking timig caused by those knock sensors or some other computer issue.
WOT switch? is there one on an S4 that does something??

thanks,

Mk

michaelathome

Mark, can you release my statement from page 1? I am still trying to get the youtube video imbedded and you are really messing the thread up.

BTW any 944 or 928 owner ought to know about the sensors if you have had your hands in the engine. They just have different names based on the car.

Again I don't know much but oh well.

Michael

mark kibort

Hey i dont know much either!

However, this list sure helps!

what did you want me to do on the prior posts to help you out with the embeddded youtube link? i edited out the link at the bottom, did that help?

mk

mark kibort

i just checked timing. at 5000rpm, pullling the cam Hall sensor didnt do anything, either did pulling the rear knock sensor. max timing is near 38-40 degrees in neutral reving it up 4500 to 5000rpm.

are they both bad, or are they really used when you are in a loaded condition.

anyway of bypassing the knock sensors?

mk

michaelathome

Good luck Mark!

I have to say that I don't know enough to be of much help.

I do think that:

Is pretty close to dead on.

Michael

Bill Ball

John speake will have to verify if you need a load. The references I see in the tech documents are a bit vague.

The GT brain will allow us to run the diagnostic tests. They will tell us which knock sensor or whether the Hall sensor is bad individually with no need to run the engine under load and check timing. Since I'm not regularly employed anymore, I can come over anytime and run the tests.

I don't know of any way to bypass the sensor signals.

PorKen

Upgrade to LH2.2/EZ-F.

mark kibort

Hi Bill,

I would love to do this at a convienient time for you. I would hate to see you make a special trip, but if you are in the area, i am available.

I'll have a lot of time over the next 3 weeks or so, (hopefully, i wont after labor day ) So, it would be great to see if i have a problem with your help.

By the way, could i get some knock sensors and plug them in to the connectors and not have them connected to the block? I dont understand the function of the Hall sensor, so im sure that could and would have to be replaced if it is not functioning correctly.

mk

UKKid35

The knock sensors can be past their best and still work to some degree. If the outer insulation is cracked/missing in places (usually next to the block) then noise can be introduced to the signal. However simply disconnecting them will still cause an ignition retard.

drnick

i will put money on this being electrical, not mechanical. the 20hp loss is just enough to notice but not slap you in the face obvious - enough to make you question but not to be sure if youve lost power - very frustrating. i wouldnt be in a hurry to loose the knock sensors, they do good work when they are ok.

John Speake

Hi Mark,
The engine relies on "normal" engine noise signals being present from the knock sensors when the engine is running, if there aren't any such signals, then the EZF will retard the igntion, blunting the performance.

The same applies to the Hall sensor.

The knock sensors are piezo type devices, they give a small output signal whose amplitude is in proportion to the amplitude of the signals they pick up from the block. The EZK has clever circuits that filter out "normal" engine noise and detect knock noise.

If a sensor isn't giving its normal output, then the EZK retards the igntion because it knows you don't have a working knock detect system.

The knock sensor output levels are very low, which is why the leads are screened. If the insulation falls off with the heat, then the screen braid can run on the heads and the EZK assumes the knock signal is compromised and so does the igntion retard.

Try a battery disconnect, and then do a test drive if possible. Thar will clear any knock codes in the EZK and may restore the performance for a short time.

Whene you say the A/F is OK, is it the same as previous ? What A/F WOT are you running ?

mark kibort

Thanks John,

Great info. so, no way to "short" out or fool the computer. I guess we just need to confirm whick one of them is bad and make sure they are all working.

yes, the braiding is (was) gone on all the wire portion just as it peaks out of the valley of the engine. i put some heat shrink wrap on it all the way to the connector and used some electrical tape too. the insulation down to knock sensor looks ok, from what i can see, but just as it attaches to the knock sensor, i can see some of the braided insulation.

the Hall sensor connector and wire looks very good. i guess we need to just see if it is producing the proper signal. i still dont understand what that signal is used for.

Do you or anyone know what the timing should be if measured in idle and reving the engine? it doesnt seem like the timing is being retarded for either of the knock sensors or hall sensors being disconnected, but if one is bad, it might be the reason there is no change.

what if i put a knock sensor and bolt it to something i can easily acces on the block or heads, so i dont have to pull the intake? (is there an easy way out if we find it is bad)

Looks like Bill solution is the best way, unless i can get some definiative data regarding static timing tests.

mk

edit: as you can see from the dyno run, the AF ratios are in the 12 to 13:1 range. same as they have always been for the last 6 years of dyno testing.

John Speake

The Hall sensor tells the EZK which knock sensor to listen too. This way the EZK knows which cylinder has just fired, and so when knock occurrs it only retards the igntion on that particular cylinder.

But where there is faulure of the knock or Hall sensor, all cylinders are retarded by 6 deg at medium load medium rpm up to max rpm.

The advance at idle should be around 10deg. But bear in mind that the 6deg retard is only activated from medium to high engine loads.

I've attached a screeshot of the 98 octane S4 cruise EZK map. The figures shown are actual deg advance. But for the reason mentioned above, revving the engine at rest isn't going to tell you anything useful.