I think that would be a perfect fit length wise in my car! Who makes the bell housing, adaptor plate, and what tail section is that? Thanks.

 

Nice pics ;o) Whose FD is that? How exactly is this engine in this setup above mated to the trans? I'd be interested to hear a breakdown of what was used in terms of: pilot bearing, flywheel, clutch, input shaft, master cylider etc, adapters etc.

thanks

 

https://www.rx7club.com/3rd-generation-specific-1993-2002-16/t56-trans-install-474847/

 

Just FYI. That custom T56 setup in the pics above cost the owner around $7500 and required a custom input shaft (to fit FD pilot bearing and align with clutch), custom hydraulic throwout bearing, cutom flywheel, custom clutch, G-Force internals, custom bell housing, etc. Hardly a pop and swap option. If you want to move the tranny back with an LS1 bellhousing you will need to have a longer custom input shaft made and you'll likely need to shim out the slave. IOW this is no easy retrofit. The best option here IMO is to have some forged internals made, for the FD trans, through a protyping firm with the assistance of one of their engineers, or through a custom gear specialist.

 

Rob, I'm sure a good bit of the cost went to G-force for their upgraded internals Also, how do you propose to overcome the issue of casing flex when using the stock FD trans? IMO, if you're serious about making lots of power, money isn't the issue; it's longevity and reliability. Having said that, prototyping is VERY expensive. There are already *several* options for upgraded internals of the stock FD tranny. Why reinvent the wheel?

~Ramy

 

You're my friend so I don't want to dog you here, but I would like for you to show me a link to forged 5spd gears that are currently available. There are no gear replacements that I know of besides the 6 speed swap so this reinvent the wheel analogy doesn't work. Also keep in mind that bone stock T56s are prone to shift fork and synchro failures. For this reason I purchased an upgraded unit with upgraded shift forks and snychros, but it wasn't cheap. Add on all the custom goods needed to bolt it to a rotary and you are up in the 5k range easily I'm sure.

A solution to target this specific problem will not only help Greg, but others as well. I am confident that custom gears can be designed and cut for significantly less than the custom tailored T56 setup mentioned earlier. I already posted a list of a ton of companies who specialize specifically in designing/making custom gears, we should be patient and see what they say. Prototyping what is expensive? I went from design to finished product on my custom cut stainless plates in 2 weeks and the developement cost was not that bad. A gear company will be tooled up to do this type of work, this is not like developing a new ECU here.

Also, it is important to delineate the differences between reality and theory. At what power levels is chassis flex really an issue? I'm not being a smart-***, I just really don't know. Does anyone here definitively know the answer to this question?

For the hand full of those who can justify a solution to address chassis flex a custom trans is an ideal solution, but as you said, have to pay to play. That trans above is essentially completely custom built, hardly a practical solution for moderately high HP vehicles. I believe Greg is looking for a solution for a 500ish? brake torque car (which is plenty ofcourse). Are we talking about Greg's car here? or some theoretical beast of an FD?

Money, performance and longevity/reliabilty are all important factors that should weighed accordingly. You wouldn't shell out to put 300M axles on a 300RWHP car would you? Well maybe you ;o), but most people wouldn't because it's overkill and requires funds that could be used elsewhere. It has nothing to do with being cheap, it's just efficient allocation of resources.

 

Woah slow down Rob. My post wasn't meant to be hostile in ANY way, shape, or form. Just an honest critical perspective if you'd be open to that

Off the top of my head (and yes you may be surprised about these...but hey, the JDM market is my niche ):

- PPG (Pfitzner Performance Gearbox): they reuse some OEM parts, and they alter the shift pattern
- Guru Motorsports/Xtreme Rotaries: the best I've seen thus far, but it also alters the shift pattern
- Quaife: in addition to the famous Grex Greddy Power box they make - 6 speed dogbox - they also make a 5 speed synchro and 5 speed dogbox
- TFR (Top Fuel Racing)
- A-FR (Anniversary Factory Racing)
- Os Giken
- Revolution/Revo-Tune
- Axia Sports
- Route6

Those are only the ones that I know about and remember off-hand. I'm sure if I had a bit more time I could come up w/ a few more

Of course...IMO, that's to be expected from any standard sychro box.

Smart man. And I'm sure it wasn't cheap. But realistically, from what I've seen out there, $5K doesn't seem to be cheap at all when it comes to serious tranny options. So I'm speaking relatively, that's all.

Rob, I'm w/ you 100%. Heck, if we could devise some kind of bolt-in kit (or at least a full list of part #s w/ prices and availability), I think it would go MILES, b/c the question of "which tranny" has haunted this forum for YEARS now, w/o a single final satisfactory answer.

Hey maybe you simply know more than I do and I'm assuming incorrectly. But anytime I think of "prototype," all I see is $$$ flashing before my eyes hehe. Now granted, if we came up w/ a solution that would be used in the long run, maybe some sort of GB could offset the costs...

But see (again, not to be negative, but...) that really isn't the problem. There are a plethora of aftermarket forged gearsets available, some dog engagement, and some synchro (take your pick). Many are toted to be so strong as to stand up to 700 or 800 ft lbs of torque. But as I mentioned above, the GEARS aren't the problem; you'll be breaking input & output shafts as well as shearing the gears due to tranny case flex WAYYY before you break any gear due to power levels...(see below).

Just to make sure we're 100% clear, I'm not speaking of chassis flex. I'm speaking of the tranny case itself flexing. It's lightweight, made of aluminum. At high power levels, the points which the input and output shafts enter/exit the tranny are under great stress, and flex, leading to grinding of teeth and eventually shearing of gears as I mentioned earlier.

Several people on this forum have questions the reliability of this claim every time I mentioned it. I've spoken to - and at length mind you - some of the biggest authorities when it comes to tried and tested road and drag racing FDs - mostly from down under. And I'll tell ya, EVERY SINGLE ONE has told me the same. Guru, Holinger (which supplies trannies for some full race FDs), City Performance Centre, Carbon Man (Ian, who designed and produced countless parts for the actual Australian Bathurst race FD, and who has a very large hand in race R&D), etc etc. All of them tote the same tune. Once you go beyond the 500 - 600 ft lbs of torque mark, tranny flex IS an issue, and regardless of what forged gearset you have in the stock tranny, the stock case is simply not strong enough to handle it, and WILL flex (at the aforementioned points).

I don't believe these are unrealistic power levels. Several ppl are flirting w/ the 500 ft lbs line, and some have already passed it. And the biggest factor comes in when you're know you're flirting w/ that power level. Having a tranny fail and lock up on you while you're in motion is NOT fun. I remember Ryan (Don't Be A Rikki) describing to me what happened when that happened to him while down the straight at the drag strip. I have NO plans of having that happen to me.

Very funny Rob

Again, I simply think you're re-inventing the wheel b/c several upgraded internals ARE available. So here's my proposal to you. YOU find me a way to reinforce the stock tranny casing so it won't flex around those points, and I'LL find you the best JDM upgraded internals set, and see if I can get a GB discount on it. Sound like a deal? That way, there's no prototyping, and we can keep all the factory compatible parts w/ the 3rd gen tranny

~Ramy

 

I have a couple of trannys I intend to disassemble and combine into one. Which means when the parts are apart and cleaned I could sit down and do an analysis of the tranny housing w.r.t. deflection, stress, and potential bearing life issues.

Naturally higher torque will produce more housing stresses, but a lower helix angle on the gears (in the case of heavy-duty synchro box design) could help some.

Anyway, it would an interesting study since we know the input shaft bearings tend to fail often - and if the calculations show the bearing life should be good, but the housing flex may be at fault, it will give a pretty good idea if the stock housings will be insufficient for heavier duty designs.

Dave

 

Dave, forgive my utter ignorance, but you can't just welt/fasten on a plate of metal around that area to reinforce it and call it a day?

Oh and Rob, there's a MAJOR drawback I forgot to tell you about. The vast majority (if not all) of these gearsets are essentially roadrace gearsets. Thus, they're a 5-speed close gear pattern. The result is a very short 5th gear (topping out around 150 or so). That SUCKS. One of the things I really liked about the T56 tranny w/ the Z06 gearset is that it essentially acts like a 5-speed close gearset + a 6th - and very tall - overdrive gear to maintain top speed. I forget exactly, but I believe CMonakar (giving credit where it's due hehe) calculated it for me, and the theoretical top speed for the Z06 6th gear was well beyond 240mph IIRC lol. Now THAT'S the best of both worlds...awesome acceleration, w/o sacrificing your overall top speed.

~Ramy

 

Heh, nope. I wish.

Dave

 

Thanks Ramy. The big benefit I see to the 6th gear is it would allow you to cruise at a lower rpm on the highway which should improve fuel economy and reduce noise. By my calculations with 18/315/30 rear tires 76mph =2,300rpms with the .56 Z06 6th gear ratio.

Wheel/Tire/rpm:
wheel dimater 18
tire width 315
aspect ratio 0.30
rpm: 8,000

FD Stock
3.48---4.27---42mph
2.02---8.28---73mph
1.39---5.70---106mph
1.00---4.10---148mph
0.72---2.95---205mph
0
4.10

Z06 Stock
2.97---12.17---50mph
2.07---8.48---71mph
1.43---5.86---103mph
1.00---4.10---148mph
0.84---3.44---176mph
0.56---2.29---264mph
4.1

Another interesting option is to use the M6 gear set with an upgraded 5th and 6th gear ratio.

C5 with upgraded 5th 6th
2.66---10.90---55mph
1.78---7.29---83mph
1.30---5.33---114mph
1.00---4.1 ---148mph
0.80---3.28---185mph
0.70---2.87---211mph
4.1

 

What about the gearset that acosta motorsports offers. It supposedly comes with the input and output shaft as well. Its all libery parts. Does anyone know if that gearset is effective? Im looking for a trans also. I have a single turbo 20b with the stock trans. It hasnt broke yet but i know its coming. Its about to blow up. lol. but anyway, ive been looking and looking into transmissions, and even after reading all this valuable information on this thread, still no one knows what will work. :/ everyone goes back to flex in the casing, Has anyone experienced this? Acosta says theyve had a few 20b cars making btwn 8 and 900 hp with this gear set in the car that fits in the stock case. But then again, hes there to sell also. This stuff drives me crazy. haha

 

No, but it cracks me up how many people would rather find a band-aid for stock parts they know damn well (or should, anyway) are weak point in the drivetrain... just to preserve the "soul" of the car.

When you reach certain power levels, you have to at least be open to the idea that using Mazda parts might not be the best solution for your application. Take the differential, for example. Many replace the stock Torsen differential unit with a TII or Kaaz unit, replace the axle center shafts with 300M axles, and add differential braces and/or replace the bushings with Nylon, poly, etc. To them, that's bulletproof, and maybe for their power levels, it is (or nearly so). But for some it isn't.

The stock differential housing is a crappy cast iron piece of ****, to put it bluntly, and it was cast thin to keep weight down. Several people have cracked or grenaded them, one of the more famous being Vosko (although I think Vosko could break a ball bearing if he put his mind to it...). In talking with Ari Yallon extensively about the problem, he said they kept welding material to the housing and hoping it would hold up, but had gone through several. He didn't have a better solution than that.

How about the inner stub axles? People replace the center shafts of the axles with 300M shafts, never thinking for a minute that the inner stub axles might be a problem. Ari said he was breaking them every two passes until they started buying new parts from Mazda and having them cryo treated first. Then they held up for 8-10 passes, on average. Granted, most people aren't at those power levels or abusing their cars by drag racing, but some are.

Ari had to keep the stock differential because of the class he raced in. So should everyone keep throwing as many band-aids at the stock differential as they can to make it hold up, or should someone throw a reasonable amount of money at replacing it with something much stronger, a little lighter, and with far more gear options? I did, and now others can too for about the cost of a Kaaz/300M upgrade.

We're talking about the same situation here. Some people seem to want to save the stock transmission no matter what, forgetting that A) it was NEVER designed for the power levels many people are making today (stock was ~220 lb-ft. of torque, remember!?!) and B) Mazda does NOT have a reputation for producing ANYTHING built like a brick ****-house. This is not a Toyota/Getrag 6-speed we're talking about. Are there any sub-9-second RX-7s using a Mazda transmission? Not that I know of.

Replace the gears if they're readily available and that's the way you want to go, but why not entertain the OPTION of replacing the OEM 5-speed with a transmission from another manufacturer? Not everyone needs a custom twin-plate clutch, and not everyone needs a custom assembled T56 with upgraded guts. Most shift fork issues in the T56 come from a sloppy drivetrain and ham-fisted owners anyway. A stock LS1 box would last most people the life of the car, given reasonable power levels and use, and an aftermarket shifter. After all, they hold up in 3,700+ lb. Camaros and Firebirds just fine, even with far more torque than most rotary owners are going to be making.

If everyone based their opinion of the cost of a Ford 8.8" IRS conversion on what I spent (~$6k because I soaked up all the R&D costs and used all new and upgraded parts where available to build a differential and axles from scratch), no one would do it. The same goes for the T56 example given so far. I know damn well people could convert for less, so let's not throw that option out the window just yet, especially when gear sets for the stock box aren't going to be cheap either.

 

Nothing against Acosta, but here's a good response to that...FYI, Ivan (20bfd3s) also broke his stock tranny at about 680 ft lbs of torque: https://www.rx7club.com/20b-forum-95/damaged-tranny-my-20b-today-368723/

Is that IRS conversion really necessary for roadracers, or only for drag racers?

 

It was intended primarily to provide alternate gearing choices for other engine configurations, although it'll work just as well for a rotary as it will for a piston engine. Drag racing was not really a consideration, although it's much stronger than anything possible with the Mazda components.

 

I'm just busting your ***** Ramy. You should know by now that I'm a sarcastic bastard. ;o)

where was this information earlier!? ;oP and which one of these companies offers forged gear cut to stock specs? NM I see this answer below (none). With that being said, the deciding factor for this potential little gear project hinges on what people really want.
Shift fork failures should be expected?

Everything is relative as you know. If this was a Z06 forum, price factors would be less influential. I guess the real question here is, what do we want? and how much are we willing to spend? I don't think there's a single person whose replied to this thread who would not spend the money to do this right, but there are trade-offs involved with all of the current solutions which make the final decision subjective.

A T56 is hands-down a more stout trans, built for high TQ applications, but there are trade-off associated with this swap. The difference in size (affecting shifter position) and the requirement for a custom clutch/slave/shaft solution. Although I agree with Jim that we should keep an open mind about all of this, I will remain a proponent of ugraded gears/shafts/bearings until housing flex is proven to be a real issue.

If it is decided that the stock aluminum FD trans case can't hack it, then it would be ideal to get one of t56 remanufacturers to build a bell housing/input shaft/slave combo that will position the rear window in stock position. While this is the ideal solution, it is probably unrealistic to assume that G-force or anyone else will do this for such a small market; not mentioning the fact that the 20B crowd will need a different configuration once again. Realistically speaking, the responsibility would lay with the FD community to develop the bell housing and slave.

I agree, and just for the record, I am not in search of a tranny as mine is already compatible with my power plant. I'm here because I find this to be an interesting discussion about something that will potentially fill a void in the FD community, which I am all about.

I completely understand where you are coming from. Prototyping can be prohibitively expensive, but it really depends on what you are trying to develope and how much time/money/work/resources are required for a manufacturer to make the end product. It will not be too expensive to have some forged gears cut because as I said earlier, there are companies that already have a wealth of expertise in this area specifically are tooled up and ready to rock. Prototyping a custom bell housing and custom slave cylinder on the other hand would be extremely expensive, there's no doubt about it.

Do you think that forged gears will continue to get sheared?

Very interesting information. I'd be interested to hear more specifics outlining exactly where the flexing occurs. Dave may be onto something, berhaps a set of gears could be designed with longer teeth and modified engagement angles to allot for this flex and remain sufficiently engaged. I assume the main problem is the instantaneous reduction of effective surface area of the gears transmitting the torque through the trans causing concentrated pressure points that are engaging slightly off-axis?

I bet all of these builders are using PPF setups... see below

Sweet jesus that would suck!

Just busting your ***** ;o)

It would be informationally useful if you could post a pic of the FD trans and point-out/label the pressure points. You could try to strengthen the trans itself by welding on a girdle. You could try to fashion some type of tranny mount to keep the tail end torsionally still while allowing a limited vertical range of motion, but it would require some really stiff motor mounts to keep the bell housing and tail end in sync. How many degrees of flex are we talking about realistically?

This bring me onto another issue: unnecessary torsional stiffening of the PPF.

I think the important thing to understand in regards to this subject is that these aforementioned driveline failures are the result of interrelated material limitations. The OEM motor mounts were not built to handle the amount of torque which is being delivered by these high HP applications. The design of the bonded 93-early 94 driver's side, aluminum arm/ bonded motor mount is extremely poor in my opinion and fails frequently. The later 2-bolt/liquid-filled puck design fails less, but still falls short because the comfort-focused design allows far too much power plant roll. As has been pointed out, this rolling motion combined with the ring gear forces test the limitations of the PPF when it attempts to transfer rotational force to the diff, which kept more-or-less torsionally static thanks to the diff mount arms. With that being said, stiffening the PPF - if it doesn't break first - is not in itself a solution. It will only make for a more efficient application of tortional force to the front end of the differential casing. The heart of the issue is that the rear most end of the differential is being held firmly in place via the diff mount while the front end of the differential is left hanging subject to torsional forces which originate at the engine. How could a stiffer PPF possibly solve this problem? it can't. With that being said it makes more sense to only reinforce the power plant frame where absolutely necessary. Reinforce the hell out of the fingers and the pass through section and leave it at that. The fact of the matter is that you want the PPF to be able to flex torsionally while retaining vertical rigidity or you will risk cracking the diff. By increasing torsional rigity you are just increasing the efficiency with which the engine transfers torsional force (via the tail end of the trans) to the differential casing. If the engine and diff rotate slightly then so be it, but allow them to do so somewhat independently. An IRS is the only appropriate solution, but a proper setup can be cost prohibitive. The most cost effective and pragmatic option is to upgrade your motor mounts and and diff bushings and reinforce the PPF just enough for it not to break. Stronger is not universally better.

The best way to address this tranny problem is to take a holistic approach. An IRS setup will allow for the tail end of the transmission to be solidly mounted. In conjunction with some high durometer motor mounts I believe that the chassis flex issue will be alleviated. The PPF setup is the root of this case flex issue because the rear end of the trans is essentially free floating. Any torsional rotation of the diff is transmitted straight to the rear end of the tranny case. The same goes for diff casing.

The more I think about this, the less blame I lay on the Fd trans and diff, and the more I blame the PPF layout. The T56's tail end is rigidly mounted in its OEM application and this no doubt alleviates a great deal of the casing flex which we would result if it utilized a PPF design.

In conclusion, I think the real issue hear is the overall power train design, not the FD trans casing per se. I'm inclined to believe that an FD converted to IRS and a solidly mounted trans with some select forged internals and upgraded input/tail shaft will do everything we need it to do.

 

Actually, high power F-body owners shear the transmission mount all the time, as well as the mount for the front of the torque arm.

It's precisely the inherent slop in the OEM Camaro/Firebird configuration that leads to a lot of the shift fork and other internal problems. There are far fewer shift fork issues on Corvettes, where the transmission is rigidly mounted between the engine and differential with a torque tube.

Replace the motor mounts and diff mounts with polyurethane or Nylon to eliminate the slop and put the OEM 5-speed behind a 20B and see how long it lives.

If they don't stay in perfect alignment? Yes. Even forged parts are destructible given the right circumstances.

 

OK, So make it stout. Gussy up the floor pan and build it to last.

That's an ideal design, but it is important to understand why. A properly mounted trans/IRS is as close as we're going to come to this ideal, solidly connected setup.

I don't this nylon bushings and poly mounts are a replacement for a solidly mounted trans and IRS. Explain how the casing will flex if neither the diff nor the engine are transmitting torsional force to it.... It's important to differentiate between casing flex and internal failure (ie broken shaft, gear etc).

I think it depends on the mount of flex, but ofcourse any material can fail given the right or WRONG circumstances.

 

With a solid mount? It's the rubber mount in the non-rigid OEM configuration that allows the movement of the transmission. Even poly mounts can shear, given enough force. If you don't want to live with vibration and noise, you're going to have to allow some slop in the connection of the drivetrain to the body.

If by that you mean eliminating the powerplant frame and more rigidly mounting the differential and transmission to the body with independent mounts, then yes.

What's this "solidly mounted trans and IRS"? Even the Corvette uses rubber bushings to mount the powertrain at the engine and differential. The only difference between the two systems is that the transmission is at the back of the drivetrain and that the torque tube is inherently more rigid than Mazda's PPF, but it's essentially the same concept and design; shifting is inherently quicker when the transmission is not moving excessively beneath the car.

No matter how rigidly you mount it, the force being applied to the gears inside the case may cause it to flex and potentially fail. That's precisely what happens to the cast iron differential housing.

It does not require outside force to break a transmission or differential housing or cause an internal failure.

 

That's right.

Agreed then

No it's not. The trans and diff are solidly mounted to one another so any torsional forces tranmitted by the engine via the tunnel of the torque tube are distributed uniformly through the diff and trans as they move in unisen. The FD diff and trans are constantly playing a game of twisting tug-a-war.

Are you sure about that? I'm fairly certain that the FD diff casings break because the front of the housing is getting cranked while the aft end remains stationary, not because of a climbing ring gear. Same thing goes for trans flex.

 

While I agree that the circular torque tube of the Corvette spreads the forces more evenly, it's done for the same reason that Mazda used a powerplant frame in the FD, so they serve the same purpose.

Excessive twisting movement in the drivetrain caused by shot motor mounts and/or differential bushings is what eventually fatigues and cracks or breaks the powerplant frame. Otherwise, it does its job just fine.

Yes.

What do you think is causing the stress on the nose of the housing? The pinion gear would rather come up and out of the ring gear than turn it and the only thing resisting that lateral force is the strength of the nose of the housing.

Not really. Any force applied to the tail end of the transmission by the powerplant frame and differential is not enough to flex the case of the transmission enough to allow gearing misalignments and internal part failure. The damage comes from inside, from the loads placed on the gears and bearing surfaces and the inabilty of the case to resist them.

 

That's exactly how it was described to me Rob.

 

The Mazda PPF is primarily for vertical alignment, not torsional rigidity. The torque tube/ transaxle and PPF configurations are far from the same; just because it doesn't always break doesn't mean it's particularly effective at maintaining torsional alignment. The Torque tube is in no way the same as the PPF, if anything the torque tube should be considered an extension of the trans case and input shaft; the PPF is a flexible coupling device.
IOW These are two completely different animals.

I think we're arguing black and white when the answer is grey. The truth of the matter is that the stress failures are caused by a combination of internal and external forces. To argue that the problem is exclusively one or the other is just plain oversimplification of the issue.

Interesting discussion regardless. Have a good one

 

Rob, I think Mazda would disagree w/ you

To quote Mazda:

"The pleasure of driving a sports car lies in the feeling of oneness between car and driver. It is a sensation felt only when the movement of the car immediately and precisely reflects the driver's input, such as the g-forces experienced when accelerating or decelerating. These characteristics are still important even though the new RX-7 is not intended to be a car that is merely fast in a straight line. Accurate response to delicate throttle openings is essential for enjoyable, controlled cornering.

Although our philosophy was to reduce weight down to the last gram, we accepted weight increases in some areas at the expense of the reductions made in hundreds of other parts. The Powerplant Frame is one such example and is bolted between the trailing edge of the transmission case and the leading edge of the differential case, thus rigidly connecting these two drivetrain components. This effectively transforms the engine, transmission and differential into one unit. The distance between the front and rear mountings has been expanded to 2150mm, so that when starting from rest the reactionary forces that act on the differential case are reduced and the torque is transmitted directly and linearly to the roadas traction. This process eliminates differential wind-up, resulting in transmission of torque that is free from lag and vibration. It is like having the engine, or more accurately the rear wheels, directly connected to the driver's right foot.
Another benefit of the adoption of the Powerplant Frame is the reduced pitching motion of the engine resulting in more precise feel to the gearbox. Furthermore, since there is no need for mounting points on the gearbox, the location of the engine and the floor pan could be lowered substantially. In the event of a crash, the Powerplant Frame is designed to crumple progressively to absorb the impact.

The Powerplant Frame offers many advantages, however, we did not accept the extra weight unconditionally. On the main frame, which is made of high-tensile steel, there are large apertures and by creating a closed section with the inner frame made of vibration-absorbing steel, the entire frame is exceedingly strong yet lightweight. The Torsen limited-slip differential is another weighty contribution which we employed to realize direct and accurate throttle control."

 

I stand corrected on their intentions, but doesn't mean I think it works ;o). Either way you got me... ;o/ lol