Torque Convertor Thrust on Flex Plate / Crank Shaft question

Swapping a newer 8 speed automatic onto an engine..

So as I understand it, there's a central "pin" on the torque convertor on the engine side, and that "pin" rests on the end of the crank shaft. This ensures the flex plate only sees rotational load, and doesn't see the face-load of the torque convertor being pushed away from the input shaft as the pressure builds. Because I have an unusually thick adaptor plate, this pin is very far from the crank, and the flex plate will see all the face load.

How much of an issue is this? Do I need to be worried about it, or is it inconsequential? This will see ~8,500rpm and ~500whp so I don't want to make a critical error here. The trans in OEM applications sees this without issue and a whole lot more torque so the flex plate as designed is solid. There are a few people using this conversion kit, but it's still really really new and hasn't been put through it's paces yet or tested for extended periods.

What says the hive mind? Will I be on borrowed time, or is this a dumb concern without merit?

I machined a hub adapter/extension for this last time I used an adapter plate. I would not run without the converter hub being lined up to/supported by the back of the crank personally. 

Sending you a message, hopefully you can help me get this done properly

messaged you back. I'd love to help but my equipment and skills are very basic. If anyone else here sees this who knows a proper machinist who could whip up a spacer from dimensions please speak up? I'd absolutely be down to help with logistics.

What torque converter automatic is happy with 8500rpm?  I'm interestedly curious.

I would be far more concerned about having nothing other than the bolts and the pump bushing keeping the converter centered.  Imagine, if you will, twenty pounds being rotated at thousands of revolutions per minute, only supported by a few 10mm bolts.  Especially if the bolt holes are elongated a bit (normal) and it is mounted just enough off center to set up a vibration.  I think you could measure the pumps lifespan in hours, at best.

Convertor is self-balancing to a degree, the fluid inside acts just like balancing beads inside a tire. But yes, definitely not ideal.

Trans is 8HP70, 8,500rpm max for 6th gear, lower limits for overdrive 7th and 8th. It's been pushed to 9k by some, but reliability is questionable there

The bolts don't locate the converter to the crank radially, the torque converter snout does. It has to be a close enough tolerance fit to positively locate the converter radially.

I'm not impressed with the transmission, it's the torque converter that tends to balloon at high RPM that generally limits automatics to 7000 or less.  Another reason why performance "automatics" are automated manual transmissions like DSG type or BMWs SMG type, no torque converter to swell up and cause weird issues.  If ZF claims that their converter can handle that engine speed, that's remarkable!  And a shame that it probably can't be retrofitted to a transmission that isn't the size of a bus 

mr2peak said:

Convertor is self-balancing to a degree, the fluid inside acts just like balancing beads inside a tire. But yes, definitely not ideal.

Trans is 8HP70, 8,500rpm max for 6th gear, lower limits for overdrive 7th and 8th. It's been pushed to 9k by some, but reliability is questionable there

You're confusing balance with run-out.

The TC could have a ton of run-out if it's not centered by the crank snout but no vibration because the fluid inside is dampening ("balancing") it.

This got a little long: rpms are fine, can use a clutch, dct is more difficult to implement. 


Legit concern. I will take a closer look at how it all fits together. This might end up being a bit more complicated than I had hoped.

The adapter goes from the BMW 8HP flex plate, and bolts to the crank, and that is concentrically located so I'm not too concerned about the run-out.

This trans, with the Turbo Lamik controller, can be run without a torque converter, using just the internal clutches of the trans and an electronic clutch just like a DBW throttle. So the end plan is to remove the TC, but I need to get the car moving and prove the combo works before taking that next more complicated step and making a replacement spacer for the TC.

RPM limit is from ZF's own specs, and is likely a bit conservative but I don't need to push past it. The 8hp trans in all flavors has been proven to hold a lot more power than their rating but I'm sure it lowers their lifetime. The Torque Converter itself has a clutch inside it, so you can choose what RPM you want to physically lock up the TC, no idea if that helps with the RPM capability but it's been proven to not be an issue.

Past 8,500 RPM you can use the DCT boxes, but reading and hearing about other peoples experiences they seem much harder to tune, the replacement clutches are very expensive and can burn out easily, the box weighs more, is larger, no overdrive gears, no skip shift and no options for a clutch pedal. A great box when it's working well and perhaps a better track box (but then a sequential is better so?), but the ZF8 will give me more of what I'm looking for at a cheaper price, lighter package, and shift damn near as quick once it's sorted out. I drove the M8 Comp with the 8hp and came away really impressed, acted just like the DCT in that application on sport mode settings but better drivability which is what I want in a mixed use car.

The TCC allows a fairly loose converter without having to have really short gears so the cruise RPM is higher than the stall speed.  That won't affect RPM, though, the strength of the converter does.

I've seen torque converters that could handle speeds like that, but the front plate was made out of 1/2" thick steel and the converter shell was extremely small in diameter.

I'd heard about going converterless but somehow thought that had been a 9HP thing.

BMW is using them in 7,200rpm applications stock, and I have no doubt ZF over-engineered them just like the rest of the trans. I want as tight a converter as possible, and I will lock the clutch inside the TC as soon as possible for a very positive throttle feel, none of that slush box converter BS.

BMW also ran them converterless in the M240i race cars, using different ratios. Called the 8P45RE, nobody knows all the changes that were made and the boxes are unobtanium.

Converter stall speed makes an enormous difference if you have an engine that comes up on cam later, or it has a turbo.  Also very important if you have an engine that makes very low manifold vacuum at idle unless you want a 1500rpm idle in neutral and two feet on the brake pedal to hold the car still at a light.

One of the most amazing things I did for my current car was the looser converter.  Turbo wakes up right away from a stop and after you get moving and the TCC is allowed to do its thing, it drives like stock.

If you have mild cams and plan on not being concerned about 0-60' acceleration, not a concern   Plus the ZF units have a much shorter 1st gear than most.

Most automatics seem to have a 7000rpm speed limit for converter ballooning safety.  It's the larger diameter, lower stall speed ones that seem to fare the worst.

Rotary with an EFR turbo, it spools plenty fast. Using a tight diesel converter, the point is to emulate the DCT boxes without the drawbacks. Might swap to a gas converter later on if it misbehaves but should be OK. Street port should hit full boost by 3k or so, probably launching in 2nd with the 4.10 diff.

Nobody has had issues with the converters ballooning, plenty of guys running them hard near and at the limit. Concern is pressure pump cavitating beyond 8,500, likely my party will be over by 8k but it's good to have a little extra, might get spicier in the future but it's already plenty of power for the chassis.

Ill be back in the shop in a few days, going to take another deeper look at the whole shebang now all the parts have arrived 

Got any pics of the adapter and crank end?

In reply to mr2peak :

I'm more than idly curious now.

My jam is bridge ports, but with a turbo, people only seem to rev them to 8000-8500 anyway.  The bridge would hate a tight converter, though, unless you played with tricks with two timing maps for neutral and drive.

Ford sort of did that with automatic equipped Boss 302s in 1970, they had a vacuum retard can on the distributor to make them work with an OE-streetable converter.

So a few things.

1) inner diameter of the adapter is off, it should snugly fit the end of the torque converter. So that is going to get a press fit sleeve from the manufacturer.

2) flex plates should flex. I'm a manual dude trans kinda dude, and learning about autos has sometimes been a bit of a struggle not being a 1/4 mile guy. So the face load is just standard normal stuff for an auto. But asking these questions pointed me to the TC run-out support so that's cool. The more you know!

It will be an extended street port, S4 low comp rotors. Twin 40mm straight gates on a shorty manifold, 8374 turbo, Rotary Revolution V Mount we are getting ready to produce for customers outside of Thailand. Bunch of cool stuff on it and around it. TC lock might be too low, but we can always swap out to one of the gas converters. The 8HP45 converters are 10mm shorter so if I want a super loose one we just plop in one from a 3 cylinder BMW (theoretically..). There's also a "clutch sleep" function to hold off on engaging the gear clutches until you are ready to drive so there's no pressure at all on the converter, but that has negative effects that go with it that I'd rather avoid.

Oh yeah, a thing I forgot that you reminded me of.  The stall speed is also ultimately a function of the engine's torque output.  I would expect that an OE converter in this day and age would have a fairly steep (for lack of a better term) stall speed point, but the upshot is with a turbo, if you are brake torquing it the engine has to make enough torque to get the speed up to where the converter is happy that it can build boost, which also builds torque and raises the stall speed further.

Stuff like this is why higher math was invented   Small converter changes (like raising the theoretical stall speed 200rpm) on the BGNs that I used to play with would increase boost at stall speed from 7psi to 9+ with no other changes, just because it coincided with the boost onset curve of those lazy old T3-60 turbos, which made the engine make a lot more torque at the higher boost, which was able to blow the converter higher.  That's just standing still, the difference when driving was much more dramatic.  'Course this is also with only four gears to play with, too!

Oh I know! We will see what happens. We are building this for good early boost, likely way too much for traction but hey that's mad fun and a great place to start!

Also, kudos to the adapter manufacturer, Domiworks, for taking care of this. When parts are new on the market, especially lower volume stuff, sometimes mistakes are missed. Once there's a mistake, it's all about how it's dealt with. I do very much like buying from people who stand behind their products!

And as always, thank you to the Hive Mind for pointing out the things I had missed.

What was their fix? Usually with an adapter or mid plate the converter mfg will just make one with an extended snout (typical th400/glide)

The inner diameter of the adapter is off. It should be smaller so the "pin" of the TC locates inside. They are going to press in m a bushing to take up the space. On future kits they will reduce the inner diameter of the adapter to snuggly on the TC "pin"

btw does that "pin" on the TC have a name?

I would call it a snout