My Knock Limited 3.4
 

Guys,

I have a 3.2 with the usual modifications:
98mm Mahle Motorsport pistons giving a static 10.3:1 compression ratio
993SS cams set at 1.7mm
Twin plugged but otherwise standard heads
Modern EFI running speed density with the standard 3.2 intake manifold
Headers and sport muffler

It runs great and is very responsive in the mid range but the top end is fairly uneventful as it is knock limited when running 98 RON fuel to the point where I can only run a total of 15.5 degrees advance at maximum load (VE at just over 100%). I’m also running it pretty rich (0.81 lambda) at load so I can run this advance safely and avoid knock.

I’d like to get more top end power but realise that any modification that increases the VE such as ITB’s will likely result having to pull more timing out. Can’t see how changing the cam timing can help me either.

So what are my options?

1. Do I try some avgas or racing fuel as an experiment to see what extra timing I can run and what extra power is available? If the extra power is negligible, I’ll leave it alone.
2. Do I first check my piston to head clearance to make sure I’m getting the most from what little squish these engines have?
3. Do I machine the pistons to reduce the dome and reduce the compression ratio to say 9.8:1? I expect I will lose part throttle power doing this but will the reduction in compression ratio be enough to allow me to run the normally accepted 26 -27 degrees advance for these engines with twin plugs and pick up more top end?.
4. Are there head modifications that I can do to improve combustion speed and reduce their predilection to knock?

I’ve done lots of searches but I haven’t come across anyone wanting to control knock after their engine is hotrodded. Plenty discussion in the prebuild stage!

Help!

AG Bell in his book "Four Stroke Performance Tuning" talks about this being a deliberate strategy employed by OEM's to maximise performance rather than maximise peak horsepower which I found interesting as street cars spend the vast majority of time at part throttle where responsiveness and fuel economy are therefore more important.

What did surprise me is his statement that this trade off only costs between 4 and 6% in peak power in his experience.

I was expecting the trade off to be more. Does anyone have actual experience with 911 engines?

Twin plug it.....

i think he said it is twin plugged already

I believe your 0.81 lambda (about 11.9 AFR) is too rich at WOT. It should be more like 0.85.

Also you can retard to cam timing and move the curve up a bit.

And why not use 100RON (some places even has 102 these days).

You should be able to run 24 degrees all in on a twin plugged 911.

I used DC24 cams in a very similar package and was quite content. Very street friendly with a little high end kick. DC 40 or mod S cam (with a 113 LC since your're running speed density) is a lot of fun and will give much more top end, but it's always a compromise at the lower end.

I would start by trying to under why it’s knocking. Did you miss you CR mark? What’s type of #’s do you get from a compression test? Do the plugs show any signs of oil? Oil in combustion chamber is pro knock. It does not take much. If you see sign of oil solve that first. Confirm both plugs are firing. Run copper plugs not platinum or iridium. The copper is a better heat conductor and helps avoid detonation. Try colder plugs too. This too helps detonation.

Assuming no oil in combustion chamber and you’ve optimized spark quality I’d try retiming cams. By closing intake later you effectively reduce compression. Not sure it would be enough though. Jumping up to bigger cams like the DC24 would help that even more. I love that cam!

Lastly there’s some great options for home brewed race fuel. Not sure where you’re located by here in the states Torco is a good option. I love E85 too! That’s much more limiting (depending on where you live)

following this one. Have similar set up with same knock issue and had to run similarly low ign advance. stock cams

10.3 compression should be no issue for a twin plug motor on 98 octane. Sounds like something else is going on. Could your knock sensor be intervening prematurely? (I.e. threshold for how it defines knock being off)

flex fuel sensor and e85?
but as above see if there is an underlying reason

I second the possibility the knock sensor may not be telling the true story. My 3.4 at 10:1 single plug with 964 cams timed at 1.3 mm has no issues with knock with 91 octane. I'm running MS2 management with advance full on at 32 degrees and light load & throttle up to 37. A/F ratio never richer than 12.3 or .84 lambda

why does some engines knock while others don't in same configuration ? oil consumption was mentioned...

It is all about the cam ... at that static cr you can run a lift at TDC of 3,0 and 2,5 mm.

This will reduce the knock and open up you engine to a 25 deg advance.

If your exhaust is good you will gain a lot of power over the complete rev range !

Best reg.

Dirk

mr edelweiss; would that be the other way round? Earlier timing should increase cylinder pressure. Which is right ?

have you confirmed its knock and not other engine noise your picking up with your sensor? Check your plugs.

Guys,

Thanks for the comments, you have given me some food for thought.

For fuel, I'm limited to 98 RON as nothing else is readily accessible.

With regard oil consumption, there is nothing untoward here with the plugs and exhaust ports dry.

The engine has been tuned on a dyno using a knock box and headset by a very experienced tuner who deliberately tuned the AFR and ignition to get best power safely and not to have knock sensor safety function trigger ignition retard.

I'll investigate the cam timing "opportunities" further and get back to you with what I come up with.

Thanks!

I would think that retarding the cam timing would lower the dynamic compression, reducing the tendency to knock. Opening the intake valve later would close it later, shortening the effective compression cycle. Retarding the cam timing is generally considered to improve higher RPM performance, at the expense of lower RPM power/torque.

Another consideration would be to go to a colder spark plug. This would improve the knock situation by reducing the likelihood of the hot plug pre-igniting the mix prior to the spark event. I believe this would also allow you to try a leaner AFR, as less fuel would be needed to cool the chamber.

Likely your tuner is not familiar w aircooled 911 engines and his listening device is picking up engine noise as knock. Granted a euro 3.2 L advertised at 10.3;1 (single plug) is actually more like 9.6-9.7:1, but they run 93 RON Euro pump gas all the time on these. 15 degrees is nothing. It will pick up a bunch of power if you raise it a little and test in small increments...keep adding advance it till power stop increasing , pretty simple. Then for safety and nuclear hot heat soak like -stuck in bumper to bumper traffic, retard it some for safety-to tuners taste. Knowing where the sensitivities are from experience in this regard is everything.
You'd also better tune it at full normal hot operating temp to know it is set properly for typical worst case conditions. This is where a dyno comes in also.

Your engine is twin plugged, and if your builder really CC;d it out at 10.3 that is a bit higher than The Euro 3.2s , however at 98 RON fuel you should have safety by a large margin to run much more timing than 15 degrees and get proper power levels normally associated with those parts.
You should run a heat range or two cooler spark plug but make sure that in your climate you know what is suitable baseline stock. As a general rule a twin plug motor will be 5-6 degrees less advance required than it's single plug equivalent, considering" the rule of thumb" that one can run 1 point higher compression ratio safely w twin plugs.
So if 9.8 is safe on pump gas, 10,8 is safe twin plug on the same fuel. I am more conservative in our hot climate here in So Cal w poor 91 Octane fuel.

Kevin
GAS Motorsport

a normal SC engine runs 22 degrees or so at higher loads. Accounting for difference in timing 15.5 degrees is not far off

Muzz,
Some interesting comments from John Dougherty here:
http://forums.pelicanparts.com/911-engine-rebuilding-forum/194042-whats-hottest-3-2-cam-you-would-use.html

trond
A guy wouldn't be running that timing on unleaded 98 race fuel and would likely make much more power w a fair bit more timing, even leaving some safety margin in there.

I am sure he didn't buy all the not stock parts to go stock on the tune. W The stock timing you note and 98 it likely makes less power than it would on 91 octane.