I have been trying to track down why my car is knocking. At WOT it is pulling up to 7 degrees of timing. On paper, I should only have 9.3-9.5:1 compression and the highest gas locally I can get is 91. I'm trying to see if a higher octane will rid my car of knock issues.

A local racing guy (runs dirt modified and such) told me that he's used fibreglass remover and says it's basically just toluene without the junk that paint thinner has in it. I'm trying to locate a source for toluene but it's been pretty difficult.

I know I can use race gas (has to be unleaded or it'll kill my O2 sensor right?) but it's costly at $58/5 gal.

Has anyone tried anything similar with good results? Is this fibreglass remover a good alternative, or will it kill something internal?

I also run nitrous, so the higher octane would be better for that too right?

 

Toluene is nasty to your kidney & liver, that is why you can't find it in the hardware store. Big brother knows whats good for you,... just ask him.

If 91 octane won't get rid of it,.... do you really expect 100 octane to get rid of it? Your compression levels are saying you don't need it. I'd suggest looking for another cause of your knock. Simple stuff like seafoam. Checking base timing. Plug condition. etc.

 

Sunoco sells Ultra 94 (94 Octane) in Ontario. I don't know where the closest one to you is. Their web site may help you find one.
[HTML]sunoco.ca[/HTML]
As cars get older, the Octane appetite of the engine increases as wear increases. You do not mention the mileage on your '89 so I do not know if this is your issue.
It would not hurt to find a Sunoco station and try a tank of 94.
The other problem with additives is that too much of them can give too hot a flame front and starting burning your valves.

I did that on my '69 when I was much younger and naive and used benzene to boost Octane. It is one of the best Octane boosters but it is CARCINOGENIC.

Try and find the underlying cause. The engine should run fine on 91.

But if you still want to go the Octane route, try an Autoparts store . The commercially available octane boosters have mixing instructions to ensure you do not harm your car. There are a number of them. Take your pick.
If they don't carry Octane boosters in Ontario, take a trip to NY

 

Forget the Octane booster-you need to look for what's really causing the problem. How's your timming?

 

A lot cheaper, easier, test is to try what I did. With premium gas in the tank, retard the time to below the base spec of 6*. THEN look for knock counts. With 91 (the highest we have here) in my tank, and the timing set to zero, I got counts, revving it in neutral (no load). That is a sure sign of "FALSE" counts. Many things under the hood can set off the knock sensor. Knock counts don't necessarily mean knock. Reviewing the method to check the knock sensor, tells you that. If you DO determine the counts are "FALSE" there are two ways to handle it. You can do an exhaustive check to find the source of the noise, Or simply disconnect the sensor. After I disconnected the sensor, I was still able determine my optimum timing, by avoiding audible knock.

 

Indeed you should try backing off base timing (and check the TPS base value at base idle) to see if the knock is real or false.

If real, you might try running a 160 deg F T-stat & fan switch as lower temps will reduce the tendency to ping.

If you are not hearing ping but seeing knock retard on the scanner, consider upgrading the ESC module per my site to eliminate false knock.

 

all those suggestions are good.
It may not be the octane.
If you do want to in crease octane however, the booster's in the store are a waste of money.
If you run Nitrous (As I do from time to time) you will increase the need for octane during the spray. I hope you retard the timing with n20.
Anyway, I use an Octane ADDITIVE from Torco race fuels.
These guys make fuel. Nitro methane, 110, 116, etc and will deliver 55 gal drums to your door. So they make an additive with instructions on the side of the can. You can mix and make 97, 100,104, 110 unleaded.
My two cents.
Matt383

 

why will they do it in neutral (knock) more than under a load?????

 

Thanks for all the replies here guys. I'll try to answer the questions posted.

I decided to get away from the fibreglass remover solvent and went and bought some Sunoco 110 unleaded racing fuel from a racing store. I went for a 10 kilometer ride and it still pulled timing from knock. Maybe I gotta let it mix better, but I could smell that sweet smell in the exhaust. It reminded me of being at the track where the crazy fast cars smell like that.

Ok, to try to answer the questions now...

Base timing at 6*, new NGK TR6 plugs last weekend.
I'll try that seafoam stuff or maybe try motorvac.
I've been to a few Sunoco stations that carry 94 and have filled up there and noticed that the car seemed to run much better, but none really close to me.
My 383 has around 30 000 kilometers in 4 years on it. Original engine was toasted - cracked block and such.
I get knock counts and timing retard from revving it in neutral as well.
TPS idle voltage is 0.63V / WOT is 4.53V (No trouble codes, I initially set it for 0.61 V @ idle, but I must have adjusted it since then.)
I run a 180 chip and the temps are always cool in my car. 175 highway and 180 - 190 ish in town.
I upgraded last year to a new knock sensor.

So... if I get knock counts and timing retard in neutral (no load) is that a good indicator of it all being false?

 

I'm not sure what your reference to "they" is. But to clarify, there is no way my 9.0:1 engine should produce spark knock (detonation) under no load, with retarded timing, on good gas. Since it can't be knocking, but knock counts were observed, the counts HAD to be the product of other under hood noises. The test was to eliminate real knock as the cause of the knock counts. The ECM reacts to "counts". One received at the ECM, it has no way of determining if the knock is REAL detonation, or false counts. To keep the ECM from retarding the timing, needlessly, I disconnected the knock sensor. Currently, I run a base timing of 14* BTDC, on 85 octane gasoline. No audible knock. I HAVE had audible knock, when it was too far advanced. I've been racing it this way every week for 5 years. It only keeps going faster.

Did that explain it? Was I able to make myself understood? Questions?

RACE ON!!!

 

Black 89, if you find the answer please do a post. I have a very similar problem.

 

You posted while I was typing, so I didn't see this until I finished. Yeah, I know I'm slow. In answer to the quoted question, ABSOLUTELY! If you have any doubt, perform my test.

RACE ON!!!

 

By "your test" do you mean back base timing to zero and then look for knock counts in neutral? Or do you mean disconnect the knock sensor and listen for audible knocks?

Problem is, my car is soo friggin loud and the interior resonance is on the border line of insanity, probably due to the mufflers not having tips and not extending past the back bumper.

 

False knock counts can come from the sensor picking up some local noise or from a bad ESC module.

You can listen for rattles or other noise sources when reving in neutral to maybe detect some metallic noise which could trigger the knock sensor.

On '85-'89 L98s the knock sensor feeds the ESC module which in turn reports knock to the ECM. Gordon Killebrew recommended a replacement ESC P/N which cured my false knock problem a few years ago. If you don't find any source of metallic noise try replacing this sensor.

 

I'm pretty sure I've replaced that ESC 65Z01. I remember reading one of your posts a while back, like last year or so and replaced it.

I'd love to just disconnect the knock sensor for a quick run down the 1/4 and reconnect it to see how much performance is gained/lost without timing being pulled.

Easy to disconnect or bypass? I am no electrical genius, so by using resistors and such, I'd be lost.

 

what kind of timing chain are u using ?

 

I looked at some of the engine specs, and I only see

Timing Chain Double row degreed with lock tabs

 

You could disconnect the knock sensor and rev the engine to see if you still get knock retard.

The ECM won't check the knock sensor for functioning till the car is moving so you won't get a code 43 doing this test.

 

Ahhhh is this true????

 

The purpose of the test is to determine if the ECM is receiving real or false counts. In order for the ECM to have the possibility to get ANYTHING, the knock sensor stays hooked up. The reason for the retarded base timing, the good gas, and the no load revving, is to have conditions where the engine CAN'T be detonating. Therefore, any counts picked up by the scanner, HAVE to be FALSE counts.

If your sensor/ESC are sending false counts, and it sounds like yours is, then the ECM is retarding the timing, needlessly, hurting performance.

The listening for audible knock comes AFTER you have determined you your ECM is getting FALSE counts and AFTER you have disconnected the knock sensor. You want to avoid audible knock, but in your case with the loud exhaust, you might try another test, that guarantees audible knock, just to verify you can hear it in your car.

RACE ON!!!