codrus

So, in prep for the track day on Friday, I reinstalled the radiator ducting, adding some weatherstripping this time to try to prevent it from rattling like crazy and ripping itself apart. We'll see if works:






Yes, I know there are leak holes on the top I haven't addressed yet. I'm planning on removing some of the factory front crash structure and taking it to TC Design to put in some DOM tubing instead. That should clear out a lot of space in there to duct the top of the radiator.

--Ian

patsmx5

iTrader: (16)

Hmm. I've been through this thread twice now.

My guess is it could be the sharp edge on the cylinder head, as that's new to all four cylinders vs the old motor.

Second thought, do you have a catalytic converter? Perhaps it's plugging up/causing very high back pressure. If so, this could explain the ping, lack of power vs old turbo.

I would also try adding fuel, it's at least possible that your wideband is off. See if adding fuel solves it, if it does, at least you're not pining any more, then you can figure out what's going on (wideband off, something else causing hotspots, etc)

Have you checked your base timing?

codrus

If you look at one of the teardown photos (in another thread I linked to), I think the sharp edges at the corner of the head were present in the old motor too. It's hard to tell with the deposits, but:



The head has been surfaced at least three times. The first time was in 2004 or so, after it burned an exhaust valve. The second time was in 2011 when I built the bottom end the first time (there was nothing wrong with the head that time, this was just a quick cleanup before putting it back together), and now after the screw.

I have an FM 3" cat in the car, I pulled it out a couple weeks ago to check for clogging, and it appears to be fine. Light passes through it, no missing chunks, nothing weird-looking at all.

The widebands on the dynos agree with my LC-2 to within half a point. They're post-cat, but at full throttle runs a post-cat wideband is pretty accurate.

I've checked the base timing and it's right.

--Ian

patsmx5

iTrader: (16)

Oh well if you've shaved the head 3 times and have 9:1 pistons, do you know what your real comp ratio is now? Probably closer to 10:1 yeah? I still don't think that's your problem to be honest.

But the problem you're facing is exactly why I'm running water injection on my car now, so I can run pump gas and all of it and not worry about detonation.

patsmx5

iTrader: (16)

Final longshot, what plugs are you running? 6's or 7's?

codrus

The plugs are 7s, same as I had before. I don't know how much the head was shaved in any of the three cases. I suspect the first couple were probably pretty small (there wasn't anything really wrong with the deck surface at that point), but it's possible the most recent one was fairly large. Comparing the before & after photos, I notice that there are nicks in the intake side of the quench area in the "before" that I hadn't seen before. I don't have zoomed-in shots of the post-machine shop head, alas, but in the shots I do have those nicks are almost gone, suggesting it may have been a significant cut.

Hm, I wonder if the machine shop kept records of how much they had to cut off.

--Ian

patsmx5

iTrader: (16)

I would at least ask, usually they do tell you/check to make sure the head can be shaved that much as all cars have a spec for how much it can be shaved.

When you shave a head a lot, it throws off your cam timing too, basically head is closer to block, so all your cams are advanced vs "stock" timing when the head was at factory height. Advancing the intake cam is going to give you more compression too. You might try dialing back several degrees on the intake cam and see if the detonation disappears. You could do math and figure out how much if you take a guess at how much the head has been shaved total.

codrus

Yeah, but the head work was done back in November, so unless they wrote it down I doubt they'd remember. I'll call and ask.

The cam gear has a diameter of 105mm, thus a circumference of 329mm. That suggests to me that shortening the distance between the exhaust cam and the crank by 1mm should have the effect of rotating the cam gears by about 1.1 cam degrees anti-clockwise relative to the crank. It's the exhaust side that matters because the intake side has the tensioner, right? Isn't that retarding the cams, rather than advancing them?

Anyway, if it's 1.1 degrees per mm, then I suspect the effect of cam timing on ping is in the noise relative to the compression ratio bump, because (if I'm doing the math right) a 1mm head shave turns a "9.0" Supertech piston into 9.83.

--Ian

NiklasFalk

iTrader: (1)

I decked my block 0.7mm and with the head shaved about as little as possible (0.1mm maybe, don't recall) gave 13 degrees retardation (about 2.5 degrees short of a full tooth).
720/46 gives 15.65 degrees per tooth.

But my head can't figure out the logic why about 1mm of shorter distance will retard about one tooth (which is about 7mm long), besides being 2.2 crank degrees per mm (it's before lunch...).

sixshooter

iTrader: (12)

Decking the head can make a substantial difference in combustion chamber volume and, therefore, compression ratio. I used to know the rule of thumb for how much each .010 effected a Pontiac 400 engine but you would need to measure the cc's of the before and after on a Miata head.

At this point it might make sense to swap heads with someone who is n/a that wants a higher compression head and you use their stock one. Or get one of those thicker head gaskets, or both since you are on Cal gas.

codrus

I did a bunch of googling last night -- I can't find all the specs, but I came up with some things that seem plausible.

Stock: 83mm bore and 85 stroke gives is a swept volume of 460cc/cyl. If you assume the stock 99 piston dome is ~ 6 cc (I can't find specs, but seems plausible), the piston is recessed .060in/1.5mm below the deck+gasket at TDC (Supertech's spec is .060 inches, and they imply it's the same as stock), and the combustion chamber volume is 52cc (949 says that their CNC head is 52cc and that it's the same as stock), then that works out to 9.5:1 for stock.

Modify this to 84mm "9.0" Supertech pistons with a 4cc dish and I get 8.3:1. Oops. Arbitrarily modifying this to take off 1mm in height makes it 9.0:1. I dunno if that's correct, but it suffices for comparison.

Because of the quench areas, the combustion chamber isn't actually circular at the top. I don't have the actual area of that cross-section, but I'll pretend it's a circle because it's pretty close. If so, then shaving it 0.5mm gives 9.4:1, 1.0mm gives 9.8:1, and 2.0mm gives 10.9:1. 2mm seems to be about the most anyone online has reported doing. Spreadsheet here if anyone wants to check my math:

http://www.codrus.com/miata/fm2r/cr-calc.xlsx

I called the machine shop this morning, and unfortunately they didn't keep track of how much was shaved and don't remember. I guess I can get an approximation using a compression tester (need to go buy one -- I only have a leakdown tester), and other that there's not much I can do to check it without pulling the head back off. Ugh.

In theory I could swap heads with the Seven...

--Ian

EO2K

iTrader: (37)

Can you sneak a caliper in there somewhere and measure the height?

I've been following along and I'm more than a little nervous. I need to look around and see if I can find some volume info for the stock chambers, or find a head to measure. Then measure my head to compare.

Thanks for documenting everything, this is the nuts n bolts stuff I feel we are lacking around here.

codrus

I suppose I could use a depth caliper down the spark plug hole, but I don't think it would be particularly precise, plus I don't have the dish depth dimensions on the pistons, so...

I do have a spare stock 99 head sitting around (belongs to a friend who hasn't come to pick it up yet), but it's probably warped so I'm not sure the measurements would be valid.

949 says that both their CNC head and the stock one are 52cc:

SuperMiata CNC Miata Head

--Ian

williams805

iTrader: (2)

I cc'd my '99 at a little over 50 after the machine shop took off .004". I did not cc before surfacing. Also I don't believe the engine had ever been apart before but no way to know for sure.

patsmx5

iTrader: (16)

The belt goes from main pulley, to exhaust pulley, to intake pulley, then tensioner. So shaving head affects both pulleys timing.

Moving the pulleys closer will shorten the distance between the exhaust pulley and main pulley, resulting in rotating both cam pulleys counter-clockwise to take up the slack. This would retard them, opps! (note-don't do math at 2:10 am...)

Comp tester won't tell you much to be honest.

You likely need to pull the head, or put water injection on, or throw more fuel at it, or back off the timing a bit. If it were my car, I'd throw fuel now to stop the ping, then decide whether I'm pulling the head or adding WI or just drive it with a bit more fuel to stop the detonation.

codrus

I think for the short term I'm going to throw octane at it, because it doesn't ping on straight 100. Certainly that's what I'm going to do (as well as turning the boost down to 180) at the track on Friday.

Longer term... yeah, sounds like I need to pull the head again at some point. Argh!

--Ian

EO2K

iTrader: (37)

Oh yeah, I know what the chamber volume is supposed to be. I believe I actually asked that question in the initial CNC head intro thread. But, you know, paranoia will make me check it anyway.

codrus

Makes sense -- I wish I'd checked mine.

A quick wash before the track day:



--Ian

codrus

Doing the final track prep this evening. I killed an FM 3" cat last year at Thunderhill, so now I run a test pipe on the track. Since the car goes there on a trailer, there's not even any legal issues with that.



Checking the coolant mixture. 1.04 suggests around 20%, which is good. Either that or I've made a nice green stout at 5% ABV.



Swapped to the track rotors and pads. DTC-60s up front, Wilwood "E" compound in the back. I need better rear pads, but Hawk doesn't seem to make them for the PowerLite yet? Went out to bed the pads. Wow, I'd read about the sparks from these pads, but that's pretty spectacular.

Hm, that's odd, I didn't know 949 made 6ULs in a dark gray matte finish.



Oh, right, they're nickel and that's 20 minutes' drive worth of brake dust. I dunno why I bothered washing these.



Enough for tonight. Still need to load it on the trailer and pack all my crap, but that will have to wait til tomorrow morning.

--Ian

codrus

Not the most convenient of commute vehicles, but work is 15 miles closer to Thunderhill than home is, so I wasn't going to go home first.



--Ian