SNAFU; high performance 318is build that lives up to the name, Turbo M42 ➞ Turbo M20

[varg]

By the look of it the front will be way easier than the rear. Rear was bad enough that after I replaced one on my own I had a shop do the other one. Front looks straightforward as long as the race doesn't separate, but even if it does it seems a 3-jaw puller should be all I need.

[digger]

Ah this weekend. It started as a fixing a simple thing, a blown out turbo gasket caused by fasteners that won't stay put:



Then this happened:



And then this:



All because this stud showed up in a search for "m8x40 exhaust stud":



I didn't bother to look up 45H steel, which is a set screw specification, and according to ISO898-5 "The use of set screws of hardness class 33H and 45H under tensile stress (axial or bending) is not recommended." For a project in school, or in my future job as an engineer I'd definitely have looked into this 45H spec before ordering, since it would be my job to double check something like that, but since this was "just hobby stuff" I didn't and here I am. I spent a few hours trying to weld a nut to the stud and back it out last night and failed, I need to get some nuts that aren't contaminated beyond what a wire wheel can fix. Now I'm working on stuff for school and don't know if I'll have time to fix this before the school week starts. I'm so glad I couldn't replace the other studs when I got my new hardware in months ago because I couldn't remove them without taking the header off, they may have all broken because I was torquing them down pretty tight to compensate for a warped header flange. The stud didn't break when I torqued it down, it stretched, and when I went to remove the nut that's when it failed.

those are the ones i was going to use. 45H isnt a bad material the specs dont call out a yield strength only hardness. it never dawned on me that they are meant as a long grub screw... they are prone to hydrogen embrittlement though from memory. what did you end up using? you can get stainless but those are not strong and need to probably antiseize the crap out of them . i might just end up using stock ones lol

[varg]

I ended up salvaging and using the old stock studs. I soaked the ones with seized nuts in penetrating oil then clamped them double nutted in my vice and eventually got all of the seized nuts off and used new nuts. Any studs that come out with the nut in the future will just get replaced with OE studs, since it seems appropriately rated studs with a hex for driving them in are not available in the size I need.

[varg]

Results of today's drive and pulls with the auto-tune running to dial in AFRs a little higher in the map. That's a lot of boost...



Head gasket don't fail me now. Because if you do I have to send the head off to get that damn stud out and get the stupid header machined flat, and that's not gonna be cheap.

[TeXJ]

I like colors :D

[LowR3V'in]

hell yeah.
15 to 20 is a big jump and imo the m20 really starts to wake up at 20.

[varg]

What enabled me to turn the boost up was the wasted spark conversion I did a little while back. It made a huge difference. Not only did it eliminate the predictable miss I would experience at high boost if the plugs were anything short of immaculate, but it got rid of a random miss I was unaware of in boost which was slowing spool. With no other modification, boost onset was much faster. I am very traction limited but the car is undeniably fast. I just need to get another spool of black PETG now so I can print a distributor blanking plate and put the cap and rotor in a box somewhere to be forgotten about.

[varg]

This car has developed the weirdest intermittent issue over the past year. It'll cut out on me randomly, dead, needles drop and everything, then start up and run fine a few minutes later, or just cut out for a second. It's a hard cut out too, if I'm braking while it happens I feel the ABS cycle and everything, same as if you turned it off with the key. It'll go a month without doing it then do it twice in one drive for just a second or it'll die and not restart for a few minutes. It did it twice on decel today. I checked my splices for my main relay wiring mod and no amount of moving them around or jiggling made it die while idling. Moving the harness from the ignition switch around under the dash did nothing, neither did wiggling the key, banging on the column, etc. I cannot replicate the issue for the life of me and it makes diagnosis difficult since the two times it has died for several minutes by the time I got my multimeter out and hood opened it was alive again. I've been through this car, grounds are good, battery cables tight, junction block connections tight, main relay has been swapped.

Looking at the diagrams it has to be
-wiring upstream of the power side of main relay (unlikely, nothing is loose and there are no connectors)
-wiring downstream of the main relay
-wiring upstream of ignition switch
-ignition switch itself (what a pain to get to)
-wiring downstream of ignition switch

For reference, as is standard with a DIY megasquirt installation, my car has the main relay wiring modified. On the adapter board, harness pins 27 and 36 are joined. 27 is hot start/run from the ignition switch, pin 36 is the pin the Motronic grounds for the main relay. Once these are joined, the +12V 0.5 RD (Figure 3) is snipped, pin 85 (.5BR) location swapped with pin 86, and the pin 86 wire spliced to a ground. Now when the key is turned on, +12V goes from the ignition switch, through a path to the engine harness, to the coil on the relay, which is grounded through that new ground. Pins 85 and 86 were swapped in case there is an internal flyback diode in the main relay. The +12V path from ignition switch is a problem.

The 325 (Figure 1) and 318is etm (Figure 2) shows it via C200 (2) -> S231 -> C104 (2) -> C104 (1) -> C101 (7). So I need to go through that path from Ignition switch to pin 27 of the ECU to eliminate a wiring issue along this path. What a pain.

Figure 1:


Figure 2:




Figure 3:

[zwill23]

When the car shuts down and you can't get it to fire back up, does the engine turn over?

I had the exact symptoms you described for nearly a year until I found a bad fuel pump connection. Intermittent no-start, dying while driving, but would always restart after some fiddling until one-day it just didn't. When I first purchased the car I installed a Walbro fuel pump and soldered the connections at the stock bracket, over the course of three years fuel sloshing in the tank fatigued the soldered connections and cracked them.

If you're experiencing no turn over at all I would be tempted to replace the ignition switch, I have an airbag car too and it only took me a half hour to replace it. The bigger pain is the price tag on the switch :)

[varg]

It's definitely an everything cuts out issue, since the needles drop and ABS cycles when it happens. Tracing the wires that go to and from the main relay I peeked up under my dash from a new angle and found this, bad sign.




Removed the knee bolster and found this.



So I ordered a new switch because to pull the harness out far enough to fix it properly I'd have to pull the wheel and remove the column cover anyway. My best guess is that someone had an alarm with immobilizer in this car, which was installed with the cheap yellow butt splice connectors. Removal of it was done with the translucent butt splice connectors, which there are more of near C302, see below. My best guess from the colors is that they belong to C103 and are a ground (brown) and the marker lights.



When I was looking at trucks last year one truck which had another major issue that kept me from buying it (53 casting block in a Cummins powered dodge is a time bomb) had an aftermarket alarm. The owner was puzzled when I considered the aftermarket alarm a liability. This is why. In my experience they're often installed in an unscrupulous manner and are a pain to remove. This isn't the first time I've seen something like this.

[zwill23]

God damn! That'll do it...

[Digitalwave]

That's some shoddy work for sure. Luckily with a new switch and some new (proper) splices, it shouldn't be too hard to fix up.

[econti]

At least it was fairly easy to find what was wrong. I hate wiring.

[varg]

Well, that's if this is the problem and I'm sure hoping it is. It sure looks like it is but I was unable to replicate the issue by moving the wires around and pulling on them while the car was running. The only hint of a bad connection at all was the burned wire and the fact that the butt splice it is coming out of gets a slightly warm while the car is running. It's possible that this wiring hackjob damaged the contacts in the ignition switch somehow as well, I won't know until I take it apart after installing the new one, which should arrive next Monday. In the mean time, I won't be able to take any more sanity drives in the E30.

[varg]

I got the ignition switch several days earlier than expected and replaced the old one today. Kind of a pain but wasn't as bad as I expected since I didn't have to pull the top column cover or the tumbler.









Not sure what these unused connectors are, I was only previously aware of two of them.









And unfortunately found some serious rust that was not visible last time I was under the back of my car. To my knowledge, at last battery change, there was no rust in my battery box. This killed the mood from fixing the ignition switch issue. Since I have a truck now, I'm just going to keep my E30 in the garage except when the weather is dry to stave off the rust as best I can. It sucks, but it's what I have to do. I can't afford to strip it down and restore it any time soon and these kinds of body repairs and paint work are outside of my capabilities since I do not have the tools to do them. Hopefully by the time I can afford to fix it the car is not beyond saving



(edit)
I really hate this new forum software. You click fullsize and it gives you whatever size it feels like giving you. All of those images should have been exactly the same width and these randomly sized images make the posts look terrible.