Help with Knock.

I just got back from the chassis Dyno (load bearing). My first engine and first time on the Dyno. The engine pulled all day without a hick-up and made just a shy less than what the cam company said it would. Power was 302 SAE HP at the wheels @ 5800RPM and Tq was a pancake flat 305ft/lbs from 3400RPM thru red-line. Tq falls off at 5400RPM and HP peak was at 5800RPM. The engine has TFS 175 aluminum heads; comp 8-502-8 cam; stock roller lifters; Comp 1.5 roller rockers; LT1 intake; stock throttle-body w/air foil installed. I did some very minor port blending on the LT1 intake to match the heads (felpro 105s). And other than a TFS SFI balancer the short block is stock. I was a little worried after the first pull that the combination was going to spin too high. But for a 305ci/5.0L TPI engine I am very pleased.

On to the problem. During the dyno we could not find the knock limit. We put 42* into the EPROM for RPMs 2400 to 3800 and all along the MAP scale. NO KNOCK!! After two tries to get some signal from the ECM that knock was happening we gave up assuming the knock sensor or ECM or something was wrong. Presonnally I never heard any knock nor saw a Tq drop during these test, but the operator said it should be knocking with these settings we must not be picking up. I don't think we put the engine into knock really. Not wanting to risk anything we proceeded tuning the engine very conservatively with a max timing of 28* under WOT conditions and around 11.5 AFR. The engine still pulled to within 20 HP of what was predicted by CompCams. Drivability was good, some trailer hitching when cruising slow, but pulls like hell to red-line. No stalling, stumbling or surging either. The cruise timing was also left a little low because of the lacking knock signal.

Now, on the way home during the 3hr drive on the autobahn I let the car run. Average speed was above 120MPH. The car has a T56 now so I was able to cruise at 130MPH below 2800RPM. On three occations with my foot on the floor taching around 4200 to 4500 RPM (peak Tq range) I heard a screetch, which sounded like a belt slipping, maybe a very fast coffee can rattle. My buddy and I couldn't agree which. Which was immediately followed by a drop on power and the SES light coming on. KNOCK RIGHT?!

The knock sensor is original for the car. Circa 1991 with rust and all. And is relocated to the driver side due to the Hooker 2210s. To get rid on the SES I had to restart the engine. ECT was dead on at 193*F with the thermostat. I have done some test
drives since and still no knock in my datalogs.

Does the ECM set a code when knock occures?
Tunerpro V5 using all the standard XDF/ADX taken from Moates, modded stock bin; there are no knock counts at all in any datalog.
The engine revs freely and is a blast to drive.

Also, a big thanks to Craig for his set-up and the work he has done. Without TP and the Autoprom the tuning of my car would not have happened.

Hotrod

 

> Does the ECM set a code when knock occures?

No. Blink out or pull the codes with a scan tool.

Are the correct spark plugs being used? They should be 3/4" reach with a flat gasket seat such as Autolite 3924's.

RBob.

 

Thanks Rbob,

Plugs are the autolite 3924s with 3/4in reach, flat gasket or ring as you recommend. The only code that has showing up is 12. And it resets by restarting the car. On the dyno we got it once but that was when we hit the engine speed fuel cut off limit we set in the EPROM.

Is there a way to disable the knock routine with a flag? All of my settings in the other tables/scalars are stock. So the flags should as well, but who knows I might have checked something one night by accident.

Maybe what I am finding is not knock. I have removed EGR and AIR, physically from the car. Only the air pump remains. I even was able to block off the EGR passage in the heads. I disabled both using the temp setting in the scalars. Is there something else I am missing?

Hotrod

 

OMG! I wanna go to!

With the car running and TunerPro connected to data hit the block near knock sensor with a hammer and see if you get a reading?

Usually the problem is over sensitive or false knock. No knock when trying is usually a failure in knock circut...

 

Can disable knock retard by setting a temperature parameter high (same as was done with AIR & EGR):


There is a forced knock routine in the $8D code. I would imagine that if it was enabled you would have hit it on the dyno. Can disable it too by setting the enable CTS high:

With the high timing test that was done on the dyno, it isn't unusual to not get knock. The Corvette '113 heads are the same way. Too much timing and the engine gets rough but doesn't detonate. Although it does pepper the spark plugs.

RBob.

 

Big thanks for the examples in the code Rbob. The plugs have not been pulled to check them. I forgot that they could tell the truth sometimes better a sensor. It's Sunday night over here, so I will have to get at them later in the week.

Hotrod

 

Well, I checked the plugs, no peppering. Seems some of the plugs are reading leaner/richer than others. My #3 plus is the leanest by showing no tanning at all and only a light bit of carbon on the ground strap and body. The other plugs read decent with the richest plug having a moderate coffee color. All of the ground straps are showing some blueing though. But the blueing doesn't reach the body of the plug.

Further info on the ignition. I have teh stock module and dizzy with an MSD cap/rotor/8.8mm plug wires and MSD TPI replacement coil. I am seeing a considerable amount corrrosion or white powder on the copper contacts between the rotor and cap. I say considerable because these parts are less than a year old with also less than 2500 miles on them. I know some of this is normal, but I don't think this much is normal.

So I don't see any evidence of knock on the plugs, I think I might be having an ignition module problem or ECM fault. The plugs are gapped out at 0.045". I am going to reduce the gaps to 0.035" and install the 3923s to see if I still get the SES under WOT.

Also, I think a limits from dyno day was the octane I had in the car compared to what I am running in it daily. I stopped at a gas station that sold 105 fuel (pump) and put that in prior to dyno'ing the car. So now running around with 95 octane at my local gas station the car may be asking for a better fuel. Compression is at 9.6:1.

Does anyone have any other ideas I may be missing?

Hotrod

 

thought:

will a misfire usually trigger the knock sensor into action?

i seem to have experienced that one myself a few times.... plugs were old/crusty and gap was getting large, in high load situations i would get some intermittent knock.... change out the plugs for new ones, proper gap, no more knock retard...

just a thought, anyways.

 

Could just be something that is loose on the car. I've been dealing with innoying false KR for some time on my car. Sometime its gone and other times its there.

The smallest little thing can cause KR. Sometimes rattling that you may never hear, but the Knock sensor does, will give you a headache because of Fause KR.

Just a idea...