V-12 Engine into 308 Build, over on Grassroots

You are way more persistent than I would have been. It sounds great though!!

 

The plan was to have a second, but I wasn't allowed to let the kids anywhere near it after the fire incident so just me again. Yes, I can start/stop the ECU from the PC and that is the preferred method as the PC stays connected that way so no data is lost.

 

Ha, my wife would just say "that'll teach'em" she's more of a school of hard knocks approach. Can't say I'm not against that.

Oh man lost data... That's one reason I ditched electromotive yrs ago, way to easy to lose our have the data corrupted or lost. I don't even trust laptop batteries, plug the sucker in!

 

I'll say it, might be too early, but I'm saying it anyway. 'The greatest Ferrari V12 Ferrari never made, but should have made..... if they were able'.

 

I tried resetting the TBs with feeler gauges this morning as a base sync while also playing with the ECU software to try to read them all....the feeler gauge method is pretty bad it turns out if my ECU programing is to be believed and this is after a couple quick roughing in adjustments

Image Unavailable, Please Login

 

Is that reading percentage open, manometer or flow rate? Real time digital reading on flow rate is very cool and useful.

 

The gauges are to help sync the TBs, they read % error from TB12 which is where the actuator is attached so it is always 0. Here it is after I played with it a bit, not prefect but much better.
Image Unavailable, Please Login
the graph below are the cylinder MAP reading that run the gauges....this looks to be working pretty well now so next run I will turn on the cylinder corrections and TBs sync'd or not each cylinder should get the same mixture.....at least that is how I THINK the code will work. With the TBs kind of matched now I should be able to get on with fixing the mixture.

Before this run I cleaned the plugs to get all the cylinders firing again and while they were out did a compression test..165-185 so not great but nothing is seated yet so it doesn't scare me yet.

I have a misfire function I'm playing with too....and it looks like its sort of working. It shows clean in the middle where I know for sure it was running well and lots of color at the end where for sure the plugs were again all loaded up and things were shutting down so helpful if not perfect I think. With the TBs kind of matched now and the misfire tool to tell me if I can trust the O2 readings I should be able to get on with fixing the mixture.

bottom graph is the 4 NBO2 sensor just kind of saying its pig rich except when there are misfires

Image Unavailable, Please Login

 

Very cool...
You could always grab an old school carb flow meter to check the TB's.

Sparkplugs, what's your gap? And then next what's the coil output? When I developed the 4L I had to find a high joule output coil, ended up with COP for that reason. Between the cam profile and CR the ignition kernal would extinguish unless the spark energy was high and the gap reduced. Had a real fight with the 3.5L @ 13:1 CR on ethanol with those cams.

 

Just the thought of trying to sync 12 TBs with 1 flow meter is why I have a screen that shows me all 12 at once

The coils are GM D585 which I think are the hottest of the LS coils and quite popular with the turbo guys. The plugs are .030" at the moment. The issue right now is soot. On the first fire when everything was clean I had a bit more time, now that its been running pig rich for 10-15 minutes everything is coated so builds up on the plugs quick but I'm going to pull about 15% fuel out for the next go and switch on the cyl_MAP correction and I think it will start to clear up so I can get on to the driving/loading part of tuning and start making some heat in the chambers to clean them out...I hope

 

LS coils should be good enough. Used them ages ago on a supercharged 328 running 20psi. Gap sounds good, any tighter and idle will get "interesting"

I know the thought of flow meter is crazy but I just can't think of anyway that's more reliable and accurate for knowing what the CFM mass is. Yeah, you can trim per cylinder but don't you want to insure you're getting the same air mass across the board so that each cylinder is producing the same pwr? I know.. PITA. It'd bug me till I did it...

 

I guess the answer is what the goal is at the given moment.

At WOT the goal is HP and all the ports are flow matched with the intake and TB in place within 1/2% total error. The TBs are 54mm and as long as the plate edges are not wider than about the shaft thickness, the flow doesn't really change, for sure there is a deadband in TB position v flow up there so small misalignments between TBs doesn't matter (so perfect on a race engine). The headers are length matched to make the pulses and low end as alike as it can be but I'm sure the various bends I had to use are inducing some flow differences and this is were cylinder trim tables will be used if needed to get the cylinder mixtures all the same to get the HP the highest it can be knowing the flows are already as high as they can be.

At idle or cruise the goal is that it runs well and doesn't foul plugs. Here matching flows is nice as it makes the engine the smoothest, but it really hard with massive TBs that are closed to where a .002 feel gauged won't go past the throttle plate....I would argue flow matching it flat impossible hence the horrible reputation ITBs have on street engines. Idle is under 40kPa so 60kPa vacuum which is the force driving the flow so big %flow changes with tiny TB angle changes. So here my goal is to get the flow similar, but match mixtures and rather than using trim tables that again depend on TB sync at a moment in time I'm going to let the ECU make the correction dynamically using the measure error at the current moment. If I can keep the TBs with +/-10% or 15% MAP reading and the mixtures all matched within +/-5% I think it should run very smooth and I THINK I can get the ECU to do this. I'm questioning if I want to write in something to taper the allowed correction to 0% at WOT but I'm not sure there is any need as WOT is basically 0kPa vacuum so the correction would be 0% anyway so not sure there is a reasong to bother? I can't think of any reason anyway... ..not sure, but I do want to get it idling and driving nice.

 

Cruise is hands down the hardest to dial in due to the gear ratios in the 308's. I hear you on the large TB balance at crack, the 4L is sporting 50 or 52mm throttles, just a hair makes a massive difference. The lowest I can get that one to idle is 1350rpm smoothest is 1500rpm. Street machine it is not.
3rd gear and 5th have big crossovers at low speed cruise and highway part throttle. Really wish I could put a gear sensor in and write specific gear maps.

 

Sounds like you had all the issue I expect to have.

Allow me to introduce you to https://enginelab.net/ where you can add any sensor or control logic you can dream up, or I can send you a copy of my model that probably already has them

I'm pretty sure I can tame this thing for the street.

 

It's the engineering and putting in the actual sensor for the gear selected bit that's expensive and difficult.

I do like the enginelab software though, wouldn't be ideal for customers though. Link does have gear map capability, just needs the input. That's the catch.

So is the wife going to be the driver or can you get one of the kids to be driver? You get the drinker seat with PC for road tuning

 

I was double checking as you were typing, I already added gear select as part of my traction control system.
Image Unavailable, Please Login

Image Unavailable, Please Login
I don't have a sensor, I calculate gear by comparing the rpm to the speed sensors when the clutch is up....which reminds me I never quite finished installing the clutch switch last winter.

For a customer, it just acts like a normal ECU, they plug in the PC and adjust whatever is needed but you can go in and setup what ever you want them to have. That little snip above is from my main build script where I tell it to add the traction control files and I also use is to document things a little bit. Add a # sign and its ignored so 1 car gets traction control and electronic throttle, another does not. Its that simple, there is no reason for anyone to have anything they don't use slowing down their ECU or cluttering up their PC screens. The AEM infinity ecus are enginelab with access to the model restricted in the inferface....when I plug into 1 with enginelab software I see and can edit everything behind the curtain but they have stuff jumbled I assume intentionally to make it harder plus its theirs, I used nothing AEM in my model) but my ECU is branded AEM as I bought it while they still had an exclusive deal. I could just never go back at this point.

For road tuning its just me and the logs. I drive a bit then pull over where I can, edit, new drive...it takes a while.

 

Totally understand and for our personal projects, I agree. With clients I have to look at that I'm not going to be around forever and in some cases these guys are in different countries and go racing so they want their local tuner to be able to work with it. GUI wins Everytime it's simple, even if it's restrictive and not fully programmable.
For some reason most people just want the easy button.

Customer interfacing... Fun times

 

This is my main tuning screen...pretty simple but you can literally put anyhjitng on it in any position to like, graphs, gauges, table just right click and place it where you want. I've been using the white background as mostly I've been inside

Image Unavailable, Please Login

or right click, black background....you can change ANYTHING to look the way you want it to look and display what you want it to display. Its pretty nice.

Image Unavailable, Please Login

 

That is heckn awesome. Fabulous Mark.

 

Ok so.....

Fuel pressure issue tracked down to a solid state relay, replaced with mechanical relay and problem solved

Wednesday I tried the cylinder MAP correction....it was outputting gibberish

Thursdays tired again, no...its error doubling not error correcting

Friday, its running pretty good with the ecu auto matching mixtures. I spent a little time on tuning so after I pull it and fix whatever is causing the rear bank to be running on oil rather the gasoline it should fire right back up so for those worried that this project was wrapping up....nope!

 

Oh ****! Are we allowed to say **** on here? Well, look on the bright side, at least it's the rear bank. Hope it's something easy. Are you sure it's oil and not your fancy coolant?

 

Well at least that answers one question

 

That's a crazy amount of something being burned... I don't recall but are you running Evans? If so it looks just like that. And that points to a head gasket issue. Or it's sucking coolant in from somewhere...

Hard to tell but oil is a bit more blue/grey in color.

 

I am running evans so maybe....it just needs stop leak then

I kept going back and forth between coolant and oil....the evans makes it very non-obvious

 

Evans will combust, so it's very likely that's Evans. I've seen it myself a couple times, failed gaskets when we were doing development on them. That same foggy cloud. Slightly sweet smell but not exactly like coolant. Leaves a residue.

Fun times.... Fun times...

 

GM tablets are the normal go to and should be compatible with Evans.

I found a better fix but that requires the head gasket coming out again.... So.... Yeah.. Try the tablets.

Fix is this, stupid common on boosted engines... Coat both sides of the head gasket in aviation sealant. No more leaks. MLS works best, haven't tried it on composite but should work.