Ugh, this makes me want to go out and buy a welder and a band saw...
Ugh, this makes me want to go out and buy a welder and a band saw...
David S. Wallens said:Wow! And thanks, John, for sharing your latest project with us. It's perfect for this crew.
Thanks David, I'm just appreciative that there is interest in this community. I've seen some really cool stuff built here and in the magazine and thought, "I could never do that."
These projects are always a few steps forward followed by a step back. That kinda happened this weekend. I ended up not liking the angle of the shifter so I cut it and had a buddy with a TIG glue it back to have just a slight rearward bend.
Then I ended up having to modify the gearbox crossmember. The driveline angles didn't end up quite right and I wanted to tilt the rear of the box down a bit. Plus I decided the one I made was a little on the flimsy side, so I remade just the riser portion, then boxed it in at the front and put in some support braces in the back that still give access to the bolts. This should be plenty strong.
I believe that the RX8 is one of the most underrated cars out there. So much cheaper than the "short version" and readily available with a pre-blown rotary to facilitate fun swaps. Seriously, if you are reading this and at all intrigued look at the mechanical damage rx8s on Copart. Love this build.
I'd never ask to kiss and tell, but, explodey bits aside, what do you have tied up in the project as far as dollars? Because enginless RXs and 2.5 durka-techs are cheap.
In reply to Appleseed :
Well until I ponied up for the G Force it was pretty reasonable. The gearbox alone was $5k, but I bought it from a professional who rebuilt the entire thing. The clutch discs were $800 on top of that. Lesson learned, stick with the GM input shaft next time vs the NASCAR input shaft. :(
Total now is closing in on $40k. I had it together and running last year for around $30k. I don't include tires in that because I have stacks of them from my DSP days. The car itself was $2k and I recouped more than the cost of the car selling various parts off of it. I spent a decent amount on a wrecked NC Miata as well. Didn't really need to do that in retrospect but it gave me a factory engine wiring harness and what Mazda calls the "Front" harness, plus a bunch of other bits that are needed (steering rack being one of them).
I've thought more than once about building a more tame copy of this car for a street car. I really like the RX-8 platform, I think it drives really well and has enough space to be a usable daily driver. Without all the expensive racey stuff like the Penskes, seats, intake manifold, fancy front bar, etc but still keeping the turbo setup (because turbo) I think it would be pretty easy to get one on the road for under $20k, and maintain emissions legality. Could be done even more cheaply if you don't want the turbo right away (the kit is $6k by itself), or the Ford diff conversion. The engines are inexpensive and plentiful, but the bellhousing to use the FD transmission is pretty expensive. It all adds up.
johnvitamvas said:In reply to Appleseed :
I've thought more than once about building a more tame copy of this car for a street car. I really like the RX-8 platform, I think it drives really well and has enough space to be a usable daily driver. Without all the expensive racey stuff like the Penskes, seats, intake manifold, fancy front bar, etc but still keeping the turbo setup (because turbo) I think it would be pretty easy to get one on the road for under $20k, and maintain emissions legality. Could be done even more cheaply if you don't want the turbo right away (the kit is $6k by itself), or the Ford diff conversion. The engines are inexpensive and plentiful, but the bellhousing to use the FD transmission is pretty expensive. It all adds up.
I read somewhere that it is possible to use an NC Miata transmission. You just have to swap the tail shaft housing from a series 2 RX-8 transmission onto the Miata trans. This allows the ability to retain the RX-8 PPF. I have no idea if this is accurate but if it is, I would imagine it would be a considerable cost savings. Retaining the RX-8 PPF may also mean serious firewall hacking.
The NC Transmission looks like a good option. The problem is it is about six to eight inches too short (bellhousing to shifter). Either the shifter ends up in the middle of the RX-8's dashboard or you have to cut the firewall and heater core out of the RX-8. I suppose someone creative could make a remote shifter to alleviate this problem but you'd end up having to relocate the heat/AC/radio controls and build a custom center portion of the dashboard to accommodate the linkage.
Another option might be the gearbox Ford uses for the Ecoboost I-4. Same bellhousing pattern as the Duratec / MZR.
Speaking of transmissions again, I have no experience with dog box transmissions. I have read a little.
How is your dog box set up? I mean some times I have heard a switch senses the shift and cuts spark etc.
Is yours set up to just lift, bang the shift lever then back on the gas, or does it require use of the clutch at all besides starting out?
I'm told it requires just a minimal or no lift to change gears. No clutch use other than starting from stationary. I haven't tried it out yet but one of my goals was to reduce boost loss during shifts. If it doesn't shift well as set-up I may want to look into a gearshift pressure switch to trigger a timing retard in the haltech.
I guess if you had an adapter plate, and then you had another adapter plate, and then you had a tube in the middle carrying a shaft on bearings, you'd have a torque tube! Moves your swap engine forward while leaving everything from the stock bellhousing back alone.
I love how this car has turned out but I'm less likely to attempt it now than i was before i read this thread! lol
Hey JohnV, just found this crazy build thread of yours - just way too cool! Its been a long time since those grad school days in Cleveland Hts working on one of your old RX7's and running the SHO down in Akron. Glad to see you are now a master fabricator! As for me I am slowly working on my C5 FRC project. Best of luck with the 2020 season!
deaconblue said:Hey JohnV, just found this crazy build thread of yours - just way too cool! Its been a long time since those grad school days in Cleveland Hts working on one of your old RX7's and running the SHO down in Akron. Glad to see you are now a master fabricator! As for me I am slowly working on my C5 FRC project. Best of luck with the 2020 season!
Blast from the past! Good to hear from you John.
Definitely still a novice when it comes to fabrication but a decent welder and a lot of hours in the garage definitely are the devil's playthings :)
Moving the ABS unit and the ECU allowed me to relocate the shock canisters to be more accessible. Made some simple aluminum brackets to hold the catch cans and the power steering pump. Cleaned up the plumbing and the wiring a little bit. I think it looks a lot better.
The RX-8 has sort of been on the back burner the past couple of months. I've been slammed at work (not complaining about that with so many folks out of work :( ) and have been spending my free time on fitness and some projects around the house. Where I left off with the car was the cooling system never got bled completely because of a minor leak. I got that tied up yesterday and since Peter is going to tune the car next Saturday, I figured I should get some miles on the engine, make sure there are no oil or water leaks and just check stuff out. I meant to put the GoPro on the car but totally forgot and it looked like rain so at the end of the drive I just grabbed a quick cell-phone video.
https://www.youtube.com/watch?v=GXvg7bIe8Mc
A couple observations.
1) Dog boxes are badass
2) The new motor is crazy responsive. I had the boost controller turned off, so basically just wastegate spring pressure (though I wanted to stay out of boost for the most part because we switched to speed/density from MAF this year and the tune may be pretty far off at this point. But even at less than 15% throttle it's building boost at 2,500 RPM. The new engine is around 10.2:1, a small jump from the nominally 9.7:1 engine that was in the car last year, but I now know that I cracked a few rings very early on, so that engine was really low compression all last year. I love these EFR turbos.
3) The Haltech's long term fuel trim tuning is really effective. In just about ten minutes of driving around it dialed in the low-load part of the fuel map. Granted, what Peter sent me was really close - he's a master - but I really like this computer.
Sorry, garbage video, but I'm excited :)
Very cool, wow. Always wanted to see an RX-8 with a turbo four.
I recognize your name from BMWs and autocrossing. (I used to autocross an E36 years ago).
Thanks for sharing this with us.
Thanks Eric. I miss my E36M3 and my DSP E46 but time marches on.
Sometimes you realize that you only exist to serve as a warning to others.
I haven't updated this in some time, but that doesn't mean I haven't been busy. I sort of decided in March that I had strong doubts that there would be a real National autocross season this year, and I wasn't wrong. Still, I was trying to get the car tuned and ready to roll in case we did have regular events. That was foiled by an oversight on my part.
It's here that I diverge into a little bit of background on Mazda's MZR engine family. What's missing from the timing chain set pictured below?
If you answered "a keyway" give yourself a pat on the back. Mazda MZR / Ford Duratec I-4 engines don't use keyed pulleys on the crank or on either cam. Like most modern engines, these engines are held in time by friction. Unfortunately for me, friction timed engines don't get along well with dog-ring gearboxes due to the shock loading incurred during normal shifts. Hooray!
So, back in July I had struggles tuning the car on the chassis dyno and eventually discovered the engine had slipped out of time. So this was the next natural step:
I sent the crank, timing chain sprocket and balancer off to Joe at Dynotronics, who machined / broached keyways into everything for me.
And now everything is back together and running happily, so the next stop is back on the dyno.
In the meantime, I came across an inexpensive Haltech-branded Racepak IQ3 digital dashboard. Given the lack of information supplied on the factory Mazda dash, I figured this would be a nice option because it can read any parameter that the Haltech (or Mazda) puts on the CAN bus. The only rub is that Mazda saw fit to NOT put fuel level on the CAN bus. Instead, the factory instrument cluster does some sort of averaging of the two fuel level sensors (one per side of the saddle tank). So I made up a small conditioning circuit to convert the two tank sensors to a total fuel level which is then put on the CAN bus by the Haltech and sent to the dash. I cut up a small piece of aluminum sheet to mount the display, painted it black, did some wiring, and installed it where the stock cluster lived.
So that's where we stand today. Hopefully, back on the dyno in a couple weeks when it's not so oppressively hot here.
Seems like a really odd engineering choice to not key the crank or cams.
It seems like it doesn't reduce much complexity on the manufacturing side. The block has to be cast and machined for the TDC stop, special friction washers engineered, and special tools designed and manufactured for assembly. All that to save milling a keyway on the crank, drilling dowel pin holes in the cam, and the cost for a key and two dowels for the cams. Now you've got 3 catastrophic failure points, and a new way for someone at the factory or tech at the dealer to bungle a simple operation.
Maybe they thought they would have an issue with oil leaking through the keyway, but that's a simple fix with a dab of RTV on the key.
I suppose instead of keying the crank an alternate solution would be to add some loctite retaining compound of your choice. Not a good fix for the cams though.
Retaining compound is okay until you have to remove stuff.
In reply to Stubbs :
To set timing, there's a pin that threads in through a service hole in the side of the block. crank balancer bumps into the pin. Then you install the cam lock plate which fits into slots on the end opposite the gears. Only fits one way with cams in the head. Slap the chain on, throw in the guides and tensioner and zap it all together with a impact. Its actually very fast and kind of fool proof to time, once you understand it. I learned it because I took the crank pulley off my wife's escape before I knew what I'd done.
Bmws are suffering through the same problems by going without keyways
Friction timing works fine on cars with torque converters. It was even fine with the synchronized box last year. It didn't so much like the dog box
This dyno session went a lot better. We spent a lot of time getting the idle solid, fixing the transient throttle, dialing in the cam control, playing with ignition timing and then finally some boost control. I can't say enough good things about Peter Florance - this car was his first Haltech experience and he's managed to make the car run flawlessly with very little time behind the keyboard.
Also, you just gotta love the sounds from the straight-cut gears.
In reply to johnvitamvas :
Nice! The serpentine belt looked funny at the end there. Probably normal, just caught on camera in a way it could be seen.
You'll need to log in to post.
