Lobo240sx wrote: ↑Thu Jul 09, 2020 12:37 pm
Nice KA Build.
Thanks!
Little later than I intended to get to this. I was having a difficult-to-recreate noise from the transmission so I pulled it yesterday and found some stuff I didn't like.
Here's a youtube vid of some dude in a miata with the same problem I've NEVER had a pilot bushing go out before. The noise didn't happen all of the time, just occasionally. Here's the thing, that pilot bushing was installed BRAND NEW with my competition clutch kit. I also found a lot of play in their throwout bearing as well. I'll try to add a caution to my original post about it but I would NOT use the throwout bearing or pilot bushing that Competition Clutches include with their kit. I've replaced the throwout bearing with a timken unit, and the pilot bushing with a brass one from Advance. Here are some pics of what I got out;
^New on left, old on right. Kind of hard to see but it IS out of round a tiny bit.
I tried to capture it but, look at those grooves on the inside of the old pilot bushing. That ISNT from removal. The pilot bushing that came out only had MAYBE 10-12k miles on it. I've only put 8k miles on the car since I rebuilt the motor. Clutch was installed a couple thousand miles before that.
I found what looked like some of the pilot bushing material on the input shaft to the transmission. Could barely see it but I could feel the roughness with my finger. I cleaned it off with some 600 grit emery cloth, blasted everything with some brake parts cleaner, then lubed up the input shaft, collar the throwout bushing housing/sleeve rides on, and the clutch fork pivot-ball with some good axle grease. I also replaced the output shaft seal while I was in there. No clue what caused this. I checked the input shaft side-to-side and deflection on the transmission and it's tight. Maybe just got a s*** pilot bushing with my clutch kit?
Took advantage of the opportunity to replace my Megan motor and transmission mounts with P2M style. You can see the transmission mount in the pic of the transmission earlier. I dropped the engine about an inch or so. It's not too low. The bottom of the oil pan sits about flush with the crossmember now. more clearance between the strut tower brace and coil packs. I get a bit more NVH now in the cabin. I don't feel it's excessive and it's not my daily. Took it for a test drive and the random noise is gone. Problem solved.
---------------------------------
Nerd alert! Megasquirt rant ahead.
---------------------------------
On to Megasquirt. In revision 3.3.0, they changed the air-density compensation scheme from using a combination of MAT/CLT and MAP density, to a consolidated table where MAT density does all of the lifting now. Good idea and makes it a lot easier to tune a single table. Problem is, they leaned on Boyle's law a little too hard for their graph. air density isn't entirely linear with temperature, as they tried to make it.
Pv=constant graph for illustration of what I'm talking about.
Link to good explanation if you want to know more.
The shortcoming in their application here, is that this all assumes a FIXED volume of air, and all of the other factors that affect air density are being taken out of the equation as well. So you wind up with the plot they have here;
What does this do? It's just like any other fuel modifier table in Megasquirt. Under the conditions you specify, it's going to give/take fuel away, in this case, based on air density which it is deriving based on your IAT sensor readings. IAT placement really affects this as well. I've got my IAT installed about 6" after the outlet of my intercooler in the cold-pipe to avoid heatsoak in the engine bay.
So we finally get to the issue I was having along with other MS users running the 3.3.x firmware or above. The issue I've been chasing my tail on, was consistancy in my AFRs. I'd get my VE table dialed in, then go out a different day when it was hotter/colder and I was running too rich when it was cold, and too lean when it was hot. Digging through the manual, MS forums, and the Miata MS forums, I figured what was going on.
So what SHOULD the values look like? Well there isn't a blanket answer. Just like there isn't a blanket VE table or spark table that is going to work for every car. They've done a good job here simplifying it for the end user but elevation and IAT sensor placement are going to play a part. What I would absolutely do is, when tuning your VE table in, set the values to 100% for a 10-15 degree range of temperature starting where your IATs read when you fire your engine up before you go tune. that way, this table won't be modifying your fuel and you can do a pure tune for VE for those conditions. Go get your VE table all tuned in. THEN, as you drive the car in different weather conditions. pay attention to what your EGOC % is. If it's having to add a ton of fuel when your IATS are well below the temperature you tuned at, you need to increase your MAT correction at that temperature range. Pay attention to what it's doing across your table, not just one area of cells.
I ended up with something like this. Apologies, I don't have my actual graph to post right now but the hotter temperatures aren't far off from this, save mine is nearly flat from 85-95F at 100% since this is where I tuned in my VE table.
So far so good, I drove a little last night and a storm was moving in. My IATs ranged from about 97F when I started to around 81F when I got to some cool spots outside of town. My EGO correction never got above 105% or below 98% peaks, riding mostly between 100-102.%. I'll call that a win.
I know that graph goes way beyond what a majority of us will ever see. I'll pull my actual graph this weekend if anyone actually wants to see it. A completely separate discussion, but, Mine doesn't go but to, I think, 140F. At 120F IATs I start pulling pretty heavy timing. I'm on pump gas (93 octane) and I REALLY don't think it's that safe to tolerate IATs that high. Lots of factors at play like, again, IAT sensor location/heatsoak, compressor efficiency, etc... I feel like, with my IAT sensor I'm getting a pretty accurate picture of what is going into the engine. At lower velocities, I'm sure I'm getting a little saturation from intake manifold heat, but I've got a Sikky phenolic spacer I'm planning to install to cut down on that. At WOT and higher loads this is not much of a factor, if any.