10:1 SR20DET in a GTiR not taking any timing!

Hey guys! I'm new here and I have a tough one for ya. I built an SR20 for my Pulsar GTiR with 10:1 pistons for autox use. Can't get ANY timing into it without (what I believe is) knock.

Here's a quick breakdown of the build:

SR20DET 54C (GTiR engine)

JE 10:1 pistons

53mm T3 chinese choochy boy

Tubular exhaust manifold with Precision Turbo 39mm wastegate @15psi

Innovate LC2 (AFRs currently hitting 11 and tapering down to 10.5 at full boost)

NIStune ECU

Frozen Boost water-air intercooler

AEM Methanol injection kit with 50/50 water meth 500cc injector

91 shite pump gas

Inadequate half-rad causing my coolant temps to be between 85C and 95C

Obviously I'm planning on fixing the cooling ASAP, but even with that considered look at this video of me doing some light pulls. I am new to using my ears. I've tuned my DSM on 30psi for years but used the onboard knock strategy. Am I going crazy?

Pulls video - https://youtu.be/JjcZkqvIh9w

I then swapped out the timing tables from my super conservative timing to the stock and did a back to back pull in 5th gear from 3000 to 5000.

During my conservative timing I thought I was still hearing little bits of knock. But then as soon as I went WOT with the stock maps my headphones just about blew my ears off. I immediately let off and aborted the pull.

Comparison video - https://youtu.be/o6i6v_54rnA

SO my question is am I being too sensitive when listening for knock considering how loud it is with the stock aggressive timing table

Like I said I have a decade of experience tuning my DSM on high boost with meth, but this whole using-my-ears thing is new to me.

I appreciate any advice you guys have

Not an aengine I'm familiar with, but some generic questions to help those that are. No offence intended with a chicken-egg thing.

I can assume you physically checked the actual TDC and the TDC mark were aligned, and that you double checked the ECU TDC is correctly aligned with the physical TDC?

What camshaft have you got and have you checked the timing is correct?

Have you tried using some higher octane fuel to see if that has an affect?

Sorry ya I've checked actual timing and its bang on. Done a boost leak test and its fine. Stock camshaft. All the basics have been checked off the list. I am actually grabbing some aviation gas (100oct) tonight to see what the effects are. Will report back

i have a metric f-load of tuned factory ecu compatible roms for gtir's (had one originally and loved it and tweaked it a tonne) but ive never come across someone running that high a compression on a gtir engine before.

Maybe check out some honda turbo maps - they usually run high compression and a tonne of boost so may give you some hints at how much you may need to retard the ignition under load?

edit : just watched the youtube vid ouch yeah i see what you mean about no timing and lots of knock sounds :\

Just for my own curiosity - whats your peak timing number at redline and at peak torque? And why move away from the factory t28 that was good for 1.4bar and 200+kw to the wheels with its good response compared to the t3?

Something you could also do is replace a plug because youve already had det on them and do a run with your conservative timing then pull the plug and see if you can see any signs of det on it - shouldnt need to swap all 4 at once for that testing (hope you dont have anything over the cam cover like factory :P)

edit2: love your channel music and cars - gotta love it :)

I decided to go with the 10:1 because the engine shop I was working with on the build has had good luck on RB's with 10:1 and we thought it'd translate. Now I'm just looking at it as the toughest (and maybe best) tuning lesson I've ever had LOL.

So I've tried right down to 2 degrees at peak timing and I still heard what I thought was knock. That's why I'm wondering if the little bits of noise in the first part of this video are actually something to try and eliminate, or if I'm more so just listening for the bigger peaks in the second half.

Here is the current timing map I'm running. 15psi gets me to the second to last load column. I'm not hearing much of the louder spikes of knock with these values, but I can still hear little bits of zappy zaps hidden amongst the engine noise.

I went with the T3 style turbo because that's what came with the bin of parts I got the car with. The car was originally converted to RWD and was built as a show car / burnout car, and I'm trying desperately to convert it to a race car haha. I'd love to go with a more responsive turbo set up but I'm currently so far into this financially it doesn't make sense right now to swap the exhaust and turbo over.

Glad you dig my channel! I have made a living off of music for the past decade and have spent most of that living on car parts so I thought it would be appropriate to try and document that :P

Just had a look at the base map for a turbo na honda and your running about 1/2 the timing they do with similar compression

Maybe take a look in the engine building forum here and ask about the specs you have and any ideas people have about the setup and why it seems to be so knock prone. From your videos it does sound like knock (loved the youtube vid you did about reactor and your knock filtering - very cool)

Whats sort of air temp are you getting after the intercooler? your not getting way too hot an intake charge and provoking knock in some way?

Is the meth kit in use or is it just connected up? Those things are meant to be a godsend in cooling the intake charge and delaying the onset of knock from what I've seen on the internet. Altho yes - it is much better to have the base sorted out with all that turned off before you start really playing with things.

If the meth kit is enabled and running - and your running that little timing and your still getting knock i would have to say "i don't know enough to say anymore" but i'd be thinking mechanical design is causing it more than tuning at that stage :\ if thats what other people conclude for you - maybe drop in a set of cam gears so you can get them set back to factory timing and slap in a thick headgasket to lower the compression a bit? That would be the cheapest thing i can think of to lower the compression / give the ecu some more leeway with timing.

Just checked an octane conversion page so i could figure out what you were running on - your 91 octane in canada to equate to 95 octane that we get here (http://www.pencilgeek.org/2009/05/octane-rating-conversions.html) - the highest pump gas around me is 98(no ethanol) here (94 in canada) so it seems you have as you put it "shite pump gas" alright :\

Would it be worth it to disconnect the boost pipe from the inlet manifold and run the engine with no boost and see if you still get it knocking when its all running n/a?

anything else i can think of would cost a tonne of money sorry :\

I need to get an air temp sensor into the set up but I live in a very cool area and the intercooler setup I have is supposed to be very efficient. I know until I have actual numbers that doesn't mean much, but I'd be very surprised to see inlet temps anywhere near 100F. If my talon is any indication, I am usually 60-80F cruising around.

One thing I did notice is that on a warm compression tests cylinders 1,2 and 4 were all pretty much bang on 180psi, though cylinder 3 was closer to 200. Thats within about 10% so I didn't think anything of it. Overall didn't think those cylinder pressures were crazy really...

The meth kit is running a 50/50 water meth mix. I run 100% meth in the Talon so I think I'm gonna try that.

Basically here's my plan of action:

-Swap in brand my new ID 1050x injectors in case maybe one of the stock injectors is a bit weak causing one cylinder to be leaner than others

-Try 100% methanol in the injection

-Try 100oct aviation gas (I have 20L in a jerry can waiting to go)

-Give up, sell all my cars, and take up competitive knitting

The cam is definitely phased correctly?

I would confirm intake manifold temp if you haven't already.

Piston to bore clearance is in spec?

Are the plugs colder than standard?

Not getting valve bounce or something silly?

Cam is in phase yep!

Will get the IATs logged asap.

Id have to check with my builders on the clearance but the whole thing was just professionally built. It is a little loose on the clearances because of the forged pistons but it doesn't sound like piston slap to me (my DSM is VERY loose so I know the sound)

In terms of valve bounce I highly doubt that since im not really revving it out and the GTiR motors have twin valve springs and solid lifters.

Im running 8's for plugs so pretty cold.

I appreciate all the help guys! Looking forward to sorting this and then having the experience to draw on in the future.

Ummm - cylinder trim #3 to be a few % richer and run a degree or 2 less timing? to try to eliminate that as a cause?

colder plugs again? ngk bcpr7es always worked well for me on a road car and were cheap enough.

totally get the iat sorted but if you have a/f ratio that looks ok - that should be ok for "not long term" testing :)

maybe invest in one of those auction site $20 bore scope cams you can plug into a cellphone and look in the cylinders to see if you can see any weirdness in one and not the others? (pitting / carbon up / wetness etc) what do the plugs look like - even across them all?

Sorry, cant help with the knitting - sold off all mums old knitting machines when she passed away - even the one that you could feed a scanned pic into it and "print" out a 3 colour pattern - that would have won a few competitions for you :P You'll have to keep playing with cars im afraid :)

Please do keep us all updated - im very curious to find out what is the cause of all this - im guessing your 100% meth and avgas will stop it (or it had BETTER stop it) but please keep us updated :)

If you are losing a lot of oil from the compressor side seal of the turbo it could cause issues too.

Were there valve pockets cut on the pistons, and were the edges broken/smothed if so? There were a bunch of guys on an Australian forum modifying LS motors with big cams, head swaps etc on pump fuel and I can't recall if they were were just relieving factory or aftermarket pistons but a couple of them did have hot spot problems and not accept timing until they smothed the valve reliefs.

Hmm that's interesting! I'm not losing oil through the turbo seals that I can tell. I've had the intake apart since running it and there's no sign of oil.

They were JE Pistons with valve reliefs as you can see below. The edges weren't terribly sharp but I didn't smooth them. Interesting point.

I put a set of ID1050x injectors in tonight. Tomorrow I'm going to try those and then try pure meth and then try pure meth with AV gas. I'll record everything and post the results here.

Also Lawrence - nistune doesn't allow individual cylinder trims from what I can tell. I will try some even colder plugs though!

OK, it seems you've some experience with the engine and have had previous builds. So, what's different this time?

If you're using the same ECU and software, are you using the same manifold pressure sender and is it working/calibrated correctly - if it isn't pulling the timing that could be why?

What I'd suggest is disconnecting the turbo' from the intake and running it in NA form to see if that makes a difference, if it does/doesn't affect the problem, might suggest further options?

Probably not the answer you are looking for but I really think those edges on the pistons are strong odds for your problem.

Swapping fuel isn't really fixing it, if it's only for racing I guess it may avoid the problem and it will make more torque but being able to run on pump fuel would be a lot more convenient?

Gord - this is my first experience tuning anything beyond the 4G63 in my DSM. The intake system makes running it NA pretty difficult (see picture) but I'll see if I can rig something up. Regarding the map signal - It's a maf based car.

Michael - the idea that the pistons' sharp edges could be causing it both terrifies me and makes sense haha. I'm just switching fuels for diagnostic purposes and will be running pump gas + meth injection otherwise. If it won't work on that combo it's pointless to me.

OK I just got back from doing some pulls while spraying 100% methanol, and then swapping over to 100oct AV Gas. With the same timing tables, there's no difference between me running 50/50 water-meth / 91 oct, 100% meth / 91oct, and 100% meth / 100oct AV Gas.

I'm starting to learn towards the little bits of noise not being significant. Again going back to the difference between these two pulls. The plugs never show signs of detonation from what I can tell.

I will put the new runs together into a video to share.

Here are those pulls https://www.youtube.com/watch?v=oEVBm5KHNYk

Here's some more info incase someone has any thoughts.

Recorded some pulls both on the knock sensor and with a microphone in the engine bay.

https://youtu.be/3sVbY6Q3MJw

Try a pull with an open wastegate. You could apply airpressure with a handpump or something on the actuator to open it up.

Ah, going back over the thread, I noticed something I missed before - "One thing I did notice is that on a warm compression tests cylinders 1,2 and 4 were all pretty much bang on 180psi, though cylinder 3 was closer to 200. Thats within about 10% so I didn't think anything of it. Overall didn't think those cylinder pressures were crazy really..."

For a fresh engine that's actually a rather large variation, to put it mildly, and definitely something to further investigate with a cylinder leakage test. If there is a relatively even leakage it would strongly suggest some mechanical variation causing higher compression which may be the root of your problem. It may be a smaller combustion chamber (did you CC the head?), deck height (did you check them?), worn or incorrectly ground camshaft lobes (did you check the lift?).

I didn’t check the videos but keep in mind sr20’s are typically pretty noisey engines. The low load cruising sounds you’re hearing are probably normal.