When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
280-350whp Track Car Gearing Inquiry (Who's Running What?)
I'm going to try to keep this post as concise as possible. Here goes.
I'm currently running 270whp on track (~295 on street) with a 6 speed Miata transmission and 4.3 Torsen diff. My gearing is too short for certain tracks already, I've already had a 6 speed failure on my current setup, and I'm planning on upping my power level even more later this year and next. I've more or less decided my move is going to be a 3.6 diff and a BMW 6 speed swap (real similar gear ratios to the Miata 6 speed) sometime this year. Is that gearing too tall at 270whp? How about at 325-330whp? For reference, my car is on 245 RT660's on 15x9's with moderate aero (airdam, 4" splitter, 60" wing).
I'm running tracks in SoCal only right now (Willow Springs, Buttonwillow, Chuckwalla) and my current top speed with the 6 speed/4.3 pairing is 130mph. Last year at WSIR, I was getting the car up to 128mph running 210whp. With the current 270whp the car is running, there's no point in going there until the gearing is lengthened out. I was also getting up to 118mph at Chuckwalla on less grippy tires and 210whp, and might be close to my top speed limit going there now.
I've decided that before swapping my GT2560r setup for an EFR, I want to do the transmission swap first and alleviate my concern about killing another Miata 6 speed this year. I milled over transmission/diff swap options and ratios for hours over the last few weeks, and have pretty much decided my end goal is to swap in a BMW ZF 6 speed transmission via Kmiata's kit, a 3.6 R&P, and then maybe a Getrag 3.42 or 3.73 further down the line. The ZF 6 speed swap will bolt up to a Miata diff, whereas the 5 speed and Getrag 260 don't have that option, and I have no interest in dropping the coin for the trans and diff swaps at the same time. The ZF 6 speed ratios are also very close to the Miata 6 speed, so I could do a 3.6 R&P swap first, go and run WSIR in the meantime, then do the trans swap at my leisure after that.
A few charts of my current gear ratios compared to the ZF and Miata 6 speeds with a 3.6 torsen:
I don't think that the diff ratio is what busts gearboxes, it is the output, both quantity and quality (spikes/reversals). You are comfortably under the 240kw/320hp threshold that I understand is the limit, beyond which the failure zone starts - that suggests maybe a look at your driving/shifting style may pay dividends, or perhaps you just pick bad boxes lol!.
By all means change the diff ratio, if you are losing too much time holding it under the rev limiter in top. The trade-off of a short diff ratio is the acceleration gain and the top speed loss, if you run on long straights often you will probably be faster with a longer diff., at least on those/that circuit. On the tracks with shorter straights, you may/will be slower. Changing the CWP is no big deal, you have options beyond the 3.6 (3.9 and 4.1) - that's a pretty big drop from 4.3 to 3.6, and an intermediate step might be a useful experiment.
One other thing you will notice going to a longer diff, is that gearchanges are more spaced out, and in some cases you won't need a change. I ran 4.1, 4.3 and 4.8 diffs at different times in my atmo racecar, and the frequency of gearchanges with the 4.8 cruelled any competitive advantage of that ratio, and in fact cost me a new bottom end after I buzzed it on a missed shift. YMMV though.
Most definitely agree with you regarding output vs diff ratio. I actually wasn’t under the impression that the ratio affected longevity to any notable degree. I was, however, under the impression that torque was the gearbox killer (although obviously horsepower is gonna correlate with torque to a certain degree). And the general rule of thumb seems to be at the 6 speed is good for up to 280wtq before you’re opening yourself up to disaster? My car’s making 270wtq on e85 but I killed my last 6 speed directly after overboosting on a very cold night.
Aside from that, I do believe I practice proper mechanical empathy while shifting, but there’s always a chance I’m a complete hack! All the more reason to bulletproof my drivetrain. Wouldn’t wanna have to improve my technique or anything like that.
My proposed diff ratio change is entirely due to the fact that I’m going to need more top end at Willow Springs. How much more top end, I’m not sure. I’m guessing my car currently will top out around 135mph there. I’m just not sure how much more it’ll need if/when I’ve got it making 325 or 350whp on that same track.
Good point though regarding the frequency of gear changes and experimentation with the intermediate steps. Maybe if I find a 3.9 R&P for a reasonable cost, I could give that a shot first and see if I’m in the ballpark. We’ve got courses here that range from predominantly tight and technical to wide open with only a handful of corners where you’re under 80mph. If I could get the most top end I’d need on the fast tracks without sacrificing gear spread on the tighter tracks, I guess that’d be the holy grail.
Sacrilege I know, I know, but food for thought anyway: ZF8-45 8 speed autos are available dirt cheap, literally by the hundreds, on Car-Part. Ratios and speed ballparks (ymmv based on wheel dia) below.
Weight ~ 20-40 lbs heavier than strong (TKO 6 speed or T56. depending on which flywheel and clutch. There are guys already running without torque converter which brings the weight down to within about 10 lbs of strong 6 speed. And paddle shift! total cost probably 2200- 3k if you buy an adapter and TCU; half that if you make your own adapter. I have a Domi works adapter on the way and will initially be using a CANTCU controller.
Funny you'd mention that swap, my buddy was just talking about doing the ZF8 swap. Looks badass and surprisingly inexpensive for what it is. What're you using for a trans support? Or will you be making your own?
This might sound dumb, but the only reason I haven't looked into it further is because I really like driving a manual as well. I know the auto could be even faster, but I still get something out of rev-matching downshifts haha.
I've been thinking that a 3.6 and 6 speed will be going in my car at some point. I will probably be at whatever power level my 2560 limits me to.
For autocross, being able to do ~60 in 2nd gear is pretty crucial. I've watch the MSMs go around and they're constantly rowing between 2nd and 3rd.
Highway RPM will be a big benefit as well.
It seems like a common enough swap for people pushing over 250, considering the 5 speed becomes less of an option. I assumed it would be fine on a car in the 250-300 HP range, but have to admit I haven't looked into it very closely.
Yeah, I think the 6 speed/3.6 combo is the ticket for a lot of people, especially dual duty cars that see lots of highway time. My car sits at ~3750rpm on the freeway going 70mph. I'd probably mind this more if I drove my car faster on the street, but I baby the thing when it's not on the track haha.
Yeah, the limiting factor for both transmissions seems to be torque output above horsepower. The general consensus on the 5 speed will hold 250whp/220wtq before you're in for a certain failure, but I've heard multiple accounts of 5 speeds failing well below this. Mine lasted a whole season at stock power, and half a season at 210whp before I pulled it out. Never ran it with more power. General consensus on the 6 speed is that it's good for 280wtq and there seem to be fewer random failures under that power level than the 5 speeds exhibit. I originally thought a 2560 wouldn't be enough to kill a 6 speed but seem to have since proven myself wrong. I'm pretty sure I did a couple pulls that peaked above 280wtq shortly before my 6 speed started failing on me.
total cost probably 2200- 3k if you buy an adapter and TCU; half that if you make your own adapter. I have a Domi works adapter on the way and will initially be using a CANTCU controller.
Did some more reading on what's involved in the auto swap. Do you have a DBW throttle body, or how are you handling rev-matching on downshifts? And what ECU are you on? I didn't think of it until reading some of Padlock's build thread but it sounds like the 8 speed swap would be ruled out for me because my Megasquirt doesn't support an electronically-actuated throttle body. Swapping to a Haltech (and all of the retuning that'd go with it) would probably put me in the same cost ballpark as a 6 speed ZF swap, and a hell of a lot more work haha.
Originally Posted by Padlock
Do you have any more details on this "without torque converter"? What is put in its place if you dont run one?
Yeah, I'm interested in this as well. Thanks for the link to your calculator! I ran across it months ago but completely forgot it was on here until now. I'll plug in my recent dyno numbers and see what I get. I'll have to pull out data from my recent track sessions as well and reference what speeds I'm seeing at each local track. A fun homework assignment!
Peered into your build thread a little. Are you still considering an 8 speed swap for your car?
As long as the mods don't mind, I don't care that this thread's getting a little derailed. Here for the exchange of ideas and banter.
The 3.6 final drive for the diff seems like a no brainer to me, especially at your power level. You might actually be able to get some traction in 2nd vs just spinning. Unless you're building the car for some specific TT class and really concerned about finishing in a certain gear, the 3.6 should be a better experience overall I would think (street and track). The gear ratios on a 6 speed are fairly close together and as mentioned you get a nicer cruising RPM on the highway.
Regarding a tranny swap. I have no clue what the options are but it'll be interesting to see what route you choose. I have read about the circlip mod (I'm sure you have too), but the ZF swap might still be the better option in terms of longevity/durability.
Regarding the ZF 8 speeds and other sort of "sequentials," they certainly look like great options as they seem to be quite strong and are fast, but I'm a 3 pedal guy myself and enjoy heel toeing (even if I'm not great). It's not something I would ever consider.
to give a bit more detail on the ZF8's: I am using an ME442 ECU, with dbw, that will be talking to the TCU, which in my version 1, will be the CANTCU trans controller. This will be with factory lockup torque converter, and will enable either full auto mode using the BMW programming, or both sport and manual (paddle) shifting. I'm pretty sure the CANTCU will also work just reading MAP for upshifts but you'd need a throttle blipper for rev matching downshifts.
Version 2 - (I have 2 of these trannies (sorry) - one with 30k miles, second with 65k miles, and I paid $175 for first and $150 for second) will be without torque converter. This MAY use the CANTCU also, depending on how much software mods they get done, but if not, will use the TurboLamik controller and software. The TurboLamik TCU requires a new board to be soldered into the Mechatronic unit inside the trans - sounds daunting but a lot of guys are doing it very successfully. Their TCU and software allows tuning of all clutch solenoids and resulting clutch pressure, clutch engagement aggressiveness, and elimination of the torque converter and lockup clutches in the torque converter. Their are at least two guys on FB who have done this successfully. I'll try to find the links when I have a bit more time - or you can go to FB, "ZF 8HP 45 8HP70 8HP90 Transmission swap and technical info", search for torque converter elimination - you'll find quite a bit of info. Because this is the interwebs, there are those who say it cannot be done, you'll fry your clutches etc bs to infinity. But of course it has been done. Re clutch wear, there are more clutches at work in the ZF8 for standing starts, therefore distributing the load, than in the Porsche DCT - it has only two. Also search Zeroto60tube on YT - bunch of Aussies that solely use the TurboLamik with the -8HP - for drift, other hotrods, and offroad/mudders/rock climbers.
I think I'm going to fabricate a separate trans mount like the T56 versions etc., but may use the PPF. Domi works said they're shipping the adapter this week. I'll post more stuff when I make more progress.
and - "without torque converter" an adapter has to be fabricated to utilize the TC splines to mate to the flex plate. There are two versions I have seen - one guy uses a straight solid shaft, the other uses a spring disc like a regular clutch with a sprung hub - I'm in the process of cutting apart the torque converter from one of my trannies har har to start that process. So what happens is you start the engine in park, with all internal clutches dis-engaged, select the gear, then energize them with a linear pot feeding the TCU.
The conventional wisdom is that with 300-ish rwhp the 6-speed + 3.9 is faster at California tracks than 6-speed + 3.6, while the 3.6 is better-suited to street driving.
to give a bit more detail on the ZF8's: I am using an ME442 ECU, with dbw, that will be talking to the TCU, which in my version 1, will be the CANTCU trans controller. This will be with factory lockup torque converter, and will enable either full auto mode using the BMW programming, or both sport and manual (paddle) shifting. I'm pretty sure the CANTCU will also work just reading MAP for upshifts but you'd need a throttle blipper for rev matching downshifts.
Version 2 - (I have 2 of these trannies (sorry) - one with 30k miles, second with 65k miles, and I paid $175 for first and $150 for second) will be without torque converter. This MAY use the CANTCU also, depending on how much software mods they get done, but if not, will use the TurboLamik controller and software. The TurboLamik TCU requires a new board to be soldered into the Mechatronic unit inside the trans - sounds daunting but a lot of guys are doing it very successfully. Their TCU and software allows tuning of all clutch solenoids and resulting clutch pressure, clutch engagement aggressiveness, and elimination of the torque converter and lockup clutches in the torque converter. Their are at least two guys on FB who have done this successfully. I'll try to find the links when I have a bit more time - or you can go to FB, "ZF 8HP 45 8HP70 8HP90 Transmission swap and technical info", search for torque converter elimination - you'll find quite a bit of info. Because this is the interwebs, there are those who say it cannot be done, you'll fry your clutches etc bs to infinity. But of course it has been done. Re clutch wear, there are more clutches at work in the ZF8 for standing starts, therefore distributing the load, than in the Porsche DCT - it has only two. Also search Zeroto60tube on YT - bunch of Aussies that solely use the TurboLamik with the -8HP - for drift, other hotrods, and offroad/mudders/rock climbers.
I think I'm going to fabricate a separate trans mount like the T56 versions etc., but may use the PPF. Domi works said they're shipping the adapter this week. I'll post more stuff when I make more progress.
Originally Posted by wombat57
and - "without torque converter" an adapter has to be fabricated to utilize the TC splines to mate to the flex plate. There are two versions I have seen - one guy uses a straight solid shaft, the other uses a spring disc like a regular clutch with a sprung hub - I'm in the process of cutting apart the torque converter from one of my trannies har har to start that process. So what happens is you start the engine in park, with all internal clutches dis-engaged, select the gear, then energize them with a linear pot feeding the TCU.
Torque converter innards coming up.
I did some research on the TurboLamik chip retrofit yesterday. Pretty cool they've got a way to way to control individual solenoid pressure and shift feel. I've never heard the term "closed loop" used in relation to automatic transmissions haha. At work (Ford Mechanic) we just input the solenoid strategy code for the specific vehicle following solenoid body or transmission replacement and then give it a short test drive to perform the relearn then it's done. I'd imagine you could get REALLY deep tuning solenoid strategy to dial in the shift feel you're going for. Personally, I have my hands full tuning my ECU and can't imagine tuning my own TCU on top of that with the amount of time I have right now lol. Would be stoked to see your progress with that project as it occurs, though! Keep a thread updated on it if you can!
The conventional wisdom is that with 300-ish rwhp the 6-speed + 3.9 is faster at California tracks than 6-speed + 3.6, while the 3.6 is better-suited to street driving.
--Ian
This is the information I came here for. Thanks for sifting through my excessively-worded inquiry and distilling the information I wanted from it haha.
Also running a 3.6 final drive with a 6 speed. I'm using a boost controller to run a flat 250ft-lb torque curve to keep the trans alive; roughly 310whp, but it's also much more driveable with a 3d map of engine speed and accel pedal vs target boost. Running DBW throttle with MaxxECU Race, which is fantastic and my new favorite (affordable) ECU.
I bumped up the limiter to 7200 on a fully built engine (rods, pistones, valve springs, etc.), so Vmax in 5th is ~135mph. Which is enough to keep from shifting to 6th at all the Mid West tracks, except maybe Road America. I was hitting the limiter in 5th at Mid Ohio in wimpy TT6 form (stock ~7000rpm redline, 3.9 final, 135whp avg). And 2nd is just shy of 60mph.
Did some more reading on what's involved in the auto swap. Do you have a DBW throttle body, or how are you handling rev-matching on downshifts? And what ECU are you on? I didn't think of it until reading some of Padlock's build thread but it sounds like the 8 speed swap would be ruled out for me because my Megasquirt doesn't support an electronically-actuated throttle body. Swapping to a Haltech (and all of the retuning that'd go with it) would probably put me in the same cost ballpark as a 6 speed ZF swap, and a hell of a lot more work haha.
I know at least one person is looking into if its possible to get a manual throttle blipper added to a MS to operate rev-match downshifts, but how the MS would (or would not) communicate with the TCU is yet to be seen. MS doesn't have DBW capability to my knowledge, which is a big miss. I admit that I haven't been keeping a super close eye on this either though because I don't have MS on my kswap, so it doesn't really apply to me. The tried and true method of doing it that I can confirm is that having a ECU that can communicate via CANbus to the CANTCU and a DBW throttle body seems to be the most logical approach. Haltech Elite 1500 and MaxxECU's are both known to communicate well based on setups I've seen.
FWIW, I've pretty much written off TurboLamik as a viable TCU solution. Not only do you have to solder on a new mechatronics board to the trans for the TCU to work, but the TCU does not provide a reliable means of smooth downshifts without locking the rear tires up. The support for issue resolution is also quite awful from the reviews/comments/posts I've seen. You'll see countless videos of upshifts and drift guys using that TurboLamik TCU, but you won't find much on downshifts because, well, it's awful... I've spoken to a few guys who have shifted from TurboLamik to CANTCU for this season and they've already stated its FAR better due to downshift smoothness alone. As an added benefit, CANTCU can control a factory (unmodified) transmission
Originally Posted by Z_WAAAAAZ
Peered into your build thread a little. Are you still considering an 8 speed swap for your car?
Any trans swap is a ways down the road for me yet due to the costs of scope creep that it will introduce (new ECU, TCU, wiring, forced induction, rear end updates, and likely onboard DAQ), but the ZF8 spd is at the top of my list for when that happens. The plan for the car is to be 500whp capable. While I am a die-hard manual fan, I'm also aware that level of power is going to be a handful in the NB chassis, so automating the 3rd pedal so that I can focus only on the steering, brakes, and gas seems to be what will make for the fastest car and easiest to drive. Given the gear ratio spread and quick shifting nature, it also lends itself to being great for both street and track driving.
Originally Posted by wombat57
and - "without torque converter" an adapter has to be fabricated to utilize the TC splines to mate to the flex plate. There are two versions I have seen - one guy uses a straight solid shaft, the other uses a spring disc like a regular clutch with a sprung hub - I'm in the process of cutting apart the torque converter from one of my trannies har har to start that process. So what happens is you start the engine in park, with all internal clutches dis-engaged, select the gear, then energize them with a linear pot feeding the TCU.
This makes sense, but is also what I was fearful you'd say. While the weight reduction is intriguing, losing the torque multiplication of the converter is unfortunate. Requiring additional hardware just to get the car moving and then trusting that the internal clutches to engage smoothly for 1st and reverse (and having the control to do that with your TCU!) is also something I wouldn't take lightly for a car that I personally plan to street drive through stop-go traffic.
Padlock - I’ll start another thread after this comment - I agree with you on almost everything - my plans are for a bit less power than you but I am continuing to research what will be required and effective/ possible for “TC less” install.
Downshift blips are a work in progress for sure. Re losing the torque multiplication, I never do drag racing starts and with the almost granny gear first I think you could lockup clutches in 1st, then light the tires from an idle. Re clutch wear, an interesting YT vid by the ZerotoSixty guys showing standing starts in gears 1-3, and the DCT trannies do standing starts frequently with only one clutch pack. Summary: you raise good points, there are issues to be addressed. I do have the TC apart and will post pics on new thread.
Padlock - I’ll start another thread after this comment - I agree with you on almost everything - my plans are for a bit less power than you but I am continuing to research what will be required and effective/ possible for “TC less” install.
Downshift blips are a work in progress for sure. Re losing the torque multiplication, I never do drag racing starts and with the almost granny gear first I think you could lockup clutches in 1st, then light the tires from an idle. Re clutch wear, an interesting YT vid by the ZerotoSixty guys showing standing starts in gears 1-3, and the DCT trannies do standing starts frequently with only one clutch pack. Summary: you raise good points, there are issues to be addressed. I do have the TC apart and will post pics on new thread.
It's very possible that I missed it, but I have watched some of their videos and it is worth noting that DCT BMW transmissions are different than the 8HP transmissions. I have not seen any videos of a direct drive 8HP launching in their media, but if you have a link I'd love to watch it.
drag racing starts aren't the only reason to want a TC. They create a nice fluid damper between the engine and transmission during those stop/go activities. It also helps isolate the trans from any 1st/2nd order vibrations the motor may be emitting (which there are many of on a k24 with balance shafts removed in my case). I actually can't find any data that explains clearly when BMW locks up their TC's. I know on my truck, it is unlocked VERY frequently. At the end of the day, I'm interested in saving weight, but not if it compromises the reliability of the transmission in any way. If 10lbs is all I can save by replacing a TC with a direct drive pressure plate that would have to be custom made ($$$), the incentive to be a test dummy doesn't seem very high and cost per lb saved isn't great most likely. That TC weight is nearly as low as you can get it and relatively centered in the chassis, so it's not exactly an awful position to have it. Selfishly, I'm already at the point of needing to ballast weight for any class I may want to compete in, so even if I add fairly serious weight with a 8HP swap, I'm okay with that. Looking forward to chatting about this in your new thread... Share the link here when you have it
I just can't imagine using the clutches/brakes in a planetary trans as the launch clutch. I can see this possibly working on a time attack only car that doesn't spend any time in low speed city driving or drag race launching. I'm imagining stop and go traffic on a hill and it's going to die. The clutches on a DCT are basically like a motorcycle wet clutch with many metals and frictions, which gives it a lot of surface area to spread out the heat. And to go with that a generous oil supply to the clutches for cooling. This whole concept scares me as a powertrain controls engineer.
The main benefit of removing the TC would be reduced engine inertia. Essentially the same benefit as a lightweight flywheel. The downsides are numerous.
I'm thinking more like this beastly 350z which has a direct drive. I say direct drive because if you were to say replace the TC with a flywheel / clutch setup it's probably going to be close enough in mass to the TC + flex plate to be of dubious value given the effort and cost. He's using the KERS mounted in the bellhousing (in place of the clutch/flywheel) to simultaneously start the engine and launch the car, which could be possible with a high torque starter.
In car video:
You can see how he launches in pit lane in 1st gear with the engine off.
02-05-2024, 06:33 PM
280-350whp Track Car Gearing Inquiry (Who's Running What?)
0
