Both the front subframe/engine and rear subframe/differential are installed. Work is now focusing on the smaller items on the to-do list. This portion is all detail-oriented and deservedly requires the most time due to its inherent tediousness.
Here is a snippet of some of the progress that has been recently made, as this swap creeps towards the finish line.
Marcus cleaned up the thermostat housing by welding up and blocking the heater coolant nipples. After consideration, I decided to obviate the heater lines and plan to eventually remove the heater core entirely from under the dash. With the air conditioning removed already, this was the next progressive step. Since my RX-7 is principally a fair weather car and I live in California, I foresee no requirement for a heater. Now there is no need to worry about the heater core leaking.
AN fittings were welded onto the valve covers. This is part of a sub-project to improve crankcase ventilation. The PCV will be obviated and an externally mounted vented catch can will be routed in.
Keeping in theme with the generous reliance on AN fittings, the crossover tube at the front of the engine was optimized as well. Before, there was a rubber line that was awkwardly routed in a loop (see the penultimate picture in this post: radiator overview).
The location of the proportioning valve had to be revised due to clearance issues with the LS3. New lines were bent and the valve was positioned further upwards.
Lastly, here’s an overview of the Samberg radiator setup in the engine bay.
Getting close! All the major parts are in, like this custom length aluminum 3.5″ driveshaft from none other than The Driveshaft Shop.
We had to place priority on getting the front and rear subframes prepared and ready for mock up in order to take the measurement for the driveshaft. I then placed the order immediately with Driveshaft Shop because the production is supposed to take 3 weeks. Conveniently, they actually finished it for me in a little over a week. Like all the other top-notch offerings from Driveshaft Shop, this one is no exception. It features a bolt-on CV joint adapter to the TR6060′s flange and then a universal joint at the 8.8 differential’s end. Incorporating a CV joint is more expensive, but features a higher range of adjustment with lower vibrations.
The fuel lines were another item on the list that needed to be measured up and then custom ordered. There will be a total of 3 “flex” lines: 2 from the gas tank to the Wix fuel filter/FPR and 1 from the fuel rail to the firewall bulkhead. The rest of the fuel system will be plumbed with a -6AN stainless steel hardline. I really wanted to overkill this area to eliminate any chance of fuel issues in the future, including fumes and having to smell gas in the garage.
I picked the XRP HS-79 hose for the lines, which is aerospace quality and incorporates smooth-bore Teflon with a Kevlar composite construction. Unlike other hoses which can be assembled at home, these have to be machine crimped. Marcus made the measurements and I gave them to XRP to pre-make. The quality of the HS-79 is certainly reflected in its pricing. This unassuming box of 3 lines was close to $400!
Here’s a look at the Wix filter with the stainless steel hardline routed.
Marcus made this bracket to hold the bulkhead connector for the long section of the hardline. Attention to detail – epitomized.
At the other end on the engine bay’s firewall, there is another bracket that holds together the fuel hardline and hardline for the rear brakes.
The goal is to remove redundant wires and relays from the engine bay to have a clean, tucked install. A great deal of effort is obviously required for this portion. Marcus has been working on the wiring harness over the past week. Any splices made were done properly via soldering and heat shrinking.
It was also determined that the spot-welded brackets which used to hold the stock radiator’s crossmember needed to be removed from the frame rails.
Alas, here is the engine mounted for the final time in the car.
The view from below… the Samberg components are truly superior to the factory Mazda counterparts.
The Spoolin headers have a good amount of clearance on both the passenger and driver sides.
While work is continued on the car, here is a mid-week update to tide things over. When ordering the MGW Short Shifter, I opted for a threaded lever in a M12x1.75 rather than the regular slotted style lever for Camaros. The stock Mazda threading is a finer M10x1.25. The different lever threading meant I had to ditch the FEED shift knob, unfortunately. On the bright side, this enabled me to try a new direction and go with a product that wasn’t quite off the shelf… I looked around at various custom alternatives from companies and ended up shooting an email to Chris at WC Lathe Werks to see what he could do for me.
After a few back and forth exchanges, I was able to finalize the exact shift knob that I visualized with him:
- 2″ diameter sphere
- 2.75″x1.2″ collar
- Blended transition
- 2 grip grooves
- Blue/Bronze coloring
I have always liked spherical shift knobs. With this being a hand made affair, I wanted to take the opportunity to elevate the design aspects. I decided on adding a taller collar to the shift knob to increase leverage and reachability. The quality and machine work on this shift knob is very high. I can’t wait to throw it on and start rowing through gears. I would personally rank a shift knob equal to a steering wheel in terms of “touch factor” and driver input importance.
The hectic pacing of the project has slowed down going into the third and fourth weeks. Much of the heavy lifting has been accomplished and the focus has turned to the intermediary steps and other details. The engine and rear end were mocked up onto the car so measurements for the new driveshaft could be made. Marcus also figured out what fuel lines were needed and gave me the information to order with XRP. More dramatic progress should be made once the driveshaft and fuel lines have arrived.
Here is what the rear end looks like with the 8.8 differential and axles fitted to the Samberg mounting brace.
I was lucky to be able to pick up this 8.8 Cobra (’02-’03) differential locally, because they are becoming harder to come by. I dropped the differential off to Samberg and asked him to do a full clean and rebuild so all the seals and bearings are fresh. A new set of Ford Racing 3.73 Gears were installed as well. These are supposedly the only gears that do not whine. If I had stuck with the stock FD rear differential, not only would I have ended up with a weaker unit, but re-gearing it would have been difficult due to almost nonexistent alternatives. Pairing up the power and torque of the LS3 with the gear ratios of the Camaro TR6060, the FD’s 4.10 ratio would have certainly been on the higher side.
The new front subframe was assembled and ready for install.
The cool thing about having the entire front subframe out of the car, everything can go back in from underneath rather than over the top. The chassis is simply dropped on top of the subframe via the lift.
A test fit… this picture aroused a couple knee shakes in me.
The front assembly with the long tube headers mounted:
And to bring closure to this post, I leave it with an eye candy of a suspended rear Ohlins DFV coilover.
With the rear subframe dismounted from the car to fit the new 8.8 components, Marcus took the time to weld on reinforcement plates. These were included with the Samberg kit but required additional refinement for a more perfect install.
The original plan was to use rubber/polyethylene differential mount bushings, but in the end they were discarded and a new set of bushings was lathed out of Delrin. Rubber bushings have been reported to fail under extended use with the 8.8 rearend. Using Delrin should offer a more robust solution with an indefinite lifespan.
As another showcase of detail, the Mazda fuel filter mount was offered up to the Wix filter/FPR. The ground strap from the Wix mount was tig welded on to adapt over the Mazda mount. This allows for a clean and OEM-like solution. Granted the OEM mounting location is rather high on the subframe, we are hoping accessibility can still be maintained enough to not warrant a relocation.
Samberg makes a nice, water-jetted pedal mount for the 5th Generation Camaro’s fly-by-wire pedal. The mount was painted a gloss black.
Moving up front, work was completed on the LS3 itself. The Improved Racing baffle was installed into the F-Body oil pan.
At the last minute, I was able to pick up Improved Racing’s newly released crank scraper which fits LS3 engines. This part was fresh off the manufacturer’s tooling and I had it immediately Priority mailed to the shop. As its name suggests, this plate is sandwiched between the main caps and windage tray and closely contours the crankshaft and rods – literally scraping off excess oil. The crank scraper will aid in oil control and ensures that more oil makes it to the bottom of the sump for recirculation. Having oil remain on the crank, where it is not needed, creates parasitic drag and windage losses.
With the windage tray and pick up tube installed:
The crank scraper should be especially helpful with the F-Body oil pan because by design of its shape, oil control is minimized for the front 2 cylinders while the crank scraper can each out to them. This can be seen from the above picture – where the windage tray is only able to cover a 3/4 length.
With no power steering or air conditioning, the serpentine system was greatly simplified, however, a modified idler pulley was necessary to fill in the gap. Marcus machined a spacer to push a ribbed idler pulley into the right track.
Here is what the pulley system (including the Katech tensioner pulley) looks like with a belt:
I had the mounting components from Samberg powdercoated in Red, taking inspiration from Ducati trellis frames. If I’m going to have the entire front subframe changed out, I want it to be recognizable with a bit of flash appeal.
Marcus also redid and straightened the Wilwood clutch master cylinder pushrod, which Hinson modified to be crooked.
The Samberg kit included new eccentric camber plates that needed to be welded onto the bolts, which were fully tig’d on.
The larger plates should allow for an increased range of adjustment.
Instead of dealing with crating and shipping the 13B-REW, I did the next best thing – trade it locally for a set of Ohlins DFV coilovers. I worked the trade with Heath from Sakebomb Garage, and he dropped off the Ohlins when he came to the shop to pick up the motor.
The Swedes know suspension…
The swap is steaming along and Marcus has made a lot of headway within the first week. I asked him to take pictures for me as he went, so I could update about the project here. So here’s a rundown of the work completed thus far…
Pulling off the powerplant frame (PPF) and rear differential. I’m not a fan of the design and construction of Mazda’s PPF, it’s an inferior way to support the drivetrain and is prone to cracking under stress. The new Samberg mounts should be vastly more robust.
I will be upgrading the rear differential to the 8.8 from a Ford Cobra (’03-’04). I want the drivetrain to be bulletproof so I can hammer on it with complete disregard. It would only have been a short matter of time before the stock FD differential blows up from the increased torque levels. The 8.8 differential is BEEFY in comparison. It will add weight to the car, but at least it’s in the rear and should help balance out the also heavier LS3 up front. Here’s a picture of the 8.8 cover, which was cut and rewelded for clearance.
Working on assembling the new axles:
To support the increased power levels, I went with Driveshaft Shop upgraded axles with bolt-on 8.8 differential stubs. The center bars on the stock axles are made from a relatively brittle and lower grade material. These Driveshaft Shop center bars are made from heat-treated, double temper aerospace grade torsional steel with rolled splines. Fancy indeed.
With the rearend apart and the axles removed, it was a good time to pull off the hubs and install extended ARP studs. Unlike the fronts, the studs are only accessible with the hubs removed. And once the hubs are removed, the bearing is destroyed in the process and needs to be replaced. There will be a new pair of Koyo bearings going in.
A comparison of the new Bosch 044 assembly versus the old Supra Denso, stock assembly.
Since the Hyperion surge tank cover was designed for a stock-sized fuel pump, the larger Bosch 044 unit wouldn’t fit into its opening. Marcus had to do a little clearance work.
Removing the old hard lines and charcoal canister:
Dropping the fuel tank:
I was explicit on maintaining a high degree of attention to detail throughout the swap, and Marcus is certainly coming through in spades. For example, here are a few unsightly holes that were filled in and cleaned up:
And with the A/C removed, the liberty was taken to properly plug up the opening for the condenser lines with a plate and nutserts.
I asked Marcus to cover the engine bay abundantly with gold.
I will be ditching the old ABS Delete setup entirely in favor for double-flared hard lines with steel fittings. The hard lines will not only clean up the routing substantially, but offers functional gains over using excessive lengths of flex lines. The brake master cylinder pictured below is from a Mazda 929 and offers a larger 1″ bore, versus the stock 15/16″. The larger bore should help compensate for the increased pedal movement required to move the increased capacity of fluid in the calipers of the Brembo Big Brake Kit.
Due to the magnitude of the swap, I decided I would leave the car to a shop. I take pride in doing most of my own work. I have a far greater degree of control and attention to detail when I’m doing everything myself. However, I simply do not have the bandwidth right now in life to undertake a massive engine and drivetrain project. I also do not have the facilities to complete the swap in my home garage in a timely fashion. Given the circumstances, I think relinquishing the car to a professional is the smart choice.
Finding a shop I could be comfortable with was crucial. I had to use deliberation and careful examination of the options available before settling. From my experiences, many automotive shops are subpar: busy, unorganized, filled with other customer cars, and with heavy emphasis on getting the job done with low refinement. There was no way I could trust leaving my car for a length of time at one of these shops. The higher end shops were either also too busy or wanted to charge exorbitant fees.
After calling around, I was given a referral to a fabricator and mechanic named Marcus Fry, who branched off and started his own shop a year ago. After my initial conversation with him, I felt like he was a good match to hand off the project to and met my criteria: small, owner operated shop and was willing to dedicate the lift exclusively to my car until the swap was complete. I didn’t like the idea of having my car moved inside and outside while work was done simultaneously on other clients. Marcus was very accommodating and had all the right skills – there are several fabrication work I will need which he can do. I scheduled in the date with him to start the project and he came by to my house on July 6th to transport the LS3/TR6060. It’s interesting that almost exactly 2 years ago was when I first received the car.
I then drove my car to his shop, Marcus Fry Racing Enterprises, and the swap was in full motion.
Marcus allowed me to lend a hand to strip down the engine bay, which I thought was cool. I never worked on my own car in someone else’s shop before. Within a few hours, we were able to undo what was essentially 6 months of my work during the single turbo conversion.
The experience was slightly bittersweet seeing all my shiny, fresh single turbo parts being systematically ripped out. All I had to do was visualize the car with the V8 swap completed, and all notions of sadness were immediately wiped away.
I collected the parts we removed and stepped out of the picture, leaving the rest of the project entirely to Marcus’ capable hands. When I stopped by the next day to drop off a few more parts, I saw that the engine bay was bare.
Seeing the rotary powerplant outside of the car gives relativity on how truly small it is. The engine and transmission looked frail and undersized in juxtaposition to the far beefier LS3 and TR6060.
Marcus also cleared out a bench near the back of the shop and organized all my swap parts there. This V8 swap is almost like rebuilding and replacing half of the car… There is another bench not pictured filled with the 8.8 differential components.
To prioritize making sure the wiring for the swap is done cleanly, I purchased a standalone LS3 harness from PSIConversion. This should help consolidate the fuse boxes and allow for a simpler install. The plan is to obviate redundant fuses and relays from the chassis harness and tuck the rest.
The provided convoluted tubing was too basic in aesthetics, so I picked up a set of Painless PowerBraid to rewrap the harness. However, the process of removing the old tubing, fitting in the PowerBraid, and redoing all the joints was definitely NOT painless. The task was so tedious it took me more than 4 hours to complete. Was it worth it? I’m not sure, but I don’t want to do it again.
Complimenting the in-tank Bosch 044 assembly from Overview IV, I have a Hyperion surge tank cover. In the RX-7′s fuel tank, there is a small, plastic “tank” that the fuel pump hangs over. The downside is this tank is 4 walls with no top. Being uncovered allows fuel to slosh over during higher G turns and creates a starvation issue. The Hyperion cover will help contain the fuel inside the tank and minimize spillage.
For the shifter, I will be ditching the stock Camaro unit in favor of this MGW short shifter.
This is the epitome of fine machine work, almost all of it is from billet aluminum. The linkages may need to be shortened and re-welded accordingly in order to center the shift knob in the tunnel’s opening.
I used a fair amount of reflective gold heat shielding during the Single Turbo conversion and I plan on continuing that theme for the swap. I also picked up DEI’s Floor & Tunnel Shield II to replace the stock sheet metal heat shields around the transmission tunnel. It has a glass fiber core with an embossed aluminum face bonded to it, which looks very similar to the stuff car manufacturers like BMW are using in their newer cars. There’s another larger box/roll of the tunnel shield that is not pictured.
I’m a sucker for a good idler pulley. They are typically an understated part, yet serve a rather important purpose. Evidently, LSx engines have a bad spot in the RPM range where the serpentine belt will want to jump off the pulley. Katech’s billet belt tensioner is supposed to remedy this issue and will be a good insurance to have. Plus, it looks nice.
Lastly, to round out this mix, here’s a picture of the Dakota Digital SGI-5 speedometer signal interface unit. This little module is necessary for V8 swaps because the FD’s speedometer requires a 8202 pulse per mile signal, whereas the T56 and TR6060 send out a different rate. The Dakota Digital unit will convert the pulse frequency at a set ratio to calibrate the speedometer accurately. It’s also nice that the range of adjustments is almost infinite, so virtually all combinations of gear ratios and tire sizes can be accommodated for.
The stock FD clutch master cylinder has too small of a bore and improper stroke to actuate a T56/TR6060′s clutch, which is where the Wilwood 7/8″ unit comes into the equation. I bought a “modified” Wilwood from Hinson Supercars which converts the threading on the master cylinder to adapt to the stock clutch pedal. And by modified, I mean they just welded on a bolt with the correct thread pattern to the end, which is quite chintzy to say the least. I don’t think the bolt is even welded on straight, to be honest.
I hope to have it redone cleaner… and straighter, before installation.
For the clutch line, I plan to use a -4AN stainless braided line that will hook up to the clutch master cylinder with banjo fitting adapters. On the transmission/slave cylinder side, I had to dig deeper to find the proper adapter for the TR6060′s different quick-disconnect plug. A call to McLeod‘s helpful tech department yielded this fitting as the correct adapter:
For the fuel system, I plan on running a -6AN stainless hardline from the tank to feed the stock fuel rail on the engine. I will not be tampering with the injectors as there is no need to. I also do not want to run an aftermarket fuel pressure regulator, but instead opted for a simpler solution in the form of a Wix 33737 fuel filter. This is an OEM replacement filter that has a built-in pressure regulator to hold the pressure at 58 lbs – easy with no frills.
The Russell fittings shown are used to adapt the filter to -6AN. For the quick-disconnect style fittings on the bottom right, I specifically chose to use Russell’s newer threaded collar design. Their older versions used a plastic retainer and a different locking mechanism that was prone to leaking. After reading about instances of engine bay fires from these older Russell fittings, it left me feeling wary. So for the engine bay side, to adapt the stock 3/8″ fuel rail quick-disconnect to -6AN, I decided to go with a Swagelok compression fitting instead. Swagelok stuff is rated to extreme specifications and are generally uprated for an automotive application, which is exactly the kind of overkill I wanted.
Since the fuel filter will be mounted under the chassis at the rear, any sort of potential leak at that location is mostly uneventful. The engine fuel rail, however, hovers directly over the driver-side header. A leak at that point is a recipe for disaster, which explains my extended scrutiny as an insurance.
My Supra Nippondenso pump currently in the car is sufficient to supply the LS3, but I will be upgrading it to an in-tank Bosch 044 assembly regardless. I had this setup made by Chips Motorsports during the beginning of the year, as preparation for what was to be my next single turbo iteration… which should now work nicely with my LS3 swap.
The hanger was sent to plating and galvanized and I had a pair of -6AN elbows welded to the cover for the feed and return lines.