The Grey Ghost

[GHSTII]

This is the Build Thread for Grey Ghost. Reply below.

 

[GHSTII]

Forgot I could do a build thread, so I’ll do a recap here

When I first bought the car a couple years back, I knew it wasn’t running. Since it already had high miles, I took it as an opportunity to “reset” and try to reinforce some failure points for the future. That kicked off quite the undertaking.

Initially, I thought I’d just replace the “weak” and “broken” stuff. Of course, I couldn’t help but make it more complicated than that.

 

[GHSTII]

I’m not very mechanically inclined. Though I’m grateful to have learned a lot since then, I do have good friends at a local shop who agreed to do the heavy lifting to get it “runnable” again. I love those guys.

It all began with pulling the engine to be sent off for a full rebuild of the long block. The motor was flooded, and it was pretty bad.

 

[GHSTII]

I knew the motor was supposed to be gone for 4-6 weeks, so I took time focusing on collecting the other stuff I already knew we’d need.

I refrained from wondering too much what to do about the sunspots, because it was already inevitable that this would be a big project. I stuck to what the guys and I agreed were of priority, and tried to read up as much as I could to figure out some details that I just couldn’t know without an assembled vehicle right in front of me.

Little did I know, the motor wouldn’t be gone for just four weeks, but rather four months.

In that time, I got antsy and a little ahead of myself. Despite me already being kind of cringe, I couldn’t do much more than visualize and speculate on what to prepare for in the meantime. I did get the things we needed, but also tried to use foresight to “get ahead” a bit.

 

[GHSTII]

Finally, the engine came back, and it was game on.

I did have to wait a while because the winter was rough for everyone, but the guys still had my back and let me help put everything together while it was in the shop.

It was really cool seeing it evolve in real time. I’d owned a GS as my first car, and a GST after that, but never got anywhere near this granular with a project. These guys made a childhood dream possible, and I can’t speak to how much it meant that they were so willing to get me up and running. If the literal heavy lifting and assembly could be completed at the shop with their superior equipment, I could pick up in more familiar territory when it could roll into my own garage.

Main obstacle was that they were mainly BMW guys, and I myself had been in an N54 for the past couple years. Despite being up to the challenge, there was a harsh learning curve for everyone that stumped momentum a few times.

A goofy issue caused by my ambitious ass led to the O2 housing and turbocharger I decided to upgrade late in the game barely clearing when we attached them to the manifold. At least they fit

Otherwise, it really did all come together nicely.

 

[GHSTII]

Soon, the car did actually roll into the garage at my apartment. Something tricky though…

It started up for the first time at the shop, running nice and strong. This was even before a proper tune, so we left it off for quite a while. However, one day it just wouldn’t start again. I was paranoid that it flooded, but there were no apparent signs that was the case.

Despite me being terrified, I made the call that it was “close enough,” towing it home to save their time while I figured it out.

Then I was immediately overwhelmed at how messy the interior wiring was. None of us had paid attention to it before. I was able to connect to the ECU via ECMLink, but none of the gauges were working. I wanted to at least be able to see the oil pressure, but the wideband seemed almost just as important.

I was nowhere near prepared to figure out the wiring, so I focused on finishing the remaining mechanical concerns (which wasn’t fun either).

 

[GHSTII]

Among other things, I actually spent weeks trying to replace axle shafts while trying to avoid reckoning with the electrical issues. I figured out the passenger side pretty quickly, but the driver’s side is what stumped me.

I avoided posting on the forums to ask for help as long as I could, as I already felt stupid enough. Nonetheless, it proved necessary, because I never would have independently figured out that I was trying to pull the intermediate shaft. Problem was solved in a day I think.

 

[GHSTII]

After a long while of scratching my head with the electrical issue, I began to entertain a new approach.

What if I just replaced the whole OEM control harness? Better yet, what if I just pulled all the wiring, and put in a new harness going straight to a stand-alone ECU?

Of course, it was insane, as I now well know. At the time, the mere task of sorting out the wiring was unfathomable, so why not go big? I was in deep anyway, and I’d replenished my budget from all the down time.

After research, I decided Haltech was too expensive, despite being clearly the ideal option in my case. I shopped around, and somehow landed on FuelTech. I liked that the ECU was the screen itself, and seemed to connect straight to a terminated universal control harness with all the sensors I’d need.

Despite ordering and even beginning to install the setup, a glaring issue arose. The base-map generator in the software made it clear that the FT450 didn’t have enough outputs to control the step-motor, which seemed essential for my car.

I also noticed that the harness wasn’t quite as “plug and play” as I naively thought. So I actually gave up and sent it back, opting to take on the spaghetti monster afterall.

 

[GHSTII]

My brother’s girlfriend caught wind of my predicament while he and I were talking, and it turned out that she was actually quite comfortable with wiring. She was an electrical engineering student, and offered to take a look despite not being into cars outside of supporting him.

The two of them came into town to help out, and she made quick work of clearing out a lot of the mess.

Meanwhile, my brother helped me figure out the Speed Density setup, which the shop made it clear was essential given some of the more drastic changes I’d made during reassembly. While he helped me with that, his girlfriend taught my wife to solder, who then taught me.

After the two left town, it still took a while, but they left us wayyy better off than before. One day, the wideband finally came on. I decided that was enough to try and start the car again, as I wasn’t getting anywhere with the other two gauges.

That said, the Flex Fuel sensor wouldn’t stop leaking during my following attempts, so I removed it despite being excited about it.

 

[GHSTII]

Despite having grown rather comfortable with ECMLink, nothing I tried was working. After replacing my wideband, I became desperate for help, and found a couple tuners who were willing to check my logs until it became clear to them that something else was wrong.

I had to stop for a little while anyway, as my wife and I managed to find and buy a house. Fortunately, it came with a big garage, as well as new inspiration.

It had been almost a year and a half since I bought the car.

 

[GHSTII]

Nonetheless, another couple months went by and I basically gave up. I stopped trying altogether.

But I did eventually decide to try calling around until I found a shop relatively close that might know these cars. I’d tried before, but to no avail.

Then I got lucky. I sent off the car an hour away, and they called me back a couple weeks later telling me I could drive it home.

They fixed a couple minor issues I missed, but the starting problem was resolved by a tuner they called in. For whatever reason, it wasn’t getting enough fuel before. They recommended replacing the injectors, but handed it back off to me to drive for a little while and put miles on the engine.

I immediately noticed it was running quite rich, and it wanted to die upon applying throttle. After reviewing the settings, driving it around and reading the log, it seemed apparent that the injectors were the problem.

 

[GHSTII]

Switching out the FIC 850cc low Z injectors for 1650cc high Z, I was able to properly scale fuel and figure out the rest of the tune. Considering it started months ago, the old injectors presumably failed at some point early on, despite me buying them new. The switch was a tricky job that was new to me, but I was happy that something finally panned out the way it was supposed to by my own hand.

Now, I could happily cruise around and properly break it in. No signs of engine damage. I was very grateful. I went and saw the guys at the first shop, who were happy to see it too.

But I still had a couple other problems to work out.

 

[GHSTII]

I finally got brave enough again to deal with the rest of the wiring and properly replace the oil and boost gauges. I wanted them working before I introduced boost.

Unfortunately, something happened beforehand. Despite my wife and I agreeing to always solder, I broke the rule and tried crimping a ground inside the car when we were already having problems. It was an unnecessary risk, and I paid for that.

There was some kind of short, and a ton of my internal accessories went out while we were on a test drive. Luckily, that didn’t include the wideband, and I knew we could make it home.

Afterwards, while troubleshooting, ECMLink stopped responding.

The OBDII port went dead. We replaced the ETACS-ECU, but that wasn’t the problem. Suspecting the ignition switch, I decided to risk installing a “push-to-start” button, which turned out well. It didn’t fix our issues though.

 

[GHSTII]

After losing contact with the ECU, I was fed up with the wiring issues.

This is a good time to mention that when I sent back the FuelTech ECU quite a while ago, it was originally to exchange the FT450 for an FT550. It was allegedly equipped to support the step-motor.

When I gave up altogether, I kept the wideband and sent back the rest. However, the shipping company messed up. After delay, the FT550 and all its accessories returned back to me. Instead of trying again, for whatever reason, I decided to pack it away for a rainy day.

That day came.

Despite yet another steep learning curve, we pulled the OEM control harness, and cut out the pigtails of the connectors we still needed. We soldered everything. No mistakes this time.

Learning I’d need a fuse & relay board for outputs, I bought a PDM instead from ECUMasters. I yet again spent WEEKS reading the diagram backwards, but ultimately figured it out.

Also, Jason Drew just by happenstance was switching to the FuelTech FT550 as well, and his job was much more invasive. I asked, and he kindly sold me the same custom dashboard platelet he 3D printed for his own car.

All in all, I’ve never been more proud of accomplishing such a tough job. It responds, it cranks, and even starts.

But it won’t idle.

After all that, the ISC won’t work with this setup. Instead, I was prompted to use a Honda B-Series Idle Air Control Valve. That comes in today, but so far I don’t know how I’m going to make it work (except for the wiring, ironically).

TBD…

 

[GHSTII]

Along the way, I’ve come to realize just how motivating it can be just to make visual changes. Of course, everyone likes that. But between the big jobs, smaller projects like these really do put a smile on my face, and they motivate me to keep pushing forward.

 

[GHSTII]

Turns out I was able to ditch the ISC for a PWC valve instead. Once I realized there was an adapter that could be used to mount it, I had no problem wiring it into the harness.

The FuelTech ECU has a built-in compression test, which was a pleasant surprise I was previously unaware of. I ran it, and the numbers don’t look bad (at least as far as I can tell).

But the component I’ve definitely spent the most time on has been the CAS. I initially had a standard non-adjustable 2G CAS, which of course happened to be inherently unusable without modification.

Kiggly apparently makes a great crank trigger set that it seems most DSMers have been using or switching to. Without it, options are limited. I had a hard time figuring out what I wanted to do, much less how comfortable I was with whatever I’d choose to change. Reluctant (and honestly scared ) to finagle with the crank trigger, I instead opted to switch to a 1G CAS. With it, I could get the crank and cam signal directly from just that device. That’s not to mention it would also be much more within reach, being right in the middle of the engine bay. My main concern though would be that I’ve only seen these used on six-bolt motors running the FT ECU. Still, it seemed like the best option in my case. Something to figure out.

Buying one off a local fellow DSMer, I didn’t have too much trouble with the install. I was confused by the wiring for a hot minute, which I felt dumb for later. Nonetheless, I finally got signal from both crank and cam. I’m now waiting for a Yoshifab trigger disk. It will be inserted into the CAS for a compatible trigger pattern, which should be sufficient to start the car.

After that comes the process I dislike the most…

Timing.

 

[DogWhistle]

Congratulations on your diagnostic, mechanical and electrical skills advancement. It’s the trifecta of owning any vehicle, let alone, a DSM.

After all of the failed starts or smooth idling, all of us are anxious for the final programming and driving.

 

[GHSTII]

Congratulations on your diagnostic, mechanical and electrical skills advancement. It’s the trifecta of owning any vehicle, let alone, a DSM.

After all of the failed starts or smooth idling, all of us are anxious for the final programming and driving.

Thanks! Fingers crossed

 

[GHSTII]

Breakthrough!

Depicted below is the stock trigger plate that I pulled out of the 1G CAS. You can apparently make do with this as long as you tape off one of the holes (I believe the smaller one). I actually first set up to try that, but I found the method kind of sketchy. That, and fellow FuelTech DSM’er Kevin Collins swears by the Yoshifab disk I mentioned in my previous update. In the other picture below, you can see where I inserted that Yoshifab 12-24 tooth trigger to replace the stock one.

Another option is to use a Mazda 4-1 trigger plate, which looks just like the stock disk except without the smaller slit you would otherwise have to cover.

Anyway, even after wiring in the Honda IACV and tying up other loose ends, the car still wouldn’t idle. I got really frustrated and felt kind of hopeless, but one of the FuelTech Australia guys hopped on a short session with me to troubleshoot.

Turns out I just had my Fuel Pump on the wrong setting

With the tap of a button, the whole thing fired right up just as it should. I was ecstatic.

My wife and I got ready to set the Timing, which I mentioned before to be my primary point of dread. However, it turns out my initial ignition calibration was on the dot.

At first it felt like luck, but it’s really just that I manually cranked the motor to TDC and input the settings most commonly recommended by the 4G63 guys with the 1G CAS setup. Almost no adjustment was necessary, as the timing light kept firing at the TDC marks without fail.

Now I can also confirm that the 1G CAS with a proper trigger disk will work on a 7-bolt motor just as well as a 6-bolt. I’ll update if any issues seem to surface, and you can correct me here if you think there’s some reason not to use it. Meanwhile, so far so good.

Despite all the grief, I do believe it is objectively a better setup than before. I now consider it a blessing in disguise, because the deeper I go with this project, the more weird little gremlins I’ve encountered and extracted that I haven’t mentioned on this thread.

My next objective is to dial in the idle and assess what else needs to get done before I can actually drive and tune. The main question is whether to shoehorn back in the spare parts of the stock control harness to salvage its original auxiliary functions, or to build creatively from my new loom to recover the lost accessories.

I’m actually leaning towards the latter… because though I don’t enjoy taking things apart, the changes so far have made it much less overwhelming to work on. The new routing is just simpler.

We’ll see though.

 

[GHSTII]

Couldn’t get my alternator to charge after multiple failed setups. It was probably just bad. So I switched to a 200A MCR Single Wire. It involved lots of grinding, but I managed to fit it in the stock location nice and flush. Added a grounding strap, and it gets 14.5 volts consistently at idle. I repurposed the stock tensioner bracket with a longer bolt, but then that somehow flew off during a road test
I’m replacing it with a Morrison Fab Saturn Alternator bracket that I purchased from someone else. Waiting on that now before hitting the road again.

I ran into some other goofy little problems, but luckily connected with fellow DSM’er Steven Johnson. He helped dial in the car, let me ask tons of questions, and at a good fair price. 10/10, would recommend. He really knows his way around a 4G63.

I switched from a manual boost controller to an electronic valve, and then we were almost ready to properly tune. I was also convinced to switch to E85, which the car responded to shockingly well.

I chased boost leaks around the throttle body for a bit with a newly purchased pancake compressor, and ended up replacing my BISS screw that had been worn down to nothing. Wired the SPAL fans up to the PDM with an analog trigger to be controlled by the FT550, and we were finally in business for real.

With everything working right, I was shocked at how quickly we got the thing running properly. There are still a few bugs, and it’s definitely still a DSM, but its performance actually blew my mind. The FuelTech unit is amazing as far as I’m concerned, and now that I understand it all better, it’s in a sense much simpler than the setup was before on ECMLink. Glad we took the shot at it.

Also, I guess I’ve decided to ditch the IAC valve. At least for now. I might try it out again later on, but after I got it running properly with no actuator, I switched to “Idle by Timing” with satisfactory results.

Otherwise, pretty sure we’ve maxed out the power potential for where the car currently is. I’ll have to upgrade my injectors to get anything more out of it, ‘cause the FP Red UHF just keeps going. I love it! It would be sick to get on a dyno and see what it’s actually making.

Meanwhile, I’m just going to try and address the little things and increase my confidence in its consistent performance. Excited for Spring.

 

[chrisman287]

Car looks great! Glad to see it up and running after all the trials and tribulations

 

[GHSTII]

Car looks great! Glad to see it up and running after all the trials and tribulations

Thank you!

 

[1cleangsx]

Just read the whole thread and I must say it’s very inspirational! I’m happy for you man and that’s a good looking car you have

 

[GHSTII]

Just read the whole thread and I must say it’s very inspirational! I’m happy for you man and that’s a good looking car you have

Thank you, sir!

 

[GHSTII]

Just a brief update.

Swapped out the Walbro 255 that was in the tank for a Walbro 525 “Hellcat” pump.

I stripped the stock MDP sensor threads while figuring out the setup, and for a while I was running with my intake manifold sealed with JB weld

I procrastinated, but did ultimately implement my permanent solution, which was just to swap the entire intake manifold.

Went back and forth about it, but I decided to relocate my battery to the trunk to make more room in the engine bay. Voltage drops a little lower than I’d like at cranking, but it runs just the same as before when it turns over. Figure I’ll see what I may be able to do about that drop later.

I put an Oil Catch Can in the bay. Set it up “sealed,” with a vacuum line going to the intake pipe.

I actually went to a dyno, only to realize I did all these upgrades just for my clutch to slip because I somehow forgot to upgrade it before the motor was put back in the car Nonetheless, I took the car to a Test & Tune for fun before swapping the clutch. Besides some mildly entertaining clips, nothing spectacular happened, but I’m glad I went. It was nice to be among other people testing their projects.

Had to drive an hour out, but I found an old-school DSM guy to install an ACT Twin-Disc Sint Iron clutch, and installed a Clutch-Stop.

It took multiple attempts, but I completed my dash by cutting some ABS board and soldering it to the stock dashboard surround (as prompted by jdxnc).

More underway…