icsamerica

I've recently used a Mazda RX7 clutch pressure plate as the base solution for a manual swap of an off topic car that was never available with a manual transmission from the factory. During this process I've come to learn the RX7 clutch has a lot in common with a 928 and it may be fertile ground as the basis for nearly unlimited clutch options for the 928. I've committed myself to developing a clutch for my own stroker project and I'm putting out an open invitation for collaborators to make this project open source and a benefit to everyone through information sharing. This is early stages, just an idea. At present I don't know what I don't know. Kind constructive feedback is welcomed.

Here are the pieces and what I presume to be true at the moment.

• Mazda RX7 is a well developed high performance platform with excellent aftermarket parts support. Selection and availability of pressure plates is considerable.
  Pressure plates from multiple EO and aftermarket sources such at ACT, Centerforce, McCloud, Sachs are readily available in various high performance configurations such as Stage 1 mild OE style to a full race 4 disk multi-plate setup and everything in between are readily available.
  
• Mazda RX8 and Porsche 928 share a 23 spline / 1 inch input shaft. This allows all clutch disk options that exist for the RX8 to be quickly ported over to the 928. And they are myriad. Everything from OE style to a Kevlar Race style 4 puck and many many options in between are readily available.
  
• Mazda RX7 pressure is size same as RX8 but RX8 has many clutch disk option. RX8 shaft size and spline count appears to be the same as the 928.
  
• As of now I'm aware of 7 stages for single disk clutch and pressure plate configurations that range from stock to 700HP.
  
• RX7/RX8 pressure plate is 9.449 inches (9-7/16), slightly smaller than the 928 S4 single plate (9-3/4) and somewhat larger than a early twin disk. Very good chance the RX7 plate will fit in the 928's housing.
  
• I've already successfully used a common LS hydraulic through out bearing to actuate two different RX7 pressure plates. This will allow the use of any available LS TOB option from OE to billet Race
  
• The RX clutch and pressure plate are highly available and seem to be priced very reasonably. I can foresee a clutch that's configured to hold up to 500Hp costing about 650$ with OE levels of effort and drivability
  
• Many aftermarket providers are willing to make RX7 based parts custom. For example, in my OT case I was able to get RX7 clutch disk with a custom spline made quickly by ACT for a reasonable price.
  This is goods news for exotic setups but customization of the clutch disk doesn't seem necessary for the 928 in most cases.

Parts that need to be Developed
1. Flywheel, - 2 options, completely new and re-drilling of existing. Both should be viable options.
2. Hydraulic TOB sleeve and mount. Should be an easy "Send Cut Send" project.
3. hydraulic line package. Should be a basic M10 x1.0 to whatever the TOB is using readily available AN3 lines in between

Plan and Collaboration
The plan is to develop the pieces and make them widely available with their specs and dimensions shared and public. This way, each 928'er can have a clutch that meets their needs and performance goals.
Once the "RX7 pressure plate bolted to a 928 flywheel issue" is resolved the rest of the project should be a straight forward with the help of a standardized TOB and sleeve development.

At this point this is all an idea. I'm inviting collaboration and constructive remarks that are kind and can help develop and bring this idea to fruition. Fruition being a sets of parts (flywheel, TOB mount and line adapters) that will allow a 928 to have many many performance clutch options by selecting readily available RX7 parts and at reasonable cost. My value add in this regard is that I'm extremely focused on machine / man interfaces so clutch feel, weight and engagement character through proper setup and TOB selection is something I can quantify and focus on. I have manual swapped many cars that never were available with manual transmissions and have developed some knowledge and skills in this regard.

Some ideas for collaboration....
At the moment I could use a OE flywheel, bell housing and intermediate shaft in any condition to do mockup or someone to do the same and provide feedback
Anyone willing to beta testing and put in some effort at problem solving fitment and feed back.
Someone with experience with send-cut-send or similar low volume prototyping service
Someone with mechanical drawing skills in order or help with documenting and publishing dimension of parts and specs for re-drilling of flywheel.
Someone with a lathe or mill to center and redrill an EO flywheel, possibly in volume
I have a Centerforce RX7 pressure plate rated to 450 ft/lbs to develop and test on.
I have no idea who can make Porsche Flywheels, a qualified contact will be helpful.

Being involved in software development has taught me that goal based collaboration can produce far greater results with valuable inputs from qualified participants. This can result in exponential improvements in end point functionality and quality. I do believe it's possible to apply this paradigm to the 928's future developments. Doing so can improve and extend the viability of the platform from a performance, customization, usability and enjoyment perspective. In the past many individuals in the 928 community have endeavored to create and resell solutions for profit or as a service to a community, That is amazing work and many great products have been developed using this paradigm. I'm suggesting a different and additional paradigm that puts collaboration and open source values at the center for the benefit of all. If the information and techniques are open source it will allow better outcomes for all involved. More profit for those that can and are willing to do the work, more options for those who want the work done. Both savings and options for those who can and want to DIY. This can lead to better outcomes for all involved and elevate the platform, in this case the 928. Over time the open source nature will allow developments from all involved to feed back as revisions and updates to what came before.

Below is a sample of what available for the RX7/8



Many clutch disks are readily available at a low price.


Many many option for pressure plates.

Alan 91 C2

Very interesting approach. Agree with your premise, for parallel users problem solving.
Alan.

Mrmerlin

I have a few questions,
does the Mazda clutch PP release pushing towards the engine or pulling away from the engine like the 928.
If pushing towards the engine then it will be critical to have a thrust bearing check preformed and monitored ,
as then both sides of the thrust bearing will become worn on a 5 speed engine.
I presume this checking will probably only be necessary till baselines are established for wear patterns.

Would a hydraulic TOB be installed in place of the original guide tube?
are the guide tube mounting arms strong enough to support a hydraulic TOB?
If so then this removes the release arm and bushing and TOB and no slave. Cool
those are a few off the top my head.

hans14914

Just a couple thoughts, since I was just looking into this myself (had a clutch slave seal fail on the road to Third Coast this year and it saturated the disc - made it for a fun drive there and back with no holding torque).

The factory 928 is a pull-style clutch. The clutch you propose is a push-style.

Pull-styles actually have more clamping force with less pedal pressure. Additionally (for whatever reason) it seems as though the 928 engine is susceptible to thrust bearing failure when pressure is exerted on the rear of the crank pushing towards the front of the engine. It may be prudent to select a clutch that retains the pull-style system.

This unfortunately does eliminate the practicality of a hydraulic release bearing. However, it does open the door to a more common release bearing when machining a new guide tube and fork. The upper pivot could be changed to a monoball with its redesign. Similarly, a clutch slave replacement could be found to make that a more common item.

Clutchmasters makes a nice 8.5" clutch system with a sprung drive hub and strapped plate in their 850 series. ACT has a similar product line with the Mod Twin 225.

A single clutch disc may be able to hold the torque you are asking, but the pressure plate is going to be very heavy. I think maybe using a modern twin-disc pull system with a designed for purpose flywheel may be the best option if you are going through the complete design exercise.

Just a couple personal thoughts since I have been noodling this for personal reasons the past two months.

Good luck with your project!

icsamerica

Mazda is a push.
Pull or push you have the same force on the center bearing albeit in the opposite direction when using a push style clutch release. Since the 928 center bearing is faced on both sides I see no reason for concern by changing the direction of force. If a 928 has lash in the in the CB and don't see how changing direction could made that progression any worse. It's a little off topic but on all my manual swap projects remove the clutch starter switch and move it to the transmission so the trans has to be in N and clutch out to start. This prevent the CB from being worn with the clutch in on start up when the CB is likely to have dripped dry. . I always start in N with clutch out and I'm old school in this regard, stick get a shake check before turning the key.

As for the strength of the bell housing casting, it's a valid concern. I'm not certain at the moment but I do believe a re-enforcement brace, plate or something in-kind would be prudent to add that runs from the guide tube mount to the torque tube bolts. This will certainly need a closer look.

icsamerica

Any thoughts on why the direction of force would matter to the Center bearing? My understanding at the moment is the center bearing is full faced in both directions. Maybe those that have raced the 928 with high HP strokers that had clutches that could hold can chime in and indicate whether they used push or pull?

I think the main concern with a Hydraulic TOB bearing is they can have over 1.5 inches of travel and if they pump up and don't fully release they can over drive the clutch past the release point and force the crank into the CB, destroying both and possibly the block. This will certainly need to be accounted for with proper TOB selection or a physical stop. If the correct TOB is selected with the right volume and location it will be impossible to over drive it using the 928 OE master cylinder. On my OT projects I use a RAM Clutches Hydraulic TOB that if over extended will fail thus saving the engine's center bearing. The stock LS hydraulic TOB will do the same. Ask me how I know. LOL

As for push pull pedal force....I have noticed a similar phenomenon. For example, I built a 408CI small block Chevy that produced about 500Ft/lbs of torque using an uprated LT1 pull style clutch. The clutch pedal effort was surprisingly light, nearly gas pedal light. It held up to DE track duty too. It's my understanding that Pull style's are easier to release becasue the pivot point of the fingers allow for better mechanical advantage for release but the clamping force on the pressure plate is the same so the opposite release force on the crank center bearing is also the same. If not the same type of clutch disk wont hold. That my understanding at the moment.

After my LT1 pull style experience I was surprised the 928 clutch was pull also becasue it's very heavy by direct comparison and I dont think an OE 928 clutch will hold 500 Ft/lbs of torque.

I just finished up an OT project using a mazda PP that's rated for 440 ft/lbs when using an OE organic full face clutch disk with springs and nacells. It's driven by and LS through out bearing. In practice pedal effort feels about the same as the OE 928 twin disk. The OT project has no helper spring on the pedal like the 928 does and the feel and modulation are very good. I've never loved the feel or clarity on the 928 clutch... it's Ok.... feels dated when compared to a modern Hydraulic TOB based system so I'm looking forward to using the same PP and disk on my 928 stoker project. If had the same engagement smoothness and pedal effort as the OT project on the 928 I'd be thrilled!

Zirconocene

I have to admit that I'm totally out of my depth with trying to help with what you're looking for, but if moral support helps, I've got you.

I hope people chime in and help move this effort forward. If nothing else, I love reading and learning about these kinds of things and, who knows, this may be something I do in the future.

Good stuff!

Mrmerlin

the reason for discussing the thrust bearing is that if its worn on the front,
then the rear side is used it might wear down enough to a total play to be out of spec.
This could be a moot point but I think its worth considering

hans14914

Converting to a push-style with a hydraulic release bearing is a fairly straightforward task with a matching flywheel. I think you need a couple design parameters in the project, such as cost, holding torque, pedal weight, rotating mass, spare part availability... etc.

Greg already has a clutch system based on the Tilton ST246 (which is a great selection), so there is already an available solution, at a fair but "expensive" price in the ~$4000 range (keep in mind that is a complete solution with bearing and flywheel). A Spec clutch is in the ~$1200 range. Are you trying to fit between those two points, or under the cost of a Spec clutch? I think having some established parameters will really help keep your project on track.

If you are going the route of a hydraulic bearing, its not as widely known, but Sachs has an aftermarket division, and they make bearings (and loose clutch components). Kind of tough to beat someone with that much experience in the OE market. Again, not cheap, but they arguably know more about clutch design than any other company on the planet.

Racing Clutch CSC / Releaser | SACHS Race Engineering (sachsperformance.com)

Again, wishing you success on the project. I will be following along as I figure out my own setup.

icsamerica

This is a open source project. I'm not try to fit to any price point but instead open up a variety of options. The end user can select the option and specific parts that meet their goals.

The key aspect are as follows...
1 Select a proper TOB, that wont over drive the clutch and damage the engine or pressure plate.
2. Develop a mounting sleeve for the TOB, and document design specs.
3. Document flywheel re drilling or source new replacement options

Once those three aspect are reached and put forth into the public domain individuals can select from the myriad of Mazda RX based pressure plates and clutch disks that meet their needs.
For example a stage 0... A pressure plate and OE clutch disk may only cost 200$ and hold 250 ft/lbs, then and up-rated aftermarket pressure plate for a Stage 2 using an EO disk, good to 350Ft/lbs that will cost about 350$, Then State 3 with a pressure plate that can hold 440 ft/lbs will add up to about $450, Subsequent stages for example, Stage X will could be multi-plate version that is light weight and can hold 1000 HP. Yes it will cost more but all varients will use the items developed in in aspects 1,2 and 3. All stages using readily available RX7/8 items that are well developed, provide serviceability, an upgrade path and will be around for a long time is a big value add.

I do believe these Mazda RX items will have good selection and availability for a long time since the RX7/8 has a cult following and many of these same items are shared with some Ford and other popular Mitsubishi tuner platforms.

Alan 91 C2

My look at the TBF issues indicate the reason for the failure to be continued pressure on the CB. The continuous pressure squeezes out the oil, from the bearing side surface Every standard transmission in use, is an intermittent loading for the CB, with little failure reported, .

GregBBRD

Currently <$3500, for the dual disc version, as I had multiple flywheels made (which is, by far, the most expensive part, including the clutch price.)
Effortlessly holds 800ft.lbs. and is smoother than a stock clutch. Starting from a stop, uphill, in 3rd gear (with an engine that has enough torque to do that) is simple...zero slippage required, with zero chatter. Just drives away!
This also includes a custom 1.125" clutch short shaft (made from 300M), to eliminate the smaller (stock dual disc) clutch disc splines and shaft, which wear rapidly, with over 450ft.lbs of torque.
And is a complete kit, from T/O bearing and mount, flywheel seal, hardware, lubrication, right down to the custom hoses for the hydraulics.
The other benefit is that I'm fairly certain that this clutch will never wear out, in the 928 application.
Jim Corenman put 5,000 miles on one (5.9 stroker engine) and the discs were still only 1/2 way from having 100% contact (zero wear.)

Incidently, that price is over $1500 cheaper than the same stock Porsche replacement parts, but certainly more than a Spec clutch capable of holding 500ft.lbs., which will shake your teeth loose.

I'm 100% certain that a replacement clutch can be made cheaper [for example, we are testing a single disc version of our dual disc clutch (cheaper), for "tamer" applications],
but doing it close to being equal or better than my flywheel/clutch, for less money, is going to be a really tall order.
After 500 really hard track miles behind a 700hp stroker engine, the discs on my flywheel/clutch assembly are still not completely worn in, on the "An Extremely Extreme GTS" vehicle.
And that clutch still releases smoother that a stock 928 clutch!
https://rennlist.com/forums/928-foru...treme-gts.html.

icsamerica

That's a good read of the situation in my view. That said the most critical part of this project is to find a TOB and locate it in such away that cant be over driven. Based on what I know so far, I dont see that as all that daunting.

Darklands

The S2 and S3 Clutches have 100 teeth on the intermediate plate. This clutches can't easily converted to later one disc solutions and are really expensive as Greg pointed out.

icsamerica

I think your are referring to the trigger wheel for the EZK? That's a good send-cut-send project.