codrus

So I'm trying to figure out why my new motor is pinging.

Supertech 9.0:1, Carillo A-beams, GTX2863, FM manifold, FM2R EWG, FM 3" exhaust, FM intercooler, 99 intake manifold with VICS intact, using 95 octane (50/50 mix of 91 and 100 octane), MSLabs Basic MS3, id1000s. 99 head, P&P, BP5A intake cam on Toda adjustable cam gear (set to 0). Build thread here:

https://www.miataturbo.net/build-thr...rebuild-81161/

Current timing map:



With this 95 octane fuel, my old motor was virtually bulletproof. The differences are the turbo (previously I had a 2560), the EWG, and the BP5A cam with Toda cam gear. In particular, the old motor also had 9.0:1 Supertechs in it.

It pings from 4K up over 180-ish kpa (aiming to run up to 220 or so). AFR targets are 11.5:1 and richer, and from the WB datalogs it's happily delivering that. With the numbers it has now it's hitting EGTs over 1500 before winding out third, which is hotter than I really want to run it.

I'm pretty sure the noise I'm hearing is actually ping, because yanking stupid amounts of timing makes it go away, as does running straight 100 octane. I've burned through an entire tank of 95, put straight 100 in it, and then put more 95 in it (from a different source) so I don't think it's just bad gas.

The motor has about 800 miles on it at this point, and the strange thing is that it wasn't pinging for the first 500 miles or so. I noticed it after modifying the timing cover to get access to the cam gear and doing some datalog runs at various settings. Initially I thought it was a result of the intake gear retarding, but even back at 0 it's still pinging.

So... ideas? I have a dyno appointment scheduled tomorrow evening, and if there's something mechanically wrong with the car I'd really rather figure that out before spending money on dyno time. I'm about 90% sure the timing belt is aligned properly (checked it while modifying the timing cover for access to the cam gear), but will double check that tonight.

thanks,
--Ian

hornetball

iTrader: (4)

What is your MAP reading when you turn on the ignition with the engine off? MS1's in particular were known to sometimes be inaccurate, although I've never heard of this issue with an MS3. It should read around 100kPa. If it doesn't, then you're not running in the part of the ignition or AFR map that you expect.

Also, what do your spark plugs look like? Any evidence of overheating (really white/clean with tiny black dots in the insulator)?

Double check timing with a timing light?

Double check valve lash as another source of noise?

Just throwing stuff out there . . . .

codrus

The MAP readings are correct.

Plugs look like this, #1 is on the right:



Not particularly white, #3 is faintly pink.

Base timing is correct.

Some of the noise is valve lash, especially above 6000 RPM, but that wouldn't go away with by retarding timing or adding octane so I don't think it's the cause.

Checked the cam timing, and I'm pretty sure it's right:




--Ian

18psi

iTrader: (76)

I keep staring at the spark map and it's uber conservative.
The plugs do look kinda white, at least way less dark than something I'd expect to be running 11.5 or richer on pump. Scratching my head over here. Is it on the OEM NB timing wheel? You had an ATI damper correct? is it on there correctly?

nitrodann

On most setups you could add 10* to the entire timing map and be fine, so its not that.

Dann

18psi

iTrader: (76)

Except
it kinda is

nitrodann

Show us a dyno sheet. I reckon you have a base timing issue or perhaps something odd like a fubar'd injector.

codrus

I have an ATI damper. The timing marks line up properly -- I lined up the leftmost mark on the damper (when standing in front of the car looking down), with the rightmost mark on the timing cover. I compared that by putting a screwdriver down the spark plug hole, and by looking down inside the lower timing cover at the crank pulley, they all agree that it's at TDC with the cams in the position shown in that photo.

Jason suggested that perhaps the pistons I have aren't the 9.0:1s that I thought they were, but I don't think that's the case. They're definitely 84mm Supertechs, and they only come in 8.6, 9, 9.5, and 11. 9.5 and 11 are domed, and obviously don't match the photos I have of mine. 8.6 and 9 are supposedly the same visually, but since I had 9:1 before, if they are 8.6 now that should be less ping, not more.

I checked the base spark timing with a timing light when we first started the motor and it was right. With the Toyota COPs I had to take a coil out, wedge an old spark plug wire in it, and use that to drive the timing light. I haven't checked it again since then, but I'm not sure how it could have drifted. Well, I guess I could have screwed up the ignition calculation settings in the MS3, but I just compared them against a file from the old motor and they look the same modulo the differences between 1.2.x and 1.3.4.

Do these look right?



Dynoing tomorrow evening -- I guess we'll see what happens there.

--Ian

nitrodann

timing light.

Braineack

iTrader: (62)

did you actually tune the spark latency to 100usec?


because I'd expect that to advance the timing quite a bit on a 99+ setup.

concealer404

iTrader: (3)

Hhammerly found that his motor didn't like a lot of timing either. My last turbo car didn't like timing either. Both cars made good power. My car made obnoxious power for a 2554.

Things all three of this cars have in common:

1) Built bottom ends
2) BP4W heads
3) BP5A intake cams

Food for thought.

Ryephile

iTrader: (1)

These would be more normal settings:

Trigger Angle/Offset = 0.0
Spark Hardware Latency = 0

The 4° offset you have right now is strange, and like Scott said, having a spark latency will only mess things up unless you have an odd setup.

codrus

The trigger angle/offset was 4.0 when I got the box from Reverant, so I assumed it was correct. The 100 latency was something we put in because when in fixed timing mode the timing light showed it moving around at higher RPMs.

I don't think there's anything odd about the crank angle sensor, it's a standard 99 sensor mounted using the standard bracket. I do have an ATI damper with an aftermarket trigger wheel, but it's FM's four tooth one that's supposed to be equivalent to the factory wheel.

I'll try adjusting those values, thanks.

--Ian

DNMakinson

iTrader: (3)

Mine was 4.0 offset as well (1999 Reverant MS3-basic), and when I went to adjust it (10* fixed), I found it to be right on the money.

Reverant

iTrader: (10)

Trigger angle should be in the 4-6 area for stock NBs.

By stock, I mean stock trigger-wise, mainly the crank wheel and sensor.

If you have the stock parts in the stock location, and you require a value outside the 4-6 window, either your timing light is off or you can't read the timing off the wheel (have someone else do it for you) or you have an engine issue.

I've never, EVER, seen an NB require something outside the 4-6 window.

Braineack

iTrader: (62)

100usec latency is a lot of advance too. on my CAS setup I tuned it to 24usec and that removed about 10° of retard by redline.

Chiburbian

iTrader: (1)

Is Harold (HHammerly) running an ATI Damper? What are his spark latency settings? Perhaps there is a common thread?

codrus

Thanks Reverant!

I checked it with a timing light this morning. 4.0 is right on for idle timing. I tried to check 0 and 100 us latency, but didn't see much difference, they both looked about right. Now that I think about it, though, it's possible it threw up the "power cycle required" message and I just missed it, so I may have only checked it at 100.

I did:
- set MS3 to "fixed timing"
- hook up timing light
- rev up to 6K in neutral and watch marks.

--Ian

stoves

iTrader: (1)

I don't remember rev'ing to 6k being part of the procedure to set or read the timing for the MS i just installed.

DNMakinson

iTrader: (3)

I don't think you will see it. Look for information about latency. I think the general recommendation is idle to 4K is sufficient.

Lock the timing, then rev engine and timing should not move with RPM. If it does, adjust the latency so that your tables are set to is what you get.