Over the weekend, I was able to accompish a lot in terms of diagnostics and understanding how the whole ignition system in my car will work. Yes, I did know that I was going to drive a trio of coils using the BIP373s as switching devices, however, there were a lot of details I was unaware of when applying this sort of setup.
Upon arriving at home, I decided that I’d started playing around with Tuner Studio’s output testing feature, which REALLY comes in handy when trying to diagnose electrical issues. I started by testing out the injectors, which fired my injectors at whatever speed I set and whichever bank I wanted as well! I was able to discover I had a blown fuse on my relay board for one of the injector banks. Following this, I gave the coil testing feature a whirl, but as I expected, I had no spark. Before walking away from this issue, I did notice I didn’t have capacitors attached between the power source and ground for my coils, so I went ahead and constructed this little guy, a two capacitor combo in parallel for a total of 20 μF (didn’t have a 25 μF unit on hand, so I used what I had). More on this later.
I turned my attention to the fuel leak issue. One source of a leak was from the fuel rail/fuel injector seal. These injectors date back to 1998, so I’m not surprised that the leak popped up. I took a trip to my local Car Quest and picked up some O rings for the seals between the rail and injectors. Unfortunately, they did not have enough in stock for me to replace the O rings sealing the injectors to the manifold, so I left the stockers in and only replaced the O rings on top. Additionally, I had the leaks at the ends of my fuel lines as I mentioned before. I was able to solve this issue by picking up the Swagelok fittings and will be using these to make the connection between my braided hose and my stainless steel fuel hard lines. I was only able to install two of them before calling it a weekend, but I will return next weekend to install the remaing two and see if I can build fuel pressure in the system with no leaks. Fingers crossed!
Now, here is the part where I talk about what consumed 80% of my weekend! Spark! Ah, the trials and tribulations of diagnosing an absense of spark, gotta love it, right? As frustrating as this process was, I learned a lot from this, and I’m glad I did.
I first began by checking my physical wiring from the relay board to the coil pack connector itself. I traced all the wires back and measured the 12 V source and everything appeared fine. Upon further investigation, it turned out the I had my wiring completely backwards on my coil connector! I learned this by measured the resistance between the pins on the coil pack. By doing a little research, I discovered that the optimal resistance should be 0.7 – 0.9 Ω across the ground and power pin on the coil pack. I went ahead and did this using the voltmeter feature on the fluke and it turns out that these resistance numbers were rendered when placing the positive probe on the far left pin and the negative probe on any other the other three pins of the coil pack if you’re looking at the coil from the top with the connector at the bottom of the coil pack. This led me to the conclusion I had my wiring backwards. I went ahead and fixed this, but before soldering everything together, I started testing my outputs one by one and seemed to have no success consistently returning a voltage when the transistors were switched on/off via Tuner Studio’s testing feature. I banged my head on the wall over this all Friday evening, and finally decided to call it a night and do a little Googling before heading to bed.
On Saturday and early Sunday morning, I occupied myself with more testing and failed attempts to get the coils to fire. Along the way, I managed to smoke the IAC2A output trace on the relay board, which resulted in discontinuity. I really thought I was SOL at this point and would need a new relay board, but I knew I didn’t have the cash for a new one (~$100.00) and this would be the second one I have ruined! So, what is a guy to do other than flip the sucker over and solder a jumper directly from the DB37 pin over to the corresponding screw terminal on the bottom side of the relay board. Voila, problem solved! That was a huge relief.
At this point, I really started to make progress. One of the big things I noticed is that I was getting constant +12 V on the Spark B and Spark C outputs, even when the key was off. I knew this had to do with the wiring on the Megasquirt board itself. I knew this because if there was a voltage, then that meant something inside of the box is grounding and resulting in a voltage potential, which is only supposed to happen when the appropriate coil is activated. I cracked open the ECU case and come to find out, the BIPs were making contact with their heat sinks and the heat sinks were making contact with the case of the megasquirt. I moved some things around, installed a couple mica insulators, secured the BIPs to the heat sinks with PLASTIC screws, and closed everything back up. This was a unique process of diagnosis because every time I would slide the case on about half way, the output would ground (I used a test light to diagnose this). I slid the case back and forth about a dozed times before I came to this realization, I was a little dumb struck after the first couple times of doing this.
Once these grounding issues were figured out, I thought I had everything working right after I tested all outputs using my test light to ensure each coil was grounding when I selected the appropriate one in Tuner Studio.
I then hooked up my connector to the coil itself and I started getting intermittent, weak, or no coil firing at all. This was disappointing considering I had just tested all of my outputs to be consistently at the same voltage at certain intervals. I knew that my wiring was right, so I turned to my coils. Throughout testing and diagnosis, it turns out I managed to screw up two different coil packs. The first coil pack would fire the Spark A and B output coils just fine, but Spark C wouldn’t fire at all. I confirmed it was the coil that was the issue by individually testing each output using a Blaster 2 single coil. I conducted this test by installing a couple leads into the coil pack connector and attaching the +12 V source to the positive post of the Blaster 2 and the respective output lead to the negative post. Every single output turned out to fire that coil flawlessy at various intervals/speeds.
So, here I am now with two blown coils and now looking to pick up another EDIS coil from the local U Pull It tomorrow, hopefully. Luckily, these things are fairly cheap, so I’m not out a whole lot of money.
Along with the Z car stuff, I helped a buddy attempt to install my cat back on his new STi, and then ended up installing it on my car (I’ve been running an open down pipe for a while, became very annoying). I also was able to wash and vacuum my Subaru for the first time in months, been too busy with school and work to get around to it.
Anyways, that’s all I have for you guys tonight, next weekend I’m expecting a start but if not, I’m not sure I’ll make it to the Branson Z Fest. 😥