Detonation / Spark Knock on Startup

[Joesmallwood]

I have a new nuisance that I am having trouble diagnosing. After running my engine hard (runs great) and then shutting off, upon a hot startup I will hear a ping (sounds like detonation) as the engine starts.

3.6 liter
10.8CR
Twin Plug
COP
Megasquirt MS3X

I have pulled the cranking timing down to 5 (from 10), up/down on priming pulses, switched from wasted spark at startup back to sequential at startup, and a few other things, but I am out of ideas. Any ideas on what this could be and/or what to look for in settings to eliminate this?

[spuggy]

Based on my experience with detonation on a single-plugged CIS 930 and a Safeguard - most notably when the advance plate stuck fully-advanced for a 25 degree idle w/ no vacuum https://forums.pelicanparts.com/911-930-turbo-super-charging-forum/611409-my-j-s-safeguard-paid-itself-last-week.html

a) it's almost certainly not detonating at idle with no load - you need load.

This would seem to be borne out by the fact it doesn't go away when you adjust the spark.

b) if it were detonating, you almost certainly wouldn't be able to hear it anyway.

To support (b), 930's have infamously brittle rings, which detonation can shatter. Plenty of guys have run many miles with broken rings - and never realized until they tore it down for some other reason.

There's no substitute for knock detection - because you typically just can't hear it at all with the naked ear.

I'd SWAG you are hearing something else, but I've no suggestions as to what, sorry.

[Joesmallwood]

Originally Posted by spuggy

Based on my experience with detonation on a single-plugged CIS 930 and a Safeguard - most notably when the advance plate stuck fully-advanced for a 25 degree idle w/ no vacuum https://forums.pelicanparts.com/911-930-turbo-super-charging-forum/611409-my-j-s-safeguard-paid-itself-last-week.html

a) it's almost certainly not detonating at idle with no load - you need load.

This would seem to be borne out by the fact it doesn't go away when you adjust the spark.

b) if it were detonating, you almost certainly wouldn't be able to hear it anyway.

To support (b), 930's have infamously brittle rings, which detonation can shatter. Plenty of guys have run many miles with broken rings - and never realized until they tore it down for some other reason.

There's no substitute for knock detection - because you typically just can't hear it at all with the naked ear.

I'd SWAG you are hearing something else, but I've no suggestions as to what, sorry.

I've heard detonation noise from years ago working with air cooled VW engines. It sounds like that smokey-and-the-bandit-steam-rollin'-out-when-turned-off type of sound. Just one "clinkety" sound right before it fires up. That said, you could be right, it may be something else and I'll dig around to see what it could be.

I have knock detection and see zero signs of detonation anywhere else in the load/RPM range. It only happens after running hard, shutting the engine off, and restarting within a few minutes and only about 1 out of every 5~10 restarts with those conditions. I haven't captured a datalog of it, but this is something I tried to do yesterday, but as luck would have it, it didn't do it yesterday with a half dozen restarts.

There are no signs of broken rings and it's running better than ever, right now (about 3k on a fresh rebuild).

[stownsen914]

I didn't see that you explicitly said this ... does the knock detection show evidence of knock corresponding to the sound you're hearing when this happens on startup? Edit - rereading your comments, I guess you are trying to replicate the scenario to capture this.

[Joesmallwood]

Originally Posted by stownsen914

I didn't see that you explicitly said this ... does the knock detection show evidence of knock corresponding to the sound you're hearing when this happens on startup? Edit - rereading your comments, I guess you are trying to replicate the scenario to capture this.

This seems obvious, and I probably missed the obvious, but I am reasonably sure that I haven't captured a data log of it, yet. My data log is set to start at the first TPS movement above 1%. I'll need to double check tonight when I get home to see if it's possible that I may have captured an event, or not. From memory, the knock is very noise on startup anyway so it may be a bust I captured something or if it's identifiable that early in the start cycle. I'll check and report back.

[chrisbalich]

Joe,
Is it possible that your plugs are one heat range too hot?
You said it's after beating on it, then shutting it down, then starting again. But you didn't say what timeframe all these happen in. If it's in rapid succession, it could be super hot plugs. If over more time, it could be injection timing or cranking timing.
FWIW, my cranking timing right now is set to 0 and I've recently revisited injection timing with excellent results.

Looking forward to seeing you in a couple weeks at RR3.

[Joesmallwood]

Originally Posted by chrisbalich

Joe,
Is it possible that your plugs are one heat range too hot?
You said it's after beating on it, then shutting it down, then starting again. But you didn't say what timeframe all these happen in. If it's in rapid succession, it could be super hot plugs. If over more time, it could be injection timing or cranking timing.
FWIW, my cranking timing right now is set to 0 and I've recently revisited injection timing with excellent results.

Looking forward to seeing you in a couple weeks at RR3.

It happens with 5 minutes, or so, of shutting off. I haven’t noticed it beyond that amount of time.

I’ll check the plugs. This something I hadn’t thought of. That said, I’ve been running the same plugs for a couple of years without this being an issue. I’ll check.

I moved the cranking timing down to 5, from 10, and it still did it. I’ll try lower. TBH, I feel like it’s something to do with fuel, timing, or something in the tune, at start up. But, I’m having trouble zeroing in on what it might be.

[chrisbalich]

Originally Posted by Joesmallwood

It happens with 5 minutes, or so, of shutting off. I haven’t noticed it beyond that amount of time.

I’ll check the plugs. This something I hadn’t thought of. That said, I’ve been running the same plugs for a couple of years without this being an issue. I’ll check.

I moved the cranking timing down to 5, from 10, and it still did it. I’ll try lower. TBH, I feel like it’s something to do with fuel, timing, or something in the tune, at start up. But, I’m having trouble zeroing in on what it might be.

If you don't figure it out between now and RR3, bring your laptop and we'll look it over together.

[Joesmallwood]

First of all, thanks, Chris. I never leave home without the laptop. Hahaha.... It's just not time, yet.

I did some more homework last night.

This is the sound:

Hot Start - With Ping

https://www.youtube.com/watch?v=dT81tHzbbbU

Hot Start - No Ping

https://www.youtube.com/watch?v=uwP6uJS77pY

I was looking at some datalogs that I had from a couple of weeks ago when I noticed. I happen to capture several.

Here is a hot start with all knock sensor reading shown.



Another screen shot of the same event with many other inputs:



You can see that cylinder 6 has a lot of noise. but I haven't figured out what is causing it. I changed several things:

1) Reduced the Cranking fuel - cut it by about 50%. It actually starts better, but the ping was still present. These are some settings that I plagiarized from someone else's tune. It's always started well so I haven't looked at it. As it turns out, it will start well, hot, with much less cranking fuel.

2) Cut the cranking advance to zero. No change. Tried 15. No change. Left it at 5.

3) Changed from wasted spark at startup to sequential. No change.

4) Changed the idle target advance from 15 to 10 to 5, etc... No change. The idle control uses timing to control overall idle RPM and usually settles around 7~8 BTDC.

5) I noticed this happens at a CLT of 98F, also - not just at a hot temp of 180~190F. But, it doesn't happen when cold.

I did not get a chance to check the spark plugs, but I will by the weekend (busy evenings this week). This could be a cause, for sure.

I am still looking at some other things in the tune related to the startup sequence. ....scratching my head....

[chrisbalich]

Cylinder 6 it typically the hottest of the set.
I don't think you can fuel trim by cylinder based on CLT...during cranking.
It sure starts nicely. IDK that I'd mess with it unless it, honestly.

[stownsen914]

A little food for thought - I just googled pinging and diesel noise on startup and saw a random post that suggested a possibility of light pooling of fuel from a prior shutdown could result in rough running on startup. Which would seem to be consistent with what you've described (that it happens shortly after recently running).

Is it possible you have an injector (perhaps on #6) that bleeds the fuel pressure into that cylinder and the noise you are hearing is a result of that?

[Joesmallwood]

Originally Posted by chrisbalich

Cylinder 6 it typically the hottest of the set.
I don't think you can fuel trim by cylinder based on CLT...during cranking.
It sure starts nicely. IDK that I'd mess with it unless it, honestly.

It does start ok, but that pinging and puff of rich smoke makes me think there is something that can be tuned out. IDK.... I'll keep looking until I run out of ideas/options. I still have some more things to check.

[Joesmallwood]

Originally Posted by stownsen914

A little food for thought - I just googled pinging and diesel noise on startup and saw a random post that suggested a possibility of light pooling of fuel from a prior shutdown could result in rough running on startup. Which would seem to be consistent with what you've described (that it happens shortly after recently running).

Is it possible you have an injector (perhaps on #6) that bleeds the fuel pressure into that cylinder and the noise you are hearing is a result of that?

This is another good idea that I hadn't considered. I did check this several weeks ago - only by watching the fuel pressure gauge. If the FP drops, that's a sign that something is leaking and I didn't notice it at the time, but I'll check again this week. Another thing I did was added a second 02 sensor the last time I had the engine out. Both banks run nearly identical (it's a cool way to balance the banks at part throttle, too).

[Joesmallwood]

Quick Update:

I've checked the injectors, sorta. I turned the key on and watched the full pressure build to 42 lbs. Once the FP turns off (about 3~5 seconds) the fuel pressure drops to about 35 psi and stays there for several minutes, but slowly drops over time (10~20 minutes it drops down to 10~20 psi). This seems ok, to me.

Tuning: I removed almost all of the priming pulse in Tunerstudio. I left about .1~.2ms to purge any any air that may get into the fuel rails (read where this is good practice), but it is dramatically reduced from 1.5~2.5ms (depending upon oil temps).

I reduced the cranking PW as low as I could and still start and also changed it to a sweep method (starts at 50%, then 75%, then 100% based on crank events) to sneak up on the starting fuel amount during cranking.

Other things: I put all new spark plugs in it, BR8ES. Here are some shots of the spark plugs that came out. I don't notice anything "out of line".


654 Top

321 Bottom

321 Top

654 Bottom


I have tried cranking timing from 0 to 20 with no effect.

I went on a rally (Ruchlos III - which was AWESOME, BTW) a couple weeks ago and it ran great all weekend with the exception of some altitude adjustments that I need to tweak. However, it did ping on hot starts several times, but only if restarting within about ~5 minutes of shutting it off.

Long story short, the issue is still present and I run out of ideas on what to look for. That said, I'll keep trying/reading until I figure something out.