Rolf's SBF -> 318iS e36 Swap

[Laminar]

Yeah, the M3 pump is a bit different than yours, but I found a place that can make hydraulic lines, I'll just have to figure out fittings and routing. It's helpful to see how you did it, I'll probably end up with a similar bracket attached to that head.

[MvnTgt]

No major update but I finally bought and installed a set of Suplex 6cyl front springs on the original Boge front struts so I now have a proper amount of front
ground clearance and bump travel and don't cringe going over parking lot speed bumps and the occasional (ahem) pothole.

I've still been fighting with the MS2extra tuning; mainly the idle mixture having a mind of it's own and trying various adjustments to stabilize it. I'd love to know
the voodoo Ford powertrain calibrators did to maintain a rock steady idle on these engines. I feel I'm fairly technically savvy but this part makes me feel like a complete dolt.

[Laminar]

I'm running open loop idle on my MS3X and it's rock solid. I'm running sequential injection which probably helps. Which IAC valve do you have? If you're interested, I can send you my tune so you can look at idle settings along with VE and timing in idle zones.

[MvnTgt]

I'm running open loop idle on my MS3X and it's rock solid. I'm running sequential injection which probably helps. Which IAC valve do you have? If you're interested, I can send you my tune so you can look at idle settings along with VE and timing in idle zones.

Hm, I"m running batch fire alternating w/2 squirts. I'm running the stock Ford 2 wire PWM idle valve with open loop warm-up and I don't think it's the problem. I adjusted the curb idle with the throttle stop screw and the
idle duty cycle is set to ~25% and I'm sure the idle valve is out of the picture once fully warmed up. I have been AFR changes (I've been shooting for 13.5) which I believe happen due to the big honking intake manifold
heating up and changing the air density but the only solutions I've seen are mess with the MAT/CLT correction and mess with the ideal gas law curve. I've also got the idle timing set for 16° BTDC and it's not changing at
idle, just the AFR.

I've just been adding some percentages to the MAT/CLT correction curve at low flow and having it slowly drop off at higher flow. It's been a guessing game and I haven't seen a good write-up on what to look for to dial it in.

Anyhow, I do have a MS3X that I assembled 3 years ago and haven't put into use yet. Send me your MSQ and I'll have a gander.

[Laminar]

PM'd!

I'd be interested in seeing an MSQ and datalog of your idle.

[MvnTgt]

Received your PM and responding with my tune and a short log file.

Thank you for the msq file. I see your using a MAF sensor?

[Laminar]

Yes - started out with MAP but figured I'd try MAF to see if I liked it. Since I had a MAF but no actual curve, I copied in one that was probably close, and set up a dual table configuration, then let autotune figure out the rest. The car needs waaay less acceleration enrichment on MAF and the throttle response seems snappier. I know most people say MAP is best for performance but I'm liking MAF so I'm sticking with it for now.

[MvnTgt]

I'd been toying around with trying a MAF setup but I've been stubbornly fighting with the Speed Density tune. I'm dense about some things. Guilty as charged.

I decided to go really radical with the MAT/CLT correction at idle (100% @ 45,000) with it dropping off steeply off idle (15% @ 77,200 ->10% @ 100,800 -> 5% @ 150,900 -> 0% @ 200,000) and fattening
up the idle cells. I have it bouncing around between 12.5 and 13 AFR. If and when I go sequential, I may be able to lean the idle out a smidge.

I don't worry about getting the last bit of performance out of this vehicle combo, I just want a non-irritating sort of performance street car that I can have some fun taking to an autocross. Earlier commenters
were right that the street tires that came on this car are garbage but driving around with 1/2" of bump travel before hitting the bump stop wasn't helping me ().

I need to redo the location of the radiator to get more space from the water pump and switch the electric fan so it pulls or see if I can wedge a mechanical cooling fan on the water pump pulley.

PS: I've heard that some types of GM rims will fit on the e36. Is this true?

[Laminar]

Some GM cars (Corvette) used a 5x120.7mm pattern that will fit.



- - - Updated - - -

What happens to your idle with MAT/CLT correction turned off? I see a rock steady advance, steady VE values, and steady AFR targets, but the injector PW is swinging up and down, leading the RPM swings just slightly. The only thing affecting fuel pulse width is CLT correction.

[MvnTgt]

I believe the fluctuations are triggered from the cooling fan turning on ( I have the fan set to turn on at 98°C w/5°C hyst). It causes just enough of a load to destabilize the idle. It gets worse
if the heater fan and defrost are on. I've played with adaptive idle timing to add timing if the idle speed drops below target and remove timing if it rises above it but can't say I've had a lot of
success/ confidence in it not making the rolling idle worse.

I've tried setting idle AFR without using MAT/CLT correction and can't get the engine to hold a stable AFR. The idle AFR likes to drift if I just let it idle parked and I have to change the idle cells
up and down (which you're not supposed to have to do).

I've been fighting this issue for a while and I haven't been able to nail down a solution yet to the idle woes short of idling at +1000rpm.

[Laminar]

So for a time period like this:



I see the idle swinging by 250rpm (top white) with a period of about 2.5 seconds.

The CLT is relatively steady at 93 degC (bottom green) so the cooling fan shouldn't come into play.

But there is a lot of movement in the MAT signal (bottom yellow), and every downward spike in the MAT signal correlates with a jump in the "Fuel: Air cor" signal, which looks like it's causing the swing in injector pulse width and also AFR.

Does the MAT show those same spikes with MAT/CLT correction turned off? If it does, that could be noise in your MAT signal. I'd check the grounding to that sensor really closely.

[MvnTgt]

I poked around on the MS2 installed on my car and double checked the bypass caps (C4 and C5) on the main board related to the MAT input and didn't see anything obvious but I reflowed the
pads just to be sure I didn't have a cold joint somewhere. I am using the relay board to make my sensor and injector connections and have the main ground point at the passenger rear stud of the intake manifold.

I see a bit of jitter on the MAT reading with the ignition on but without the engine running. Something is adding some noise and the ADC is seeing it. Probing the MAT and MAT return points was showing a steady 4.77V with the sensor
disconnected.

I'll read through this document and see what, if anything, I can do.

[Laminar]

the main ground point at the passenger rear stud of the intake manifold.

But the IAT is grounded at the MS sensor ground wire, right?

[MvnTgt]

Yup. The TPS, IAT and CLT grounds are connected to pin 19 on the MS as per the relay schematic. I agree though, the jitter doesn't look right.

[MvnTgt]

I got tired of of belt squeal due to the insufficient wrap on the alternator pulley so I made a plate to mount an idler
pulley and slotted it to allow the alternator to swing and take up the belt slack. As almost an after-thought, I added a turn-buckle to make belt adjustments easier. I'm going to eventually redo the radiator mounting so
I can move the electric fan to the engine side and have it work as a puller instead of a ghetto pusher fan. Another time though.

Hopefully this arrangement is better than a kick in head...

IMG_20170916_173015[1].jpg

[ponchiz318]

That configuration is similar to how I acquired my engine. The previous owner deleted the idler pulley up top. I put a new idler and tensioner on mine with a belt that's about 1/4" too short. Definitely a pain in the ass to get it on.


ForumRunner_20170918_100110.png

[Laminar]

For comparison's sake, here's the stock Explorer setup.

[MvnTgt]

Another small update:

After I made the idler pulley locating plate and put the car back together, I decided to deal with a long standing coolant
leak at the front of the engine. Coolant was pooling in the low spots on either side of the timing chain/water pump cover.

So, I drained the coolant a second time and pulled the upper and lower intake off to find this:
IMG_20170930_183807_large.jpgIMG_20170930_183819_large.jpg

Fel Pro 1250 gaskets.

Anyway, I installed gaskets for a 2001 Ford Explorer V8 (5.0) and just took it for
a couple of trips around the block and all seems good so far. Yay?

At some point, Ill see if I feel like redoing the picture links.

Photobucket, phooey.

[MvnTgt]

I've been driving the car to and from work and keeping an eye on the coolant level and it appears that the gasket change was successful.

I have made two changes to the electric fan setup (Taurus/Sable fan running in reverse as pusher) I've been running. First off, I removed the reinforcing bracket between the horns the bumper mounts to to put the fan right against the radiator (no more foam sandwiched in between the fan shroud and radiator as an air seal -- it didn't work well over time). Secondly, I switched to using the high speed winding instead of the low speed one. I had been using the low speed due to my issues keeping the engine from getting into an unstable idle when the fan turned on. This appears to have corrected itself mostly (tightening the ball/socket on the downpipe with the 02 sensor, slight intake leak fixed with gasket change? Dunno.)

The car is running the best it's ever run since I started driving it. Pinch me, I must be dreaming!

[MvnTgt]

I found something that I guess has been dogging me since I first got the car running on the MS2. While reviewing a datalog
I had made while adjusting the tune on the car, I happened to select the Batt V as one of the parameters since I was trying to get to the bottom of an apparent charging issue. From all my lurking on the msextra forums, I've never gotten a good idea of what an obviously noisy signal would look like but I was seeing some really jumpy spikes and valleys. I confirmed the voltage I should have been seeing via my DVM hooked up to a cigarette lighter adapter plug.

I did some researching on the msextra forums, found out this noise can be caused by the fuel injectors being fed by the same 12V powering the MS2. The easy solution is to solder a capacitor (I had a 1uF tantalum in my junk box) to the H1/Boot header (used by the MS1 processor; I'm using the MS2 so the header is not used this way). The capacitor bypasses any noise at the ADC pin (that the H1 header is connected to) to ground.

http://www.msextra.com/doc/general/ms1v3schems.html

The better solution would be to have the injector drivers 12V supply separated from everything else but this works well enough.

I'm going to have to double check my car's alternator. I thought it was a 130 amp 3G but it seems to be having trouble keeping up when the electric fan is running on high speed.

Sigh, if it's not one thing, it's another.

[Laminar]

Nice find. Electrical noise is a friggin ghost and a pain to fix.

[ponchiz318]

I'm at the point where I need to go stand alone. I don't trust the sct chip, so I'm looking at either a ms pnp computer or FAST EZ EFI 2.0.
I'm leaving more to the EZ due to its set it and forget it function. I have been looking at pros and cons for both systems.

[Laminar]

MegaSquirt has an autotune function that will dial in your fuel table for you. Once your tune is dialed in, you can activate EGO control with a long-term fuel trim table and it will keep the engine running at your specified AFR without you needing to do anything.

[MvnTgt]

To be precise, the registered (ie. paid) version of Tunerstudio MS has autotune; VE Analyze Live and Warmup. The paid version comes with some other perks too.

I went for the cheapest solution. It works but i don't like the db connectors that much among other things. That's what the disclaimers are for, right?