all the tests were not taken in the same manner... note the throttle position on some.

 

You can also use Delphi's Ion sensing knock control. Mmmmmm ions.

http://delphi.com/manufacturers/auto...gnsys/ionized/

 

Really?

Then Rabbit, you have logging fail.

When street tuning always do the following:

Log on same stretch of road
Drive the same direction (if the road is not ruler flat)
Only go wide open throttle (adjust boost controller to tune different load)
Note your AIT and coolant temps. Also ambient air temps.

 

So are these basically COPS that have knock sensors built in?

 

Gotcha.

The short answer is that you simply can't run that much boost and timing on a high-compression motor. With a 2554R on a stock '99 motor (9.5:1) I was limited to ~12deg at 15psi on 91, so it's not really a surprise that you're seeing detonation on your setup (more compression and a worse head) at those boost/timing levels.

The '94 head isn't helping things, either. When I switched from my 8.8:1 '94 longblock to the 9.5:1 '99 longblock, I was able to run MORE timing since the head flows so much better (presumably causing lower EGTs).

22psi on pump gas on a 10:1 '94 motor doesn't work.

 

Ah yeah head design. I always foeget about that and assume all miata heads are the same.

My head is heavily ported and has big valves.

Rabbit there is a sick labonye water intjextion system on the classifieds here. Use it and run 5 defeees more toming!

 

They were all done on the same stretch of road but I got tired of beating on my mill with these settings. Hence I eased into the ones I figured were going to ping. Also I did check base timing before all of these runs – 10 degrees using the pulley tick mark that lines up with cylinder 1 @ TDC (passenger side mark on my mill).

I did get the other plugs today. I will throw them in and see what happens – I am not expecting a miracle, but I am curious.

I never gave much thought to the knock being picked up at redline and throttle shut as driveline lash - makes sense.

 

What was your AFR?

So you have a boosted '99 motor now? What PSI are you running thru it? When I put this mill together 3 years ago I wanted to use a 99 bulider seeing as they were the holy grail in CSP. Yet at the time I couldnt justify the premium. Is it a truly better head design, or is it the VICS?

 

What do you think of the Snow's performance setups? With method injection do you typically just add timing or do you pull out fuel as well?

 

I have gotten a fair amount of pms asking status and questions regarding this thread, so I figured I would give an update. The knock kept getting worse, to the point that I could no longer run 12 psi even pig rich and no advance. Over the holidays I tore down the motor to see what was up. I found heavy detonation in cylinder 4 (see pic) and light on 2. Pistons and head on 1 and 3 looked as good as the day I buttoned up the engine.

There was no smoking gun for the cause of the detonation. I am thinking it is caused by either injector 4 running lean, or I need a coolant reroute. I am not 100 percent sure on the reroute because I would expect the detonation to get progressively worse from cylinder 1 to 4, but 3 was pristine. I checked the head for taper and all of the oil squirters were intact.

Some of the Al on piston 4 melted over and scratched the bore in a few spots, yet the motor had excellent compression all the way across the board (170-175 psi at 5000 ft elevation). My machinist said the scratch might be ball honed out, but he couldn’t indicate if it would be problematic down the road. So here is my plan forward:
1. I got .020 over Wiseco’s with a custom CR of 9.2-9.5:1. My goal is to have the motor, post machining, to have a CR in the 9.2 range. I will tune the CR by decking the tops of the pistons again.
2. The motor was dropped off to get punched out. My contact who had a torque plate at the time of the original build no longer had his. Hence I machined one up at work out of some 7071 scrap Al billet I bought. $80 bucks and 5 hours I got one that beats the cast crap online. Will post pics when I get this back from the machine shop.
3. I got a 99 head that is also at the machine shop getting the works. I asked that it be minimally decked.
4. Send back the injectors back to RC for blueprint, or get a new set.
5. BEGI coolant reroute housing, KIA junkyard T stat housing, and my own hoses.
6. Clean up the homemade dual feed fuel rail. I am thinking of getting a billet one, and I am definitely running all braided lines to the rail to mainly clean up the fuel injection hose rat nest

The machinist says 3 weeks until I get my mill block, so more to come.

 

I have little sympathy for people who don't understand the glory of 8.5:1 compression. Learn from your mistakes now.

 

Not looking for sympathy. This car is competitivly autocrossed, hill climbed, and tracked, hence I don't want a slug off boost. The lowest to go in my mind is 9.0. The evo9 was 8.8, and the 10 is now 9.0. My old 85 saab 900t was 8.5:1. I guess if you run 85 octane that cr would be nessary.

I definitely won't push it as hard on pump gas on the street, or use the crap maps posted up here. Yet I will let it rip on 105. Seeing as pistons 1-3 look decent at 9.8:1, a drop of .6 should be good with the other mods listed.

 

I have little sympathy for people that can't keep a turbo OR n/a car running

OP: I had a very similar experience as you on my 00 with stock engine years back: when it died, cylinder #4 ringland was slightly melted. I rebuilt the engine with OEM pistons, and ran much better injectors + re-route. Could not replicate the problem no matter how aggressive I got, so I'm guessing one of those 2 things (or both) was the problem.