Half-Life | '91 DS 318iS Slicktop | Track & Weekend Warrior

[Digitalwave]

Update time!

I've been working on installing my Megasquirt for the last few weeks, but work has been busy so it was going slowly. I did pick up a new set of wheels! I found a good deal on a set of 17x8" square, forged M-System II's. WITH TURBINE COVERS! The wheels are straight, but they need to be refinished. I actually don't have any pics of them yet. They are heavier than my Style 4's, and I am not sure how tire selection is for 17's for an E30. These may end up just being "street wheels." I'd still like to find a set of ultra-light wheels for the track to replace the Style 4's.

I have the Dederic Motorsport plug-n-play MS2 kit, which is really nice. It has adapters for everything, so no cutting re-wiring has to be done, and it's easy to convert back to Motronic if needed. It uses the stock O2 sensor plug for the wideband (Spartan 2), and has a wiring adapter + plate for the E36 variable TPS, and a wiring adapter to use the stock AFM plug for the MAP sensor and IAT sensor.

The Microsquirt-based unit is built into an E36 ABS housing, with the same 55 pin connector. The ABS unit can mount right up to the stock ECU location with a little creative rearranging of the stock speed nuts (it uses 3 instead of 4, and the spacing is different).




One of the GM 3-bar MAP sensors I got in the mail was cracked open. DIYAutotune replaced it no problem:





I had an AFM delete pipe welded up and powder coated. I also sourced an early model intake manifold vacuum cap that has the extra vacuum port on it (I think for L-Jetronic?). I routed the silicone vacuum line below the manifold, and mounted it along the stock wiring harness bracket under the manifold. It's accessible, but pretty well hidden.











The car fired right up with the Megasquirt base tune. After some basic calibration, the car will idle, but it's quite rich. Going to get it tuned pretty soon. Other than the AFM delete tube, you would never know that the car isn't running Motronic any longer.







Running two widebands, one for the gauge, one for the ECU. I like the redundancy.





I did a baseline dyno a few weeks ago. With "80 smoothing" the car had a peak of 173whp and 156wtq. I was happy with those numbers. I'm looking forward to having a proper dyno tune with the Megasquirt setup. These were 4th gear pulls.

Keep in mind the AFR on the chart is from a sniffer at the tip of the exhaust, which may not have been reading very well. My wideband wasn't showing such erratic numbers, but the car is leaner up top than I want it to be. Sorry for the low quality. I can try and scan the sheet if anyone is interested.





Other misc stuff. Condor drilled door handles for extra scene points.





Did a track day in October. Nothing to report, other than I did my best time at PBIR, with a passenger in the car actually!





VIDEOS!

I don't think we can embed anything anymore, so here are links to some videos.

Dyno pull 1 (behind car): https://flic.kr/p/E4tQww

Dyno pull 2 (in front of car): https://flic.kr/p/ZQXRYi

Dyno pull 3 (next to car): https://flic.kr/p/229CnyS

First startup on Megasquirt: https://flic.kr/p/229CkpS


You can really hear in the videos how the cam comes on around 4,700 RPM. In the startup video, the car didn't idle at all because I had it configured for a 4 BAR MAP, and had some other settings wrong. The car will idle now, but at about 12 AFR. The cam and bigger injectors are too much for the "stock startup tune" I am certain. I don't know my way around MS, so I am not messing with it anymore. Hoping to get it tuned in a week or two.

[zwill23]

Your videos are private!

Nice work on the megasquirt install, the early model intake manifold cap is very clever. Much cleaner looking than the barb that is typically added.

[Digitalwave]

Thanks!

The videos should be working now.

[potatomash]

Looks good man, one thing I'd say is heatsoak. Warm starts were an issue for ages with mine, kept leaning out big time, tried heaps of air temp sensor placement and still did it. In the end I used EGO correction to get the idle pinned under any temp. Just something you'll come across at some point I'm sure..

[Digitalwave]

Interesting. Do you think heat soak of the IAT was the issue on your car? Or it just needed more idle & startup tuning in general. I won't be tuning the car, I'm having a local Megasquirt guru do it (Linfert Performance).

[paynemw]

I had an AFM delete pipe welded up and powder coated. I also sourced an early model intake manifold vacuum cap that has the extra vacuum port on it (I think for L-Jetronic?). I routed the silicone vacuum line below the manifold, and mounted it along the stock wiring harness bracket under the manifold. It's accessible, but pretty well hidden.

So, did you do any engine harness repair? I'm guessing you're running a mixture of M1.3 harness with your MS2 and other goodies. I'm having issues with my harness and my buddy is as well, and we were looking at building new harnesses, I was curious to get your input given the cleanliness of your M20b25 and the fact that you have also changed out a handful of sensors, and alot of stuff looks mixmatched, like earlier style ICV, newer style boot with minimal vacuum lines from the booster, L-Jetronic vacuum plate, etc.

[Digitalwave]

So, did you do any engine harness repair? I'm guessing you're running a mixture of M1.3 harness with your MS2 and other goodies. I'm having issues with my harness and my buddy is as well, and we were looking at building new harnesses, I was curious to get your input given the cleanliness of your M20b25 and the fact that you have also changed out a handful of sensors, and alot of stuff looks mixmatched, like earlier style ICV, newer style boot with minimal vacuum lines from the booster, L-Jetronic vacuum plate, etc.

My engine harness is actually a 100% stock Motronic 1.3 harness. Not a single spliced wire. The sensors that I added are all independently wired.

The T and L style ICV's are interchangable and use the same connector. You just have to match the intake boot to the ICV. I prefer the looks of the L shape. The silicone intake boot doesn't have the nipple for the brake booster, but it isn't necessary. The two lines coming off the side of the throttle body still have the stock check valve, and they go directly into the brake booster now (without the extra L off to the intake boot).

The L-Jet vacuum plate was added simply to get a clean looking additional vacuum source for the MAP sensor.

It's nice to have a clean, untouched wiring harness. My last E30 had some shoddy work done to the engine harness for a Miller MAF conversion. It also had some issues within the harness for the CPS wiring, that were causing a break-up at high RPM. That was really hard to diagnose and fix. That is all part of the reason why I went with a fully plug and play MS2 setup, where I wouldn't have to modify the wiring harness at all.

M20B25 wiring harnesses usually sell for less than $100. Sometimes it's easier to just buy another one, rather than trying to fix one with issues.

[paynemw]

M20B25 wiring harnesses usually sell for less than $100. Sometimes it's easier to just buy another one, rather than trying to fix one with issues.

Thanks for the info. I have yet been able to find a decent Motronic 1.1 harness for anything less than $200. I'm actually picking up one tomorrow, but I will most likely have to re-create it, which is my intent.

Regardless, clean clean build, and thanks for the tip about changing boots to allow for the T style ICV, I didn't even think of that. I was interested in removing the unnecessary vacuum lines and converting to the M1.3 style intake boot.

[mitchlikesbikes]

God damn, the attention to detail in here is insane. Thank you for some inspiration.

[Digitalwave]

Thanks for the kind words all! Glad to see this thread help others or even just inspire.

Just a short update today. After the initial hardware installation (MAP, wideband, VTPS, etc.), the Megasquirt conversion slowed down a bit. The initial configuration was pretty easy, but I started getting confused after finding conflicting information about rising edge versus falling edge input capture. I figured that out, and posted the solution in this thread. Shortly after, I had a tuner start working on my car. We made some progress with the idle, and cruising at low load/low rpm as well. Shortly thereafter, we realized there was a crank signal issue (VR signal) at 4,800 RPM. We did some things to diagnose it, but didn't have any luck fixing it. Long story short, a brand new OEM CPS fixed that issue, as documented in this thread. Cliffs: make sure your MS has a proper resistor in the VR+ input, check your CPS air gap (1.0mm +/- 0.3), and when in doubt... new CPS?

It was odd that this CPS worked fine with Motronic, but it seems that Motronic is much more lenient on a poor VR signal that Megasquirt is. Glad that got sorted out.

Today, @varg from the forums took over tuning duties, and it went really well. We've got the idle down, cruise is good, and we did a lot of WOT pulls. Things are looking and feeling really good. We are going to work on the tune some more later this week. My injector duty cycle is in the 85-90% range, so these 20# injectors are pretty much tapped (leaving a margin of safety). Don't have many pictures to post since my last update.

The two pins on the black connector as part of the engine harness (previously A/C related) are wired for launch control and table switching. Pretty neat. I've already got a game plan for implementing the launch control nicely.





My old CPS. The new one didn't have that dent you can see on the top. I have no idea if that was related to the issue or not. Resistance was exactly the same between the old unit and the brand new one.






While changing a jumper on my daughter board, went ahead and soldered in CANBUS outputs for later AiM race dash usage... :devil:




While trying to figure out the VR signal issue, we used the proto area to add jumpers to multiple resistors for easy testing. Since the CPS itself ended up being the issue, I don't need this any longer, but I will leave it until if/when I need the proto area for something else.






The new rimzz:




Shoutout to Derek @ DedericMS and @varg for the help figuring out the VR signal issue. Derek has been very helpful in supporting his PNP Megasquirt product.

[rzerob]

Style 21's, nice.

[Digitalwave]

Getting really frustrated with the car at this juncture. We started tuning on Monday, and it went really well. Car was running great. Yesterday, we picked up where we left off, but from the get go something was wrong. On startup, the car had a lean misfire, and wouldn't even driven until we enriched the VE tables considerably. The car was not happy, with AFRs jumping all over, and overall much more rough. The car hadn't been touched since we parked it on Monday. The temps were a good 20* colder, but it shouldn't have had that much effect.

We came back and started looking for vacuum and exhaust leaks (didn't find anything). We pulled the plugs, and they didn't look too good. #5 and especially #2 looked fouled. I don't have a bore scope, but from peeking into the cylinder you could tell it was wet in there. There was some oil on the threads, but I think my valve cover gasket is seeping some.

We had also noticed that my valves sounded louder than normal (which could be injector tick), and if you remove the valve cover cap and sniff in there, the oil definitely smells like gas.

After putting in brand new spark plugs, the car ran awesome again. AFR's were stable, and the car was MUCH smoother (no more misfiring). Drove down the road and less than a mile later, the new plugs fouled again.

That being said, I think I have a bad injector or two (probably cyl. #2 is the worst). Maybe leaking, maybe stuck open, I don't know. Since my IDC's were hitting 90, I am going to take the opportunity to move up to 24# injectors instead of rebuilding these. I will probably replace the FPR as well, since it's of unknown age/origin. I am going to have the injectors tested locally, which will hopefully confirm an issue with at least one of the injectors.

Haven't ruled out the possibility that there is oil in the combustion chamber, rather than fuel, as the "wet" substance you can see through the spark plug holes. However, my car has never consumed oil or blown smoke from the exhaust, so it seems less than likely that all of the sudden I have a piston ring issue.

I will definitely need to do an oil change as well before running the car again.

Gotta say, after we dialed in some WOT pulls on Monday, the car was feeling phenomenal. Pulling really hard to redline. Noticeable power difference with the AFM delete and better fueling than my Motronic tune had.

Lastly, I leave you with a 2-step launch control video (volume up!). I can't get it to upload in the right orientation for some reason: https://flic.kr/p/Dynzpc

[Sh3rpak!ng]

I'm sure you'll figure it out. Goodluck getting it sorted quickly. Launch control is so damn cool. I'm a little jelly :p

[Digitalwave]

Worked on the car a little bit this weekend. I pulled out the injectors to get them tested this week. I just want to know if they had any issues or not. I already ordered new 24lb injectors as I need the head room anyways (especially when I move to ITB's). My fuel rail looked kind of bent, so I ordered a new one, and a new FPR as well. Now my entire fueling system will be new.

Here's an example of my cylinder #2 and #5 fouled plugs:







Fuel stuff out:







I started to wire up my switch setup for launch control. It is going to have a "master" arming switch on my switch panel. When that switch is turned on, it will active a switch on the clutch. I am using another brake light switch as it operates in the way you'd want for LC. The LC is a "going low" wire that normally has a 5v single. LC is activated when you ground out that single. The stock clutch switch is normally closed (when the clutch is out) and opens when you depress the clutch. That means it would be grounded and LC would be activated when the clutch was released, and it would be disabled when the clutch was depressed. That's the opposite of what we want.

The brake light switch is normally open (when the clutch [or brake] is out, and it closes when you depress the clutch or brake. So first, you have to hit the master switch, which will bring the signal to the input of the brake light switch. Then, when the clutch is put down, and the other settings in Megasquirt are met, then launch control will be active.

I used pin 40 on the ecu, which runs to the C136 connector, a small black 4-pin auxiliary connector by the ECU connector.







Now my mystery ??? switch has a purpose!





I wish I had some better photos, but the next project I've been working on is a Nexus 7 tablet as a digital dashboard. This is actually surprisingly easy with Megasquirt, and MUCH cheaper than something like an AiM race dash that I was originally planning for. Shadow Dash MS is an Android app that uses TunerStudio dashboards, and can display just about anything that the ECU or GPS can see. I also bought an external 10Hz GPS for lap timing and for other data (like MPH for the cluster). This is where I'm at for the dash so far. Exhaust Gas Oxygen should just be "AFR", but I am dealing with a bug in the software with the developer.





Lastly, while messing around with TunerStudio and the tablet dash, I noticed that my MAP sensor was reading all over the place with the engine off. It should read steady at atmospheric pressure with the engine off, basically 100 kPa at sea level. I noticed my sensor was wandering all over, but at the most, reading around 82 kPa (and jumping as low as 50 kPa). Soooo, I think my MAP sensor was faulty, and possibly a big part of the problem we ran into with tuning recently. I had a spare sensor here, plugged it in, and it was a rock solid signal at 102 kPa. We will see in a week or so once I get the new injectors, spark plugs, and valve cover gasket installed and can drive the car again.

All done after a long night of work:

[

[dasmanschaft012]

What wheels/tires are those? Alpina reps? Interested in width and fitment.