27 Jun 2023 (Pics added 01 Jul)
Summary of work in the last several days:

1. Did some fab work on the car on the left side of the radiator, to give more room to allow the Needswings DCAI piping to fit. This is very tight on every Crossfire, and with the relocated radiator package in this car, it became impossible. The sheet metal to the right of the left headlamp was sectioned, and them replaced with a "C" shaped piece of stainless steel, and welded in place. This was then painted, and once everything is in the car, almost no one will be able to tell that a change has been made, but now the DCAI will fit properly!

View that shows how a 3" pipe will not fit in there. Squishing a hose resulted in it simply sucking all the way closed and flat during hard acceleration.


Detail showing the metal that was cutout


And here is the final result (before complete cleanup and paint). The 3" pipe fits right in there now. A 3" ID hose will still squish a little, but just barely. Should not suck closed anymore.


2. Did heavy maintenance on the engine. Both oil pans and the front timing cover were removed. Of course this meant removing all of the accessories from the front of the engine, the crank pulley and both valve covers. The engine actually looks pretty small! The oil pump was removed to replace the check valve, which is inserted into the block. After working on the E55 (find that thread in the MB section), it was found that all rubber pieces, including this check valve turn to stone after all these years. The check valve presses into the block with a rubber seal holding it in place. When the oil pump was removed, the check valve fell onto the ground. Needless to say, it was no longer doing its job. New one installed. New oil pump chain, tensioner and spring installed. M113K oil pump chains wear and stretch, causing rattling noises, especially at idle. This was was doing that lightly.

What your engine looks like with both oils pans, timing cover and valve covers off. All M112's and M113's are pretty similar. ALL get that awful brown varnish color to them.

3. Timing Cover coolant o-ring seals. I will be posting pics of this. ALL M112/MM12K/M113/M113K engines have these. They start life as squared off, hard rubber seals. These were working, but when they fail, coolant dumps into your oil - not a good thing. Here's the other hard fact: These seals WILL fail in your engine eventually. No doubt about it. When the cover came off this engine, one seal was turning a bit gelatinous, but it was ok. The other? Mushy, swelled and partially torn. A ticking time bomb! The timing cover is a ton of work, especially on a M112K/M113K (supercharge adds extra fun here), but I do know that all of my Crossfires will get these serviced at some point. Pics coming on this. It will shock you!

These are the seals that fail. This one was working, but on its last legs. You can see that it is distorted. It is actually swelled, and gelatinous. Also, it was beginning to tear on the inside edge at about the 10 o'clock position.

4. Oil pans both cleaned, and resealed. Rear main seal replaced. Front timing cover resealed, and new front main seal. Everything reassembled. Valve covers got new gaskets and top sections were leaking, so they were resealed, too.

5. Spark plugs and wires replaced. Found one spark plug with a severely cracked insulator. Could explain the miss I was hearing at idle.



6. "Stock" exhaust manifolds replaced with new NeedsWings big stainless manifolds with black Jet Hot coating. Getting the old ones off was a snap. Putting the new ones on required removing the right side engine mount and the starter to gain access to the nuts to tighten with a wrench. Sockets would not work on most of them. This was a decent chore out of the car. In the car? WOW! Not sure anyone wants to tackle that. Looking forward to them, though, that's for sure!

All dressed up and ready to go!


7. Attached the transmission, no drama there

8. Installed powertrain in the car to test fit several things. Also to take measurements on the driveshaft and crossmember. Details on this forthcoming.

Back where it belongs!

9. Completely changed the plans on the Killer Chiller. I really have to say I am disappointed with that whole company. They were nearly a year late on delivering the KC. The tank they sent was not the one on their website. They shorted me several A/C line fittings. Their mount literally broke as I was just looking the thing over (one of the bolts welded to the tray popped off from poor weld penetration). The unit itself looks like it has been upgraded to the same core that Forced Induction is now using. That's about the only good thing to say about it. I will not be mounting it up in the front bumper, but with good ideas from @CL770, I will be putting in in the left side fender well, just forward of the left door. Inside there is a nice little platform that the KC core can sit on. Up above it where the A/C lines run into the firewall. I dropped a spare set of A/C lines off at Advanced Auto Air in Pensacola, FL today, and they are going to modify those lines to have a Tee in them, and from that Tee, I will feed the KC core. There is a convenient oval shaped hole more than big enough to feed those down to the core in the fender, and the hole is even beveled to prevent chafe. It's like they thought of this! For the water side, I will eventually have a tank fabricated that is the same size as the battery, and put it in that location. So picture this. The stock heat exchanger has been removed completely, including all of its plumbing. From the cold side discharge of the KC, a 3/4" hose will go through a new hole inside the windshield cowl, and travel across the car, to another new hole right where the tank will be. This will dump into the tank. From the tank, a short hose will go to just behind the right headlamp where it will find a newly installed Pierberg CWA100 water pump. Discharge from the pump will go into the feed side of the intercooler. Coming out of the intercooler, the hose will drop down to the OE type Bosch 010 pump, which will discharge into another 3/4" hose that will travel across the front bumper, underneath the left headlamp, and inside the fender liner, up and over the left front wheel, and down into that inner fender area where the KC is. Routing like this makes a big loop of the engine bay, but it also helps keep the hoses mostly away from the engine. Mostly. In theory, this is going to work great. Back to Killer Chiller, don't send them your money. Just don't. Go for an FI Interchiller, or if you have the scratch, go to ChemCool, who are really innovating some neat stuff.

 

01 July 2023
It's 130 in the morning, and work has just wrapped up for the day. Work starts late in the shop and runs deep into the night, because it has been nearly 100° and humid lately. Better to work in the night and be cool. Tonight was a revisit to the batter relocation project, which had been mostly done, but not completed. Pics of the trunk are coming, but a second set of hands are needed to tighten hardware and bolt everything down. Hoping to get that in a day or two. Up front is a small detail that took a lot of time to get just perfect, and that was the battery jump post and cable termination. It makes more sense to be able to jump the battery from the engine compartment than anywhere else. But as more and more things get crammed into this car, there is less and less room for even the simplest things, like jump posts. To simplify, just a positive post was added and the negative cable would just clamp onto a convenient ground if it were necessary. To install this, a jump post pair was purchased for a UTV from Amazon, and then the negative post was liberated from the package with a chop saw. The remaining post and the rubber divider were retained, and installed here:


This made for a super sanitary install, overall. I am exceedingly happy with the overall battery relocation. You'll like the trunk setup when you see the pics! Believe it or not, finishing the new battery tray, this post and the terminals took the entire evening. At this point the car would have power if the cables were attached to the battery, which they will be in a day or two.

Now, it is finally time to let the cat out of the bag. There was another big reason to remove the engine and transmission. That would be the transmission. Don't get me wrong, the NAG1 5-speed was working perfectly well when removed, but it was time to step up the game! In place of the stock type (but very built 722.6 transmission, a later model 7 speed transmission is going in. Actually, it's already in the car! And not just any 7 speed - the big daddy - the 722.931 AMG Speedshift Multi Clutch Technology transmission. There is no torque converter. Instead, it uses a wet clutch. Ultra fast, ultra high performance shifts! Never heard of MCT? No worries, check this out:
CT
Here's a better pic of the engine and trans, right before it all went into the car.




This transmission came from a 2013 SL63 that was wrecked with less than 5k miles on it. It was purchased ridiculously cheap, due to a misprint in the ad. Kudos to the company for honoring the price, even though they realized it was a mistake. No modifications were made to the trans, save one, where the original ISM module that shifts gears electronically was removed, and a simple mechanical shift lever from an earlier 722.9 was added. Many thanks to @Dave2302 for sending me one over!

To make this work, several parts had to be fabricated. First off, the flex plate. The later model transmissions bolt their torque converters and in this case, the wet clutch up differently that the older cars. This requires bolts that come in from the backside of the flex plate, commonly called the slant style flex plate. ALL slant style plates use the later crankshaft bolt pattern that I simply call 4+4. There are two groups of 4 bolts, and there is a gap between the two groups. On the earlier cars, the 8 bolts were spaced somewhat evenly around the crank. As luck would have it, 4 of the holes and the alignment dowel hole line up between them. To remedy this, the four non-matching holes were welded up and redrilled on the earlier crank pattern. Voila! A late model flex plate that bolts to an earlier engine



There are a few other problems, too. One is that the output flange is now a 4 bolt flange. Another is that that flange is 40mm further back in the car, due to the transmission's longer length. This necessitates mods to the driveshaft. Another issue is the mount. The mount is right at 2" further back, and as a result, the existing crossmember will not work. To fix this, I took a stock Crossfire/R170 crossmember, and a W211 crossmember, and broke out the chop saw and welder. After tacking up, checking fit, cutting apart, changing, tacking up and checking fit a bunch of times, I finally got what I wanted, and made this truly ugly piece



For anyone else looking to put a 722.9 into their Crossfire or R170, even though the driveshaft will have to be shortened 40mm, the mount has to be lengthened about 51mm. I went into this thinking 40 mm, and made my initial cuts all wrong, which resulted in filler metal and lots of extra weld. Ugly.

Other things that have to happen mechanically are the transmission cooler lines, which were not photographed. The ends at the transmission are different. A used set from an R171 SLK were purchased, hoping they would bolt up. They won't. The solution in the end was to chop the ends off the R171 lines for the correct hookups to the transmission, and then do the same to the Crossfire lines, to mate up. Not very elegant, but should work just fine.

Next is shift linkage. A shift rod from a W211 was expected to make this work, when in reality, all that needed to happen was to loosen the adjuster screw near the shifter, and run the adjuster all the way in, and tighten back up.

The real magic in this is in the electronics. The 722.9 (or NAG2) comes in a lot of flavors. Early ones will 'speak' to the Crossfire's stock ME2.8 based ECU and other electronics. The MCT (and other later 7G transmissions), on the other hand is exclusively for ME9.7 and newer electronics, and they do not speak the same language. After working with a few real sharp guys in this area, codings and programming in the transmission's computer (the VGS) were made, along with a couple of key changes inside the car to modernize its CAN bus. I won't detail everything here, as this is not a how-to thread, but suffice it to say, this is not a simple project. To prove out the electronics ahead of time, a special wiring harness was made up to place the MCT's VGS and Valve body in the passenger seat of the car, to do the development. Once proper communication was establish and all configurations were made, that special wiring was all removed, and put up for this stage of the project. In other words, we have a warm fuzzy that this is going to work. This process here took over 6 months of tinkering to make it work.

The "hot mock up" in the car, testing the MCT programming, well before disassembling the car for this project

So, what are we looking to gain here? Well, a few things, really. First, the extra gears mean a wider dynamic range for the car. Second, lower RPMs while cruising. Third, performance. The MCT shifts gears within 100ms. That's 1/10 of a second! That's blazingly fast! If it doesn't work for some reason, the 722.6 and all of its pieces have been retained for that contingency. More on the MCT portion of this as time goes on. There were a few others involved here and there, but the main guys, @Dave2302 and @Deplore deserve a LOT of credit for their help and encouragement through this sorting out process. Dave has his own 722.9-in-an-R170 project underway in Scotland, also. We're all anxious to see if this works, as there are no MCT transmissions behind an M113K to be found - anywhere. Getting this to work opens the door for use behind any other engine in the family, including M112, M112k and M113.

Saturday will focus on the water plumbing for the Killer Chiller, and on getting the cooling package back in the car, and buttoned up. If time permits, the headlamps will be replaced with another set here, and the bumper cover will be put back on. Goal is to fire the engine up this weekend, but will not be able to drive it as the driveshaft is still another week out from the shop (they had problems). Exhaust will be wide open, as the pipes don't go back on until the driveshaft goes in. More updates to come!

 

02 July 2023
Slow progress today. Worked on relocating the oil cooler, and will have to work on it some more. Still not quite happy with it. The -10AN fittings are pressing against the bumper beam, and this will not be an ideal solution, as wear and tear are going to happen. Also, I am mindful that even the most minor accident would cause damage that would not normally occur. If something like this happened far from home, then what was already a bad time for any Crossfire owner could turn into a real problem for this car. Each thing that is done to this car has an element of this type of thinking involved. What if this system broke far from home? Can replacement parts be easily found and installed? Are the parts being used common enough that a fix can be applied? If not, what spares need to be in the car? Do we want that many spares? What about tools? All of these questions need to be answered, and a good risk mitigation approach needs to be applied. Obviously, a catastrophic wreck nullifies this all, anyway, but common breakdowns happen to everyone. No sense in making a bad situation worse!

Also worked on plumbing the intercooler circuit lines. Going from the original intercooler pump location, since most people know where that is. It is re-plumbed the following way: Water comes OUT of the supercharger/intercooler, and down to the Bosch 010 pump. This feeds a 3/4" hose the runs underneath the front bumper beam, into the left fenderwell area, up and over the wheel arch, and to that space at the bottom rear of the fender, just ahead of the door. This is a mirror of where the washer fluid tank is on the right side. This goes into the inlet of the Killer Chiller (KC). The outlet of the KC feeds a 3/4" hose that runs up through that nice oval hole below the fuse box, and runs between the fusebox and the fender, to a new hole that goes into the cowl. From there, the hose runs across the car to a matching hole in the cowl, putting the hose directly behind and above where the battery used to be. This is where a water tank will live (hoping to have it made in a couple of weeks). This tank will hold about 2.5 gallons. Another 3/4" hose will attach to the tank, and then runs under the tank to a mirrored oval hole that currently sits under your battery. This hose goes through a large number of 90° elbows to prevent the hose from kinking, but follows a path to the front of the car, and just under the right headlamp, it feeds back to a new Pierburg CWA100 pump. This pump then feeds the inlet side of the intercooler. The dual pump arrangement provides a lot more flow, as well as some redundancy in the event of a failure.

Tomorrow will finish the oil cooler, and then move onto a project that I neglected to do when the engine was out. I will drill and tap a 1/4" NPT fitting into the upper oil pan, a little behind the A/C compressor. Why do you ask? Another item to be added will be a Mann ProVent oil/vapor separator. This is like a catch can, but better, as it filters and automatically drains oil back to the crank case. Moisture leaves the separator and is vented off and/or sucked into the intake. This prevents oil in the intake, while preventing contaminated oil from returning to the engine. I have used these separators with great results in SRT-4 projects. Check them out: Provent 100 While this project will not be completed immediately, the drain hose is the tricky part of this, and better to do this with no oil in the engine. The time to have done this should have been last week when the pan was off the engine, but alas, it was forgotten about! Lower oil pan will be removed to ensure no metal chips will enter the engine's bloodstream. The hose will be installed, but capped for later use.

 

Is the black exhaust a ceramic coated version? And if so, does that truly work? (as in: would that be a viable upgrade for the regular V6 as well to direct heat away from the CPS and enginebay in general?)

 

I haven't fired the engine with these headers. However, I have had various jet hot coating on exhaust manifolds and headers in other cars, and it has made a marked difference in underhood temperatures/

 

12 Sept 2023
It has been quite some time since there has been an update. Things have been happening, but progress has been slow. There are two experimental aspects of this latest round of the project:
1. The rear end
2. The transmission

Lets talk about #1. The rear end is a HAG215, from a 2008 Dodge Challenger SRT-8 6.1L Hemi. It has a 3.06 gear ratio. It is essentially the same as the HAG215 found in a number of MBs, but it is laid out somewhat similarly to the Crossfire's rear end, with the exception of the dogleg rear cover, found in later MBs. The front mount of it is very similar to the Crossfire. As you may have read in previous posts, using it required quite a bit of fabrication. If it were not for the fact that I already had the HAG215 and a Wavetrac for it in hand, there is no way I would have gone this route. Instead, I would have gone with a HAG210, found in the W210 and other cars, and I would have gotten a Quaife LSD for it (no Wavetrac for the HAG210). What's done is done, and we moved on. The mechanical aspects of the HAG215 have been worked out, and it is in the car. What was not reported in the earlier parts of the thread is that the tool that was constructed for measuring the shimming actually broke during reassembly. This invalidated all of the readings that were taken. Given that, best guess on shimming was made, and a wear pattern check was taken. In short, there is too much clearance (backlash), and now the rear end whines quite a bit. There is still light at the end of the tunnel, though. There is a plan to fix the whine! The 3.06 gears were always a temporary solution, until a set of 2.65 gears could be found. These will work with the 7-speed transmission more properly. Right now, 1st gear is not especially useful, and just leaving the transmission in Comfort mode, leaving in 2nd gear is ok. So let's look at a chart, that is part of a larger spreadsheet. This sheet was developed by me, and is nothing magical, but surprisingly accurate.

What needs to be pointed out are 1st and top gear for each transmission. Notice the spread in gears of the 7G vs the stock Crossfire 5G. First is much lower (numerically higher) and top gear (7th/5th) is much lower. In other words a much wider range of ratios in the 7G. For now, let's look at Final Drive. Final Drive (FD) is the product of whatever gear you are in multiplied by the differential ratio. So, on a stock SRT6, 5th gear is 0.83 and the differential ratio (3.07) is 2.5481. Now let's take a look at 1st gear. Stock SRT6 would be 11.0213. Actually, a 1st gear multiplier of between 11-12 in a street car is considered ideal, so it is very well matched. Even a Limited with its 3.27 ratio is 11.74. Great! Now this car is technically using a 3.06 ratio, but for all intents and purposes, directly comparable to the SRT6. So, we know what first gear looks like on the 5G already, how about that 3.06 against the 4.3772 of the 7G 1st gear? 13.394. Yikes! Way too high! And this is in a car that makes a ton of torque? Can we say tyre-fryer? So, now we step off to a 2.65 differential ratio, and now we arrive at a much more usable 11.5995 multiplier. Will the car still light up the tires? Well, yes. But now let's look at the chart, and see what the highway RPM looks like. The chart has 75MPH entered, and this car's rear tire size. If you look at 7th gear, and follow it across to the 3.06 ratio, you'll find 2181 rpm, which is about dead-on for what it really is doing on the highway. Now follow that over to the 2.65, and you find 1889 rpm. That sounds low, but remember, there's a big V8 pulling this along. For additional reference, look at the 5G trans and 3.06 that was in the car previously: 2488. So between the 2.65 gears and the 7G, this engine will turn 599 LESS RPM going down the highway. That will reduce cabin noise substantially, increase economy, and result in a little less stress on the powertrain. Wins all around!

(For those wondering about all of the differential ratios, this spreadsheet was originally started while researching gearing on my old Mustang SVO, and there are gear ratios in the table to cover those, along with several other transmissions found in other models.)

Now, onto #2. The transmission. This is the big thing that is different in this round of changes. Some pretty significant challenges with the physical fitment were apparent from the start, requiring a modified driveshaft, fabrication on the transmission crossmember, modified transmission fluid lines, and several other items. That portion all worked out great! Now, on to the transmission electronics, fluid, valve body and more. This transmission is actually a 2014 722.931, which makes it to where it originally ran as a start/stop ECO transmission, on far later electronics (PTCAN FBS4, for the nerds), and the later 236.15 "blue" fluid. Not going to get into it all, but a lot of mistakes in this regard, but that happens every time new ground is broken. Someone has to blaze the trail!

This is a 2004 Crossfire, which means it is using 'leftover' R170 SLK320 ESP (ESPMk203 ABS) module and stability sensors. Very quickly it was realized that the 7G would not talk to the ESPMk203 properly, and an upgrade to the SRT6 ESPMk25 would be required. They look about identical, and the brake lines go to the same places, so no problem, right? Um, no. This required the main connector to be swapped, as they are different (soldering almost 50 wires), different stability sensors (in center console, behind the shifter), and a few changes to even how the brake lamps are wired. All of that was done in early August, and that made the 7G happy again. This is required, because the 722.930/931 does not use a torque converter, but rather a wet clutch (called NAK in MB parlance). This is pretty much what it sounds like. A clutch, just like in a manual car, except that it is controlled by computer and hydraulics. It does not engage when the car is stopped, just like you don't have a manual car in gear, and clutch engaged when stopped. So, if the computer cannot see wheel speed and brake system status, bad things happen. In short, if you decide to put a MCT transmission in your car, start with a 2005+ Crossfire, or plan on swapping the ESP. @Dave2302 is working on swapping a 7G into an R170, and his tests so far show that this may not be a problem for a "regular" 722.9, which uses a conventional torque converter, just like your stock Crossfire does.

About two weeks ago, we had a breakthrough (I say we, because there is a small team of absolute MB diehards with a ton of knowledge contributing to this), and the car began driving almost normally. A couple more breakthroughs later, and it can now be reported that there are about 200 driving miles on the 7G transmission, including highway and heavy stop and go traffic. It is a different experience driving the NAK equipped transmission. Somewhat like a manual, you can feel engagement as the brakes are released and the NAK engages. Kind of cool, actually, but not as smooth as a torque converter. However, once engaged, the NAK does not release unless there is something wrong, or the car is stopped again. Much more like a manual. The gear changes are very precise feeling. More confidence has to be built into the whole package, but it is truly beginning to come together!

A lot more to do. The 7G uses a different protocol to talk to the mode selected switch (W/S switch on the shifter), and work is ongoing for the electronic device that will translate that message. Once done, there will be four modes, Comfort, Sport, Sport+ and Manual. Yes, swapping in a different steering wheel with shift paddles is in the works, but that is likely not to happen until winter time.

There have been complications with the Killer Chiller. (DO NOT BUY ONE OF THESE! Get a system from Chemcool, FI, or just about anywhere else. There are plenty of vendors out there.) Still working through those challenges, but the solution is a known item that just requires time to get to it.

As the car is not 100%, it was decided to leave it home for the Fall Crossfire Rally, and instead I will ride up in @QuadPrism (William) SRT6, and just enjoy the event without worrying about any potential teething issues.

One last thing for now, in case you are wondering. There has not yet been a single full WOT run with the transmission. I am working up to the power very gradually, and need to be very measured in this, as I do not like broken parts! That said, the throttle has been rolled into a few times, and the feeling is very good. Much more direct and powerful feeling over the 722.6, if for no other reason than that the NAK is 100% engaged, like a manual. There is no lockup clutch turning off, and running the driveline off of the torque converter itself. Acceleration feels instantaneous, as do heavy decels when manually shifting the trans. Basically, the benefits of a manual, but no 3rd pedal.

 

Neil, I wanted to point out your extremely gifted in keeping your projects in simple terms (well explained) so that almost anyone can follow your progress. I enjoy your abilities and vigor in 'breaking new ground'. Thank you, for allowing the membership the opportunity to follow your progress (and fails). I continue to look forward to your posts regardless of where they travel! George (GG)


.

 

George, thank you for the kind words! I was essentially raised up as an engineer, and became one in life. One thing I was taught early on is that if you can explain your ideas clearly enough that the average person can understand and conceptualize what you are talking about, then you yourself have mastered those ideas. I also admit that at times during this blog I have been stuck on a particular technical item, and as I am typing these mini-novels out, it trips something, and I have a new way of looking at a problem I may not have even gotten to explain here. There are tons of details to a project like this, but most of those are boring, and no one wants to read about them. Heck, lots of people don't want to read what is already here! My hope in this is that I get what I have done out on 'paper' and can use it as a reference later on. I also hope it sparks an idea off for other people on things to do, but what I hope the most is that others can learn exactly what not to do, and therefore save time, money, and effort. Lord knows how many mistakes have been made, or wrong avenues have been traveled since the summer of 2021 when this began. However, I do not want people to be turned off from doing really cool things with their cars. The Crossfire offers the opportunity to harvest parts and ideas from so many MB and Chrysler vehicles, and actually have them work. In a lot of ways, the very basic electronics (compared to its MB contemporaries) make it a little bit of a blank canvas, and therefore a possibility in adding new and modern features. I definitely see 722.9 (7G) transmission swaps becoming the "in" thing for Crossfire mods within the next couple of years. What has made this particular one difficult is that I didn't just go for a basic 7G, but instead went for the gold on the MCT version (much more complicated). If that bar can be cleared, then the 7G swap will become altogether commonplace before we know it.

I see you are in SC. I hope we get to meet in person at the Fall Rally!

 

29 Sep 2023
Please read all the way to the end. I have been fighting a persistent drivability problem in this car for quite some time. It is related to what the E55 crowd will call the "hiccup" problem, where between 1000-2000 rpm, a temporary lean condition manifests, and the car stumbles a bit. All M113K cars seem to do this, but if you have a supercharger pulley change or a larger throttle body, then the problem is exacerbated. My E55 has a 168mm lower pulley, and it does this. It is an annoyance, but livable, as you learn it is just a quirk and the car is fine. This CF had this problem for the better part of two years. It was far more severe than the E55, but it was livable, also. Passengers always noticed it, and I would shrug my shoulders. More than one tuner took a stab at fixing it, but nothing seemed to work. Now enter the MCT transmission. This uses a clutch rather than a torque converter, which gives a more or less direct connection to the driveline, much like a manual transmission does. It has no fluid connection that can dampen these hiccups, like a regular automatic will do. In this case, the hiccup went from an annoyance, to a total pain in the rear! The car was borderline unmanageable on the street, especially in traffic. The stumble was a violent jerking in the driveline that made the car difficult to manage. The decision was made to leave the car home during the Crossfire Fall Rally 2023, mainly due to this!

Throughout the course of things, several gaskets had been replaced, another 82mm throttle body, another set of fuel injectors, and several other items were replaced. Tests were made with the supercharger belt taken off, to see if it was an engagement issue - it wasn't. Spark plugs changed, no effect. Fuel pressure was changed, no effect. Different ECU was tried, no effect. CAM and Crank sensor were replaced, no effect. RCM was changed (yeah, we were throwing the long ball!), no change. Air leak checks performed, no result. After my tuner made several, and by several I mean on the order of 30 tunes later, we figured we had one of three variables remaining:
1. Bad MAP sensor
2. Bad supercharger bypass valve
3. Air leak somewhere
Preliminary checks showed no issue with #3, that I could find, and #2 was checked in various other ways. #1 was more difficult. The MAP sensor on an M113K is notoriously hard to change in the car. I decided to take a shotgun approach and replace everything at once. My son came home from college to work on a few niggles on his SRT6 after the absolutely-incredible-and-you-better-not-miss-it-next-year Fall CF Rally, so of course I put him to work on my car. Last night we pulled the supercharger assembly off the car to replace all gaskets, a new MAP sensor, and a replacement bypass valve. The bypass valve is actually a throttle body from various MB four cylinder cars. I bought a nearly new one for a M111. We had the supercharger off in just over an hour (team work is awesome!) and found that the "donut" rubber seal on the bypass valve had not been seated, and was pushing itself out from one side severely. We could not tell if it was leaking, but if it wasn't yet it was going to before long. It had to be leaking under boost at the very least. From earlier diagnosis the bypass valve seemed to be OK, but not taking chances! Replaced it and all of the gaskets and a brand new MAP sensor from MB. Put it all back together.

Took the car out for a ride and sure enough, the hiccup problem was back to "E55 levels" and none of the severe jerking/bucking bronco business from earlier. I acquired a pressurized smoke tester from Amazon and will smoke test for leaks Saturday just to be sure.

It should be assumed that if you take on a modding project, or just own unique car like a CF, that your time making repairs is donated to the cause as part of the hobby. However, we sometimes will not fix a problem because we think it will cost too much, or take too much effort. Let me give a breakdown on costs for what was done last night:
- Used M111 throttle body, eBay, $30 shipped
- Gaskets, seals, new from MB: ~$50
- MAP sensor, new from MB: $51
Total: $131

My point here is that the cost of this fix was less than $150, and about 4 hours of time. Which leads me to the next point:
Problems like this can be difficult to trace, no doubt. As a long time tuner myself, I used to hate being used by the customer as a diagnostic tool, by writing tunes that I knew full well meant that the car had problems. The tuning process was a waste of my time and would not fix the root problem. The shoe is on the other foot in this case, and I owe my tuner a great deal of gratitude and an enormous apology. He stuck with me and kept working on tunes, as I was insistent there was no problem with the car. There were problems! Was it a single item that was replaced or a combination? Who knows? But the fact remains that a lot of valuable time was wasted by the tuner on problems that were not of his making, and that is downright abusive. I am publicly apologizing to him (he knows who he is). I also want this to stand as a monument to others that no matter how clean, no matter how perfect you think your car is if the tuner tells you there is a problem with the car, then you need to go back and dig deep. This problem was in the car for almost two years. Two years! It was fixed in one evening. Don't be me!

I want to thank @Dave2302 , @CL770 and @Deplore , who all three were telling me that they thought there were issues with the car that needed to be addressed, and what the fixes should be. I appreciate all of your advice! Thank you, guys, I'll move on advice much faster in the future!

 

30 Nov 2023

So I've been MIA on this topic for a couple of months. There's some valid reasons for that. We have been whittling down our rather excessive fleet of cars here, and that included a CF 6spd Coupe, an SRT6 coupe, my very last SRT4 (which I have owned since 2006), and a PTGT parts car. I am also in the process of prepping our 1962 Triumph TR3b to be sold. We're down to a much more reasonable fleet around here, and I am proud to say that everything here can run and drive on its own at the moment! This was not the case just a few weeks ago. A couple of weeks after the fall rally, testing of the 7G setup on the CF resumed. Progress was really moving along, and the car was really starting to come into its own. One day, a CEL came up with a code for a misfire on #7. It had been running fine. Clearing the code made the car run well, until it would come back. A slight miss at idle was noticed, but after that it was OK, until the light would come back. A few days of clearing the code, driving, light coming on, and clearing again ensued. The ME2.8(.1) in these cars will shut down a misfiring cylinder after x number of misses. Coil was changed, which is the typical failure for this code. Same results. Tried the wires, same thing. Pulled the plugs, they looked fine and clean, changed them, code came back. Checked fuel injector, and it was OK. uh oh. Break out the compression gauge, and check surrounding cylinders for reference. #5 and #6 yielded 170psi. #7 yielded 70psi. oh boy.

Three main possibilities. One, the head gasket is blown. Two, valve is burned. Three, piston is cracked. Good idea to take a peek inside and see what we have. An endoscope camera attached via USB-C to a cell phone went right in, but could not see the valves. Pulled it out, turned the engine to bottom dead center (BDC) on #7, and take another look. OK, now we can see the intake valves, and they are OK. Top of the piston looks wet (fuel?), but no visible damage. Cannot quite see the exhaust valve for some reason. Push more wire into the cylinder for the camera, no joy. Pull it out, let's try from the other plug hole. Can't. Camera is jammed! 90 minutes of pushing in, and tugging out, no luck. Push in one more time, give it a tug.... it's free! Camera comes out. Yay! Not so fast. Camera comes out, aluminum body of the camera dropped in the cylinder. UGHHH! Time to take a break. Get on Amazon, order a much better borescope, with an articulating 4.5mm end. Idea is to get a better look around, and then "spear" the camera body, turn the articulating arm, and pull it out. Two things on this: New borescope is 10x better than the cheap endoscope. Much, much clearer picture, and the controllable end piece made it easy to look to see that all three valves were intact. It also gave a clear picture that the wet stuff on the piston was oil, not fuel. Second, there was no hope on spearing that camera body. No matter. Oil is likely from past the rings, due to a cracked piston. Engine has to come out and come apart, anyway, so this part of the exercise is now over. Engine is blown. But what happened? More on that later. Out it came for now


Few things here. Those headers? Yeah, they were in the car for just about 600 miles. Notice that the ceramic coating looks about 20 years old and is peeling off. Not too happy about that one. Both headers were about the same. Engine oil came out black as ink, backing up the cracked piston theory. There were no leaks of any kind, any where, so the resealing did do its job, despite the overly short life.

Due to life commitments and a conscious decision to approach this without being angry at it (easy to do, right?), the timeline of this stretched out for two months. Fortunately, a 'spare' M113K had been acquired this past summer. It was an unknown quantity from a non-running 2003 S55, but the idea was to tear it down and have it sleeved and built up for abuse, so condition really did not matter. Now it does. Now it might be the engine savior for a bit. It was decided to treat this as the unknown quantity that it was, and therefore needed to be gone through. Engine was stripped down to the long block to be inspected and resealed. The treatment that the old engine received just this past summer of being resealed and normal wear items replaced happened to this one. The difference was that this engine was remarkably clean inside. Like gorgeous clean for a used engine! Very little varnish, and looking at the bores, no scratches could be observed.





There's those timing cover seals. This was supposedly an 80k mile engine. Looking at it, I agree it was. Those seals, however, are 20 years old. Be aware that *every* CF out there has these, and most of them probably look like this. Expect to start seeing posts about overheating CFs and water in the oil. Only a matter of time. It's a huge job to do in the car, but it is feasible, so plan on this as preventative maintenance on your CF coming up in the next couple of years. BTW, this engine did have trace amounts of the blue coolant in it. Every car I see with the blue coolant seems to have rubber issues like this. I run the (evil) GM Dexcool orange stuff in my project cars with excellent results, even if that makes MB purists mad. It was designed for aluminum and rubber. Back to the main story.

Timing cover seals replaced, front and rear main seals, oil pump chain, tensioner and spring replaced, oil pump check valve replaced, oil pans thoroughly cleaned, and then resealed. I have opinions on this last one, too. The upper pan is a machined flange to the block. It is not a place for the factory RTV. RTV is not appropriate here, even if MB/AMG used it. The appropriate stuff is anaerobic sealer. Yes, that red transparent stuff that virtually all other manufacturers use in these cases. No more leaks between the oil pan and block. I waivered back and forth on the timing cover, and finally used RTV there, but that was only due to the coolant and the head gasket areas. Otherwise, it, too, would have gotten the anaerobic sealant. The rear main seal cover is cast, not machined, so RTV there. New spark plugs, and then it was time to park the two engines next to each other to swap parts. Straightforward, but time consuming. No drama there until the very end. A mistake was made. Turning the engine over by hand, a sharp "pop" was heard. Turning more got a second "pop" until it was realized that the timing chain tensioner that screws into the side of the timing cover was forgotten. Rookie mistake! But I made it. Time to see how badly the cam timing got messed up. First, a new tensioner was threaded in to prevent any more skipping. Getting the tools out, and dialing the crank to 40° ATDC, per the manual, and then the cam timing tools are slid into place. If all is well, the tools will slide into the slots on the cams, and lay more or less flat against the cylinder heads (different tool for each bank):


OK! Finally some good news! The sprockets did not jump time, but rather the slack in the chain was biding up on the loose timing tensioner.

Before reinstalling, one last item in the car to tend to, and that was the steering idler bushing. There was significant play in it, so it was replaced, and geometry rechecked (it was off, too!). Generic steering damper was replaced with a Bilstein. Power steering hoses were pre-emptively replaced, also.




Will move into the next post, as we are out of room for pics in this post!

 

30 Nov 2023 (con't)
Engine bay was cleaned up (really wasn't bad - only went 600 miles since last cleaning!


And reinstall the engine:



Suffice it to say that it is all back together and running. So what happened? Well, that is often a process of elimination. The tune was the same as what had been running, so probably not that. Next will be a look at fuel delivery. Fuel pressure is monitored, and it was ok, so on to the fuel injectors. The logical thing to do here is to have the fuel injectors flow tested, and cleaned, if necessary. These are very low mileage (<10k miles) Bosch 630cc/min injectors. However, failures happen, dirt happens. A few phone calls around revealed that testing and cleaning could be done for $38 PER INJECTOR! $38 x 8 injectors = $304! Time to consult the AliExpress app! For a grand total of $354, I had an Autel fuel injector test and cleaning machine and 1 gallon of cleaning fluid at my door step! Left over from a previous injector flow project were a few gallons of n-Heptane, which is the fluid that the OEMs use to flow test injectors. All set then, so flow test the injectors.

Perfect machine? Well, no. Adequate? More than enough. The injectors were all very even, and if they were not, there is an ultrasonic bath on the backside of the machine to help clean them out. For roughly $50 more than a one time flow test and cleaning (which wasn't even needed in this case), the machine now lives in our shop. Every time injectors come off of something, they will run through this machine, now!

So, it wasn't the injectors, what could it have been. The answer is that it was not a single thing. The first known issue was the intercooler circuit. Remember back in an earlier post it was mentioned that there were issues with the Killer Chiller (KC) ? The issue was that the KCwas over-taxing the entire A/C system. In the summer heat, neither the intercooler water, nor the HVAC were getting cold, and a ton of heat was being dumped into the radiator (via the condenser). The root cause was the TX Valve (TXV). This had to be replaced once, but Kincaid Performance (makers of the KC) made an odd choice. Normally in a refrigeration circuit, the TXV sensing portion is tied to the suction side of the A/C core. This is the large hose that gets frosty cold in an HVAC system. The idea is that this get cold, and the TXV starts closing a bit to moderate the refrigerant flow. Kincaid elected to wrap the sensing bulb around the water outlet. In essence, they were looking for ice cold water to exit the core, and until that happened, the TXV would be wide open. This resulted in a huge pressure drop, and no hope for the regular HVAC to get enough flow to work, which is exactly what happened. Unfortunately, the KC core was tucked away on this car, and difficult to access, so in the short run, I cheated. The fuse for the water pumps was pulled to stop the water flow, and allow the HVAC a chance to work. This results in no flow through the intercooler, and instantly hot IATs. Yes, the SC will eventually disengage, and the knock sensor will reduce timing if it knocks. However, this is not a "feed forward" system, and it only reacts to problems. In other words, the ME2.8(.1) only reacts AFTER the engine knocks, not before. This is why I refuse to put low octane fuel, even in the N/A CFs. In this case the engine was likely knocking, and then the knock sensor would reduce timing. As I left the car in this state for a long time, plenty of knocks occurred. 100% self-induced problem that should have been addressed. During the rebuild, the TXV was moved to the suction hose of the KC core. Now there is cold water and the HVAC works. Amazing! Don't be like me, and leave a problem like this. Also, NEVER BUY A KILLER CHILLER! Go to Forced Induction or ChemCool.

Remember when I said it was not a single thing? When the tools came out to check the cam timing on the 'new' engine, it seemed like a good idea to just check the old engine. First, set the crank to 40° ATDC:


Put the cam tool into bank 1 (passenger side). Not too bad, maybe some chain stretch?


Now to Bank 2 (driver side):

Uh, Houston, we have a problem. Your eyes do not deceive you. The tool most definitely is not sitting flat. That cam is advanced 1 tooth! Yes, this is the bank that contains the failed piston.

So there you have it. Cam timing off, high IATs, boom! So now what? Well, we do have good cam timing and good IATs. Testing of the electronics has just resumed on the 7G, with a couple more good usability breakthroughs. Weather has been less than ideal for this area, so only a little driving has occurred so far, but things are moving forward in a very measured way. The new engine is a LOT more rev happy than the old one ever was (go figure!), but no power pulls have been made. Right now, driveability is the main focus, getting the 7G MCT to behave on this engine with this computer system is not a plug and play deal. It is taking a lot of programming work. The good news is that regular 7G is far less complicated, and also that when the code is cracked on this setup, it will be easily reproducible in other cars, and it will also open the door to other exciting things. Right now, we are back on track from where we were in September, but just in a much more healthy way.

My parting thoughts are these: The cam timing was an issue, no doubt. Odds are that this issue existed from before I put the engine in this car the first time. I never touched the timing, and I had no issues with it (other than a tuner telling me something was off, which he was right). However, I 100% KNEW that the coolant was NOT flowing through that intercooler. I 100% KNEW it was a problem waiting to happen. I was 100% bitten by my OWN negligence. The worst part? It took 30 minutes to correct that problem with the KC. Swapping the engine took well in excess of 80 hours (admittedly being OCD on details).

In the end, it turned out OK, but only because I had a spare M113K that I intended to tear down to build up. That was *the* happy accident in this whole thing. Now the 'old' engine will serve in the rebuild capacity, and effectively I am just out the time and a set of engine swap materials (fluids, seals, gaskets, etc). Had I not been prepared with a spare engine, this could have been much, much worse. Don't let things that you should take care of pile up and then become an expensive bill, not to mention major headaches!

Onwards, charge!

 

Wonder why that particular number piston went, it's not uncommon for just one to fail in other engines but still, maybe that cylinder gets the most air. Maybe get the manifold flowed. And just in case you don't know, others put off doing something just because and then self administer kicks when it fails.

 

WHy did #7 fail first? Unsure. Maybe the piston was the weakest? Maybe it ran just a touch hotter and knocked the most? Who knows? I'll treat the remaining 7 like they are about to go, and they will all wind up in the trash, tbh. These engine cost too much to cut corners on rebuilds.

 

07 January 2024
First post of the new year! Work has been ongoing for the last month or so, but more so on fixing little issues with the car that have been annoyances since June of last year, and then work has focused on electronics. Being a nerdy engineer it is always interesting to me to pack plenty of modern tech into every car build. Some previous builds saw various options that did not exist when the cars were made being integrated in. Most got some sort of custom electronics that would add new functions, or enable newer devices to work in the car. This would range from modern OEM radios, to cruise control, to entire engine management systems. For instance, while it is an "old" concept now, I have a Megasquirt running my 1962 Triumph TR-3B, but it wasn't so old when it was first installed in 2007. (Yes it still works, and in fact the Triumph was driven today) So now the Crossfire is getting a similar treatment. As you are reading this thread, you know it has a M113K V8, but if you didn't catch the last few updates, you might have missed that it now has a 7-speed automatic transmission from a 2014 SL63. It took some crafting of electronics to make that work. But now it is time to add some more. As noted much earlier in the thread, the plan is to add a digital dashboard system, but before that can happen a couple of other projects must be accomplished to lead up to it. First, electronics need to be added that will gather data from many sources and digitize it. The Crossfire is still an old school car in many ways. When you turn that headlamp switch or turn signal lever, you are moving a switch that is directly manipulating the device you are wanting to use. Turning the headlamp switch directly switches the headlamps on and off. The same with the turn signal lever, and even the ignition switch. In modern cars when you do the same, a message is placed onto the vehicle's CAN bus network, and that message is received by various modules that will then trip a relay or other device to turn on and off a lamp, fan or move a motorized door somewhere. To bring the Crossfire into the next decade or two of automotive development, some of these functions need to be brought on to the CAN bus, or at least set a status of each lamp on the bus. Fortunately, the board I designed and showed earlier (below) is equipped to do that.

This board was designed and built a couple of years ago. It is mostly outdated now, but it still works. (There is already a more modern, easier to use and tailored PCB on the drawing board) By connecting to the wiring in the car for each of the lamps and switches, the status of those can be digitized and added to messages that newer devices will understand. At the same time, we can add newer devices to the car, and read their data from the CAN bus and turn those into usable functions. Enter the second item to accomplish: a modern steering wheel with buttons for the radio, dashboard and paddle shifters, just like a newer car! Several steering wheels were purchased and tested on the bench. In hand now is one from a C63 (W204), E300 (W212), E63 (W213), and another C63 (W205). The silver of the W204 wheel matches the CF the closest, but it has red stitching and it will clash with my black and grey interior (besides, my son claimed it for his SRT6 that he swapped a cedar red interior in to). So, the E63 (W213) wheel it is!

CATUION: Before proceeding with this story, you must understand that a steering wheel is a safety item, and it contains another safety item, the driver's side airbag. These are engineered and crafted for each specific vehicle, and then tested relentlessly to ensure the highest level of safety possible. Changing these is inherently unsafe for most situations. In other words, it's risky. If you read this, and decide to follow anything you see in this, do a lot of research, and then anything you do, you do it at your own risk. No pics or step-by-step instructions of this process are included as a result.

The 2003-2006 Crossfire uses a single stage airbag, and the 2007-08 Crossfire uses a more modern dual stage airbag. Pretty much everything made after 2007 features this dual stage airbag, including all of the steering wheels listed above. The subject Crossfire is a 2004 - single stage airbag. Eventually this car will to go to the later Crossfire dual-stage setup, but for now, it is a 2004 single stage system. Next, the Crossfire uses a clockspring mechanism with four wires in it. Two for the airbag, one for ground, and one for the horn. That's it. The W213 steering wheel has four wires for the airbag, and three wires for the button controls. Plainly, something would have to be changed to add more wires. In this case a W203 (C-class) clockspring was used, as it has four wires for the airbag and four wires for everything else the W203 had in the steering wheel. It is very close to the Crossfire/R170 clockspring, but it has slightly different mounting, it is thicker and one of the connectors is different. If you get one of these, make sure you get the connectors that run down the steering column to the car. The turn signal switch has a bracket to it that holds the electrical connector, and this has to be modified to allow the W203 connector to fit. Once that is on, you'll find that the splines of the new wheel are the same but the outer diameter of the shaft is too large to fit the clockspring, so the wheel itself will have to be turned on a lathe to get that to fit. Then you will see that the spline section is about 10mm longer than the Crossfire, so you will need a longer M16x1.5 bolt to hold the wheel on. You will also have a nearly 3/4" gap between the wheel and the steering column that is ugly as sin. My son @QuadPrism , a mechanical engineering student and wanting the W204 wheel in his car, designed a sleeve that was 3D printed, and is simply glued to the steering column plastic, covering this gap up nicely. So that solves the mechanical aspects of the wheel. This is what it looks like installed:



So it looks quite good in there, but it feels absolutely perfect in the hand! The thumb cutout areas of newer steering wheels really do a lot for positive feel and control. That alone is worth it. The yellow lights on the A-pillars are for blind spot protection. The system works quite well, but unfortunately will not work with the Crossfire out of the box, as the Crossfire does not put needed data onto the CAN bus to make it work. The system itself is an off the shelf device, but designed for a more modern car. It will light up one of the yellow LEDs when someone is in your blind spot, and if your turn signal is on and someone is in the blind spot, it will beep at you. Also, it will sense traffic and people walking behind you when you are backing up, again beeping at you. Very nice setup! Available here: Unfortunately, it has gone up in price pretty significantly since I purchased it. Also, it will not work with the Crossfire as-is, because the Crossfire does not put data it needs onto the CAN bus. I hope to offer some electronic devices in the future that will enable everyone to add more modern tech to their Crossfires (and other MB products) at reasonable prices and low installation complexity. If you have a newer vehicle that needs blind spot protection, this system is pretty good.

I have a whole M113K V8 Crossfire with a 7g MCT transmission mocked up on a bench to develop these devices out. Every control module is wired up, programmed, and all "think" they are in a car. This speeds up development and testing tremendously! Once the device for this function was working on the bench, it was moved out to the car for testing. Here is a video showing operation of the paddles, directly after getting it to work:

A road test happened shortly after, and it worked perfectly! Now back to the red circuit board above to enhance the functionality, and bring all the various devices that are deployed in the car into a single unit that does it all, and is tightly integrated. More on that to come in a bit. Thanks for reading!

 

"it now has a 7-speed automatic transmission from a 2014 SL63. It took some crafting of electronics to make that work." Is this crafting of electronics unit going to be made available?
Flipping a switch "a message is placed onto the vehicle's CAN bus network, and that message is received by various modules that will then trip a relay or other device to turn on and off a lamp, fan or move a motorized door somewhere." Ah, that's how the magic happens.

 

Possibly. It will be a while, as I keep finding new requirements and desired features as I go. I also need long term testing time for reliability.

 

.


What a post! Pleasure to read and see what is possible! Thanks for your curiosity, your ability to want to, & to share with us!


.

 

13 August 2024
Change of some plans, so editing this post. Have a few maintenance items to address on the M113K Crossfire in the next couple of weeks, and I will post up about them here, soon. This car is currently scheduled to go to the Crossfire Fall Rally 2024 in Asheville, NC. Have some plans for this car for over the winter. Some of what will be done will be able to be easily duplicated by other members, and the goal is to make some of those changes available.

In addition, there is another V8 Crossfire in the works, and I will start a separate thread on that car when work begins on the conversion. Right now, that 2008 Machine Gray coupe is keeping me ATMOTBS (At The Mercy of the Body Shop - old Mopar Mailing List term, for the ones who have been online the longest!), but should come back looking stunning! Once here, a new thread will pop up documenting that one from start until, .... well, whatever happens!