Nissan Juke : Juke Forums banner

281 - 300 of 319 Posts

·
Registered
Joined
·
1,182 Posts
Discussion Starter #281
@pboglio

*Correcting the sec pressure with a fixed linear offset will be incorrect as the solenoid pressure vs current profile is not linear
"No, the pressure vs signal output is linear, I have the map. I'm also not using a fixed linear offset, it's a scaling factor using a voltage trigger value to activate the circuit. Hopefully that's clear, I used a comparator to analyze the input signal."

*The maximum clamping is indeed still limited by the system pressure, a stiffer spring will not bring anything as the spring force can be neglected compared to the hydraulic clamping force, however the system might still have some margin.
"I have sufficient oil pump pressure available anyway (4-5.5 Mpa). I rarely operate above 3-3.5 Mpa at the Secondary Pulley, thus there is sufficient pressure available to still drive all hydraulic circuits.
Not sure I fully understand you, but the stiffer spring within the valve body is to increase the pulley pressure. I have never done the modification so I can't verify exactly what spring and what metering orifice in the valve body sealing plates were modified. Normally I'd say the ECM/TCM would simply compensate for the increased spring rate by altering the control solenoids duty cycle, but I can't comment beyond that. Somehow Level10 made a CVT hold some pretty serious torque.

*Overwriting the sec pressure will potentially decrease ratio stability as the primary pressure should follow the higher secondary pressure, but the system does not know this hence it will need to rely on the ratio PID to compensate for this ratio as well but only very limited.
"Not sure I agree with this. Without the PID control, the car won't get out of park. It's continuously relying on feedback control. The only thing I'm doing differently is forcing the ECM/TCM to trim another 5-7% from it's expected settings, most likely it won't throw a TCM code and trim the solenoids accordingly.

*Increasing the input torque and compensating the sec pulley pressure for this might maybe work, but will you do the same for the drive/reverse clutch and TC? Probably the TC control both controls the switching from TC mode to lockup mode and hence simply increase or decreasing the current is unlikely to work. I also assume the adaptive control on the TC only works for the lockup clutch biting point and does not impact the TC switching current.
"Good questions. No, I'm not modifying the drive/reverse clutch pressures or TC. There is nothing to be gained on the Torque converter if a good CVT fluid like AMSOIL is used, it has a generous sized clutch. If you read my build, I modified the forward clutch packs to handle something like 320 lb-ft of torque with the same clamping pressure. I'm good to go."

*fatigue of the pushbelt and variator shafts might become a problem due to the higher clamping.
True. The higher clamping is a necessary evil. Without it, the belts would slip and wear out rapidly. Did you know the factory already adds 20% safety margin on pulley pressure to account for pulley wear? I'm simply adding another 5-7% pressure on top of that and only at WOT or a certain torque value where I would expect some slippage might occur.

*Increasing the rev limit will lead to higher centrifugal forces on the prim cylinder parts, this in combination with higher clamping pressure as a result from the reaction (through PID) on the increased sec pressure might overload the pulley pressure cilinder parts.
"Yes, increased RPMs tends to make the belt slip. Increased torque breaks parts.......and increases slippage. Which is better? This is precisely why I designed my CVT to be prevent slippage, because higher rpms doesn't break rotating parts like torque does.

I came accros this thread by accident, nice project! Would you happen to have a re-build manual which you constructed yourself or a mapping of system pressures vs solenoid currents? Basically any information would be of interest to me. What oil do you use? Where do you source parts, website?

Regards,
Xander
Xander,

Thanks. I do have a couple rebuild manuals I can point you to. Yes, the Pressure vs sensor output maps exist, I'll look for it. I use AMSOIL CVT fluid. I have too many sources to list here. I'll see if I can dig up all the vendors I used.

Gene
 

·
Registered
Joined
·
4 Posts
Xander,

Thanks. I do have a couple rebuild manuals I can point you to. Yes, the Pressure vs sensor output maps exist, I'll look for it. I use AMSOIL CVT fluid. I have too many sources to list here. I'll see if I can dig up all the vendors I used.

Gene
I would be very much interested, I have send you a private message with my mail adres if needed if the manuals or other info would be too big.

Keep up the good work!
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #283 (Edited)
To follow up on some of the questions Xander asked.

The fact of the matter is, it's nearly impossible to predict what the TCM will do without on-vehicle testing if changes are made. The TCM is very adaptable and something really has to be out of wack for the transmission to not work correctly, the PID control is that good. The TCM uses a CANBUS communication system for some of the non-mission critical communication, but it relies heavily on engine rpm, throttle position, input torque, gear selector position sensor, primary pulley position, secondary pulley pressure, input shaft speed, secondary pulley speed, wheel speed sensors, oil temperature, and so on.

This adaptability is both good and bad, makes it harder to trick it. I've done a ton of datalogging and also built a computer model to predict what types of pressures I should be seeing for varying input torque, gear ratios, fluid friction coefficient, etc. While I don't have the actual ROM maps of the CVT, I essentially have mapped the actual outputs under the driving conditions I'm concerned with, the computer model uses that data. This is the only way I could build a circuit to fool the TCM and double check the outputs. The ability to shift a signal 2%,3% or 8% is a good way to sneak up on the TCM to avoid it throwing a fault. On these transmissions, even a +8% extra clamping force is like +20 lb-ft extra of torque capacity. Throw in some Amsoil CVT fluid and add another +20-25 lb-ft capacity. So together, it'd push the slippage point another 40-45 lb-ft higher on the belt.

Regarding electronic pressure control, I've modeled my mod with what Sonnax are doing. The pressure control circuit is similar to what they offer on automatic transmissions. Sonnax didn't divulge how their control circuit worked........so I designed my own from scratch. It's basically a piggyback circuit anyway. There are valve-body modifications that can be done, but it requires a deep knowledge and understanding of the hydraulic circuits and a LOT of trial and error. I chose to keep it simple and reliable for this round.

As I've mentioned already, the factory Bosch belt is good to 350 N-m if it's got solid pressure control from the TCM. I'll repeat, the belts can TAKE the extra pressure, the belt elements can't handle ANY signficant slippage, though they can take some extra clamping force to a point. I won't go into detail about the life testing that Bosch does on their belts, but the max overload tests are brutal and 350 N-m is a conservative rating. I should probably link this, but Hondata did some amazing R&D testing of the Civic CVT and that vehicle run's the badass Bosch 901074 (12-band, F7 miraging steel, high inertia belt-element), rated for something like +400 N-m. Well, surprise surprise they could not come close to snapping a belt even on a road course, but the belt elements wore out immediately once they kicked the torque up and well below the belt rating I might add. All because Honda don't run enough clamping pressure when they really would have benefited from it.

The main problem for us is belt element degradation over time that is below the 350 N-m point, and that is a real world problem for belt life and CVT life. Those few guys running modded Juke's, the CVT don't stand a chance. Those that thought otherwise.............they found out the hard way. Personally, 240 lb-ft torque maximum with AMSOIL fluid and a great CVT cooler + religious oil/filter changes.....done. In the real world, the CVT fluid degrades, people run the CVT hard when the transmission is cold and there is major belt slippage, CVT fluid overheat anneals/softens the belt elements, people under-fill and there is belt slippage, valvebody wears out and there is belt slippage, oil pump flow control valve fails and there is belt slippage, over extended oil changes wear out the belt due to solid material contamination, etc. The factory build in a 20% reserve of pressure to deal with these derating factors, because it's the what actually happens. Unfortunately, it isn't enough and the belt goes to 0% reserve........and then it starts micro-slipping. A modded CVT will slip immediately at the 400 N-m point, and that is very bad for the belt for longterm durability. Bosch CVT belts don't really catastrophically fail anymore, they just wear down and start slipping like mad. I have the S-N curves somewhere, but trust me, the belt bands are mega strong. The old F5 or F6 generation belts used to snap, but Bosch have gotten these new F7 generation miraging steel belts to be almost impervious to cycling failure as long as the 350 N-m limit is observed. Basically, these belts need some extra pressure to keep the slippage in check.

Regarding my transmission build, all of the strength/reliability mods for the most part are either available commercially in kits, or I custom design myself from what performance automatic transmission shops are doing on racing/performance transmissions. Again, the reason for this is very simple: I have a limited R&D budget and I get only (1) chance to get it right. Any modifications must work and work reliably for 100,000 miles. In fact, you will find my modifications are actually mostly fixes to the factory design flaws, with some of my own custom fixes thrown in for good measure. I will be putting this to the test in the next 6-8 months.

An important point here: Not all the commercial available CVT kit items work, and in fact I list which ones would actually cause a CVT to fail if they were installed. Please read CAREFULLY my CVT build and you'll see the pitfalls. I can give you all the part numbers I bought for the rebuild, but my build thread shows what aftermarket parts I used, what aftermarket parts I didn't use, what OE factory parts I re-used, and what parts I modified. I'll state right here that any of the aftermarket teflon sealing rings for the "pulley sheaves" are junk, they lack the "2-direction" overlapping gap ring design the OE factory uses that produces a perfect air tight seal. The factory OE teflon pulley sheave sealing rings can be reused if they measure close to the 2.0mm thick spec (from memory) by flipping them over 180* and given a good cleanup. All the smaller aftermarket teflon sealing rings in the kits are excellent quality and match the OE design and seal perfectly. Like I said, you could not slap one of these CVT together off a build list and have it work, it would actually not work. Having said that, most will agree that the CVT is mechanically actually a pretty simple but deceptively dangerous rebuild without all the tips/tricks known. The CVT is very similar to a standard automatic transmission in many ways, and unique in others.

I'm reiterating these points because the mention of ratio control stability was brought up. The PID is sophisticated but adaptable and quite reliable when things aren't all worn out mechanically.
 

·
Registered
Joined
·
4 Posts
Thanks pboglio. I am really amazed by the amount of information you have been digging up, thanks. In what format is that computer model? In matlab?
I would also like to jump into a detail: using other CVT fluid with beter friction properties might indeed improve the torque capacity of the CVT, the lifetime however can be drastically reduced due to the internal friction contacts between components of the belt not transmitting the load in a correct way. I do not know this specific oil, I am just stating it's a real risk. If things would not work out, increase clamping further and use the OEM oils instead.
 

·
Registered
Joined
·
4 Posts
If I am correct you are not changing anything on the torque converter and you assume the lockup clutch can take the higher torque of the engine (after it learns again) ? Correct?
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #286
Spreadsheet and solidworks for the model. I suppose I could use matlab.

No, the Amsoil CVT oil testing was conducted on (2) vehicles to 100,000 miles and belt, pulley wear compared against standard NS2 type oil, no comparison. Real world the difference in traction is night and day, I ran it when the transmission was already dying to extend its life. The difference in traction on the belt and clutches is immense. When I would slip around 250 n-m, the amsoil pushed that to 310-330 n-m.

I have a remanufactured torque converter. Mine supposedly runs an aftermarket carbon clutch disc, better than the cellulose discs in terms of maintaining a stable torque capacity, but considered a factory style replacement for the reman market. The factory torque converter is sufficient if the fluid friction coefficient is good and it’s not brake torqued with a good cooler. There are upgrades to the torque converter clutch that I was going to make using a carbon Kevlar clutch disc from Exedy, but I didn’t want to risk it with the rebuilder as it’s tough to find a reliable shop. Either way, the torque converter never slipped on my car, it was actually the forward clutches and the CVT belt.
 

·
Registered
Joined
·
29 Posts
Spreadsheet and solidworks for the model. I suppose I could use matlab.

No, the Amsoil CVT oil testing was conducted on (2) vehicles to 100,000 miles and belt, pulley wear compared against standard NS2 type oil, no comparison. Real world the difference in traction is night and day, I ran it when the transmission was already dying to extend its life. The difference in traction on the belt and clutches is immense. When I would slip around 250 n-m, the amsoil pushed that to 310-330 n-m.

I have a remanufactured torque converter. Mine supposedly runs an aftermarket carbon clutch disc, better than the cellulose discs in terms of maintaining a stable torque capacity, but considered a factory style replacement for the reman market. The factory torque converter is sufficient if the fluid friction coefficient is good and it’s not brake torqued with a good cooler. There are upgrades to the torque converter clutch that I was going to make using a carbon Kevlar clutch disc from Exedy, but I didn’t want to risk it with the rebuilder as it’s tough to find a reliable shop. Either way, the torque converter never slipped on my car, it was actually the forward clutches and the CVT belt.
thanks for this whole thread. im in the middle of pulling a motor myself and have found all of this so helpful! you really are a beast!!!
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #288
Glad you enjoyed the build. The engine removal wasn’t so bad actually. I bought storage racks to put the externl parts on, then for the motor I bought a (3) level cart that had all the internal engine parts stored neatly. Engine/trans unit is pretty easy to pull out once the upper/lower radiator supports are removed.

It’s just patience and taking your time. Start an engine build thread.
 

·
Registered
Joined
·
29 Posts
Glad you enjoyed the build. The engine removal wasn’t so bad actually. I bought storage racks to put the externl parts on, then for the motor I bought a (3) level cart that had all the internal engine parts stored neatly. Engine/trans unit is pretty easy to pull out once the upper/lower radiator supports are removed.

It’s just patience and taking your time. Start an engine build thread.
I would but im not doing anything close to what you did. just replacing all the gaskets, a head rebuild, timing chain and tensioner (if needed). no up grades and as few part replacements as possible. doing it for my sister who works at petsmart and has no money. I probably could have done it all with the motor in the car but I have bad arthritis and absolutely hate tight spaces. also could have done a motor swap with a salvage motor but you never know exactly what you are getting. this a away I know all the gaskets are fresh. the head and valves are rebuilt and by my math going to cost less. I guess I could start a thread, just wont be a whole lot to put.
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #290
Sounds like a good plan. I always wanted to rebuild the motor while I had space so I can dial in the cam timing, and definitely having it out helps tremendously. The engine rebuild kit is approx. $147, they come with new cylinder valve guide seals, so you can save money on the head rebuild if you give them to the guy from the kit. I'd highly recommend a brand new RS oil pump (P/N 15010-BV81A) for $143. New water pump (B1010-1KC0B) is about $58. Double check P/N for your make/model year though cause there are differences.

Are you planning on pulling the pistons/rods out of the block for inspection? If so, carefully check the cylinder bore diameter and maybe get some new piston rings and just lightly re-hone the block & a fresh cylinder head deck cut. I really didn't see much wear in the cylinder bores when I measure them vs. the cylinder block bore I.D. stamp codes. Maybe a few tenths wear and well within spec, just don't mix up the pistons since they are matched. If you're in there you might as well replace the main bearings & rod bearings & rod bolts, they are pretty cheap.

It's definitely possible to get a bang-for-buck rebuild if you target the right areas.
 

·
Registered
Joined
·
29 Posts
Sounds like a good plan. I always wanted to rebuild the motor while I had space so I can dial in the cam timing, and definitely having it out helps tremendously. The engine rebuild kit is approx. $147, they come with new cylinder valve guide seals, so you can save money on the head rebuild if you give them to the guy from the kit. I'd highly recommend a brand new RS oil pump (P/N 15010-BV81A) for $143. New water pump (B1010-1KC0B) is about $58. Double check P/N for your make/model year though cause there are differences.

Are you planning on pulling the pistons/rods out of the block for inspection? If so, carefully check the cylinder bore diameter and maybe get some new piston rings and just lightly re-hone the block & a fresh cylinder head deck cut. I really didn't see much wear in the cylinder bores when I measure them vs. the cylinder block bore I.D. stamp codes. Maybe a few tenths wear and well within spec, just don't mix up the pistons since they are matched. If you're in there you might as well replace the main bearings & rod bearings & rod bolts, they are pretty cheap.

It's definitely possible to get a bang-for-buck rebuild if you target the right areas.
thanks. I think as long as everything goes as planned, this should work out great.
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #292 (Edited)
Sounds good(y)
 

·
Premium Member
Joined
·
2,200 Posts
If you're a corporate shill for PetSmart posing as a Juke owner, you're doing it wrong.

I would but im not doing anything close to what you did. just replacing all the gaskets, a head rebuild, timing chain and tensioner (if needed). no up grades and as few part replacements as possible. doing it for my sister who works at petsmart and has no money. I probably could have done it all with the motor in the car but I have bad arthritis and absolutely hate tight spaces. also could have done a motor swap with a salvage motor but you never know exactly what you are getting. this a away I know all the gaskets are fresh. the head and valves are rebuilt and by my math going to cost less. I guess I could start a thread, just wont be a whole lot to put.
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #295 (Edited)
Long time since the last CVT update.

Oil pump replacement:
Well, looks like I need to pull the CVT oil pump back out again and toss it in the trash. I found out there's an issue with the Transgo oil pump relief valve design causing some very high pitch whine and it's a very common problem on the JF011E with this aftermarket valve, especially on the Nissan Rogue and Jeep Patriot rebuilds. Sonnax have the improved relief valve design with the hard anodize coating and it's proven to work perfectly on an these OEM setups. The pump I'm buying is machined oversized for the Sonnax valve and installed ready to go as a reman unit. This crap drives me insane but since there are so few people who have actually done this stuff, it's whatever works that matters. The JF011E oil pumps are fairly generic, so it should drop right in after I inspect it and upgrade the pump cover bolts with high strength alloy bolts. It never ends man, I've had that tranny assembled/disassembled so many times I've lost track. The pump removal is actually pretty quick with the oil pan, valvebody, and main case removal it drops out with 4 bolts, no huge deal. Dropping a tranny off the car......not so easy.....lol.

Further Upgrades:
I have been going over and over the Juke CVT transmission design and have come to the same conclusions: My upgrades can handle the torque without slipping, but the forward clutch drum can't handle the stress without fracturing. I keep coming back to that same design limitation. Based on this, I've been crunching the numbers and I've settled on a power output of 320-340 h.p. and 257 lb-ft @ 6500-6900 rpms. This 257 lb-ft is the max reliable limit of the Bosch CVT pushbelt.........that's it. The forward clutch drum can't even handle that torque reliably for very long, so something else needs to be done.

So, to deal with this I've finalized my forward clutch drum reinforcement insert design for the JF011E. I've done extensive FEA analysis and finally have a simple solution to the problem. This part will be made out of 300M VAR high strength steel (i.e. vanadium/nickel), the strongest steel my machine shop can cut. I'm currently running the calculations on the strength improvement with additional heat treatment, but it's about +20% stronger. However, this part is going to require me to machine oversize the primary pulley spline shaft I.D. that the forward drum sits inside of, about +.040" I.D. to a depth of about 0.250". All this while the transmission is assembled. There is very limited material I can machine without weakening the primary pulley shaft, so I have to be careful here. Honing will allow me to cut thru the shaft I.D. bore slowly but very accurately. This is critical because in park/neutral the input shaft/forward drum and primary shafts are disconnected and spinning at a high relative speed to each other, proper gap clearance is absolutely crucial. This'll then create a clearance pocket for this slip fit reinforcement ring to sit within. Finally, the reinforcement ring will also be honed and matched fit for a light slip fit onto the forward clutch drum, so as not to stress the spline hub during install. Finally, I'll use bearing retention epoxy to bond/glue the parts together so they can transfer loads effectively across these separate parts.

My machine shop will be quoting maybe (4-5) total parts in case I decide to build future transmissions for myself or anyone else........if all goes well that is. He did a great job on my custom 4130 steel Intake/Exhaust valve spring seats, so this part will be cake. I'll post up some pics of the 3D-design in the transmission assembly. I'll be working on the engineering drawings and the tooling needed for the precision honing over the next few weeks. I'm thinking 4 weeks for machining and maybe 2 weeks for final fitment/clearance checks/reassembly in the transmission.
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #296 (Edited)
OK, gonna be another long one but a good recap.

Latest update:
I just ordered a brand new factory/dealer available JATCO CVT oil pump from WIT transmissions (P/N D212500A), (Pump, RE0F10A/RE0F10B) for $292.13. The Sonnax oil pump relief valve hard anodized (P/N 33510-N02, standard bore) was also ordered for $28.12. This upgraded pump piston valve will swap right into the new OEM pump without modification, 2 minute job. They make an oversize version for use on worn pumps but it requires some reaming/machining on the old pump......no thanks. As I mentioned way back, this will prevent/eliminate early failure of the CVT belt due to insufficient oil pump pressure feeding the pulleys, a sure killer of most failed JF011E transmissions. Sonnax know their stuff and are proven, they are recommended by mechanics rebuilding the JF011E based on excellent field results on customer vehicles. The Transgo oil pump relief valve is proven junk, I'm glad I caught it before the install, too bad I didn't know about it until recently.

I put the list below for those members who asked me if my CVT transmission is modified/upgraded. Yes, it's most definitely upgraded. I'll backfill the missing P/N and suppliers, this was just a quick rundown and the suppliers and parts are always getting updated and new releases available (i.e. oil pump, etc.).

Will it hold 500 w.h.p.,............no it won't. The horsepower is a silly rating for a transmission. Torque and rpm are typically used. Where I can, I list specifically what the component torque ratings are. I did this for (1) main reason: The weakest link in the chain breaks the chain. The factory CVT pushbelt is probably way underrated but Bosch are very conservative, the rest are highly accurate engineering calculations based on clutch disc capacity, clutch diameter, clutch clamping pressure, fluid coefficient of friction, or FEA analysis. What instantly broke my transmission..........a $.30 snap ring on the input shaft. Everything else was mainly collatoral damage, except where noted. There were many failures, but this was catastrophic.

It'll hold more than a stock transmission would, and last longer.......that's the goal anyway. Every part I replaced was a fix to a known failure mode on my transmission or common to all JF011E. Had I to do it over again, I'd start with a factory fresh new CVT with zero mileage but this was done to prove a point that these CVT transmissions could be rebuilt. Anyway, what the CVT needed were some well chosen factory/aftermarket upgrades to bullet proof it for a mildly upgraded Juke.

To recap on the transmission status:

Brand new OEM CVT factory parts:
new Nissan/Jatco JF011E CVT oil pump (KA6X0623E9)
new Nissan/Jatco Valvebody (solenoids, pressure sensor, ROM pack, etc.)
new Bosch CVT pushbelt (901083) factory rated at 250-350 N-m, (original is older revision 901066) (original belt worn out from brake torque, abuse, lack of maintenance, high engine torque, etc.)
new Transtar CVT master rebuild kit (teflon piston seals, oil lip seals, o-rings, etc.) (P/N xxxxxx)
new Exedy forward/rear clutch packs & steels (OEM supplier) (Transtar kit)
new SAP blue bonded forward piston (reverse piston didn't fit) (Transtar kit)
new NSK roller ball cartridge bearings ((Sonnax 33235N, 33236N, 33238N, 33239N),(primary & secondary pulley)
new NSK-NTF taper roller bearings (NSK-NTF R60-32A), (large differential side) (note: Could NOT source, reused old taper roller bearing in great shape)
new Timken/NSK taper roller bearings (Timken 30206M-90KM1 (qty1), NSK HR 32008xJ (qty 2)), (differential, transfer gear, etc.)
new Mitsubishi torrington thrust bearings (forward/reverse clutch assemblies), (P/N xxxxx)
new Mitsubishi factory endplay torrington thrust bearings (forward clutch drum), (P/N xxxxxx), (very hard to get, special ordered from Japan)
new NTN oil pump sprocket roller bearing (NTN 6901NR), (purchased from www.mitsumi.com)
new Mitsubishi oil pump sprocket retaining clip (Mitsubishi #2791A018), sourced from (www.megazip.net)
new oil pump sprocket thrust oil bearing (Could NOT source, looking into it, reused stock unit in great condition)
new Nissan/Jatco oil pan & pan gasket, bolts, drain plug, etc.
new Nissan/Jatco oil filters (oil sump and high pressure filter cartridge)
new high strength alloy case bolts, oil pump bolts, pulley end case, etc. (Mcmaster Carr)
etc....etc...etc.

Reman/replaced CVT parts:
reman/rebuilt CVC torque converter (DA30N) with new lockup clutch disc, new seals, new shaft oil bushings, etc., est. rated for 312 lb-ft (412 N-m), (factory pressure and spin balance tested) (original was fine, always replace torque converter on transmission rebuilds)
reman low wear/mileage Altima RE0F10A input shaft, (original broke/wore from failed clutch drum)
reman low wear/mileage Altima RE0F10A stator housing (original broke from failed clutch drum)
reman low wear/mileage Altima RE0F10A forward/rear clutch sun gear (original was failing, poor design)
reman low wear/mileage Altima RE0F10A forward clutch drum. (original failed, cracked in half, input shaft snap ring launched, input shaft hammered and cracked drum)

Refurbished/reused CVT original parts:
refurb/refinished original primary & secondary pulleys (very bad scoring, resurfacing required)
refurb/reused original clutch planetary assembly (heavier duty 4wd design vs. weaker RE0F10A design)
refurb/reused original differential & transfer gear (very heavy duty, zero wear)
reused oil pump chain
refurb/cleaned original aluminum cases.

Aftermarket CVT upgrades:
Custom modified 4-clutch forward drum (stock is 3-clutch) est. rated for 330 lb-ft (449 N-m), (custom machining and extra clutches purchased)
Custom forward clutch drum spline hub reinforcement insert (pending***), est. rated for 297 lb-ft (400 N-m) torque
SAP upgraded cobalt pulley axial slider roller pins [(6x12mm, qty3 & 6x25mm qty3), (stock are prone to fail cobalt 6mm dia. ball bearings(qty 18)]
Spirolox (Smalley WSM-98) input shaft/drum heavy duty spiral retaining clip 25mm dia. (+8500 rpm rated) (stock failed @ high mileage, launched out of clutch drum, destroyed entire trans)
Sonnax (33510-N02) hardcoat anodized oil pump flow control valve (fix for jammed/worn stock oil pump relief valve, my stock pump pressure relief failed 50% open)
OEM sump pan magnets, added a 3rd magnet (stock has 2)
Setrab external CVT oil cooler (pending***)
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #297
Here is some progress on the clutch drum reinforcement insert. Previously I had attempted to make a straight walled insert sleeve due to space limitations. This posed a problem in having to accurately machine straight hole in the primary pulley shaft, while the transmission was assembled. The other problem was the insert sleeve wall thickness was around .040", which was too thin to accurately cut reliably on the lathe. The new design is a tapered/drafted sleeve that can be easily machined and is much stronger. It doesn't reach the top of the spline hub as before, but it doesn't need to for full strength. To cut the clearance pocket for the insert to sit in the primary input shaft, I'll use a 75 degree Neway carbide valve cutter to create a steep chamfer clearance pocket. This'll maintain full strength in the primary pulley shaft without weakening the external splines that mate with the sun-gear assembly.

Then, the upgraded reinforcement insert sleeve (i.e. black sleeve) gets slipped onto the clutch drum with some bearing retainer epoxy or silver solder to form a good strong surface bond (see pic below). Final gap clearance might be something like .001-.003" so the epoxy can fit in-between for a strong bond. I'll probably set that clearance by honing it myself and having the machine shop undersize the I.D. a bit. I'll then put clay on top of the insert and reassemble the drum to check the gap clearance with the primary pulley shaft, which will measure about .020" minimum. In the pics below, the large pockets/gaps between the large sun gear and the various parts are for the (2) torrington thrust bearings that keep these high speed spinning parts from contacting, I don't have those shown but they go there. Finally, I added the 4th clutch pack to increase the clutch torque holding capacity, that's been done and sitting in the transmission currently. The last pic is the stock drum assembly that broke, thanks to the failed snap ring.

I'll be creating the drawings and sending the CAD over to the machine shop while I prototype an SLA rapidprototype of the insert on my spare clutch drum sitting on my desk. Hopefully I'll have something to test fit over the next 2 months or so. The end result should be a very tough drum assembly that can take nearly 300 lb-ft (400 N-m) of torque and still survive +100,000 miles, that's the goal anyway.





187525


187526


187527
 

·
Registered
Joined
·
29 Posts
Here is some progress on the clutch drum reinforcement insert. Previously I had attempted to make a straight walled insert sleeve due to space limitations. This posed a problem in having to accurately machine straight hole in the primary pulley shaft, while the transmission was assembled. The other problem was the insert sleeve wall thickness was around .040", which was too thin to accurately cut reliably on the lathe. The new design is a tapered/drafted sleeve that can be easily machined and is much stronger. It doesn't reach the top of the spline hub as before, but it doesn't need to for full strength. To cut the clearance pocket for the insert to sit in the primary input shaft, I'll use a 75 degree Neway carbide valve cutter to create a steep chamfer clearance pocket. This'll maintain full strength in the primary pulley shaft without weakening the external splines that mate with the sun-gear assembly.

Then, the upgraded reinforcement insert sleeve (i.e. black sleeve) gets slipped onto the clutch drum with some bearing retainer epoxy or silver solder to form a good strong surface bond (see pic below). Final gap clearance might be something like .001-.003" so the epoxy can fit in-between for a strong bond. I'll probably set that clearance by honing it myself and having the machine shop undersize the I.D. a bit. I'll then put clay on top of the insert and reassemble the drum to check the gap clearance with the primary pulley shaft, which will measure about .020" minimum. In the pics below, the large pockets/gaps between the large sun gear and the various parts are for the (2) torrington thrust bearings that keep these high speed spinning parts from contacting, I don't have those shown but they go there. Finally, I added the 4th clutch pack to increase the clutch torque holding capacity, that's been done and sitting in the transmission currently. The last pic is the stock drum assembly that broke, thanks to the failed snap ring.

I'll be creating the drawings and sending the CAD over to the machine shop while I prototype an SLA rapidprototype of the insert on my spare clutch drum sitting on my desk. Hopefully I'll have something to test fit over the next 2 months or so. The end result should be a very tough drum assembly that can take nearly 300 lb-ft (400 N-m) of torque and still survive +100,000 miles, that's the goal anyway.





View attachment 187525

View attachment 187526

View attachment 187527
Ddddddaaaaaannnnnggggg
 

·
Registered
Joined
·
1,182 Posts
Discussion Starter #299
Yep, it’s definitely an R&D project.

New oil pump came in, so did the Sonnax relief valve. I’ll post up some pics of that pump rebuild. It’s good to just have new components in the CVT where possible.

Finished the clutch drum insert upgrade design. Spent most my time running stress analysis to improve the reliability at higher torque levels. I’ll be using StayBright high strength solder to bond the insert to the drum. The silver solder braze rod I was looking at was 1250*F melt temp, but its too high for the drum heat treatment. The solder version is 550*F melt and has a tensile strength of 10,000 psi. It’ll work to reinforce the entire assembly and keep it secured.

The input shaft snap ring that popped out has already been upgraded with a Spirolox spiral snap ring, it’s actually rated for 10,000 rpms. I’m then going to solder the retaining groove with the snap ring in place to reinforce it and keep it tight. They usually fail be spreading out at high rpms or leaning out of the groove at higher thrust loads. I’m not entirely certain the cracked drum stressed the ring and popped it out, or if the ring popped out first and the loose input shaft hammered the drum and then cracked it. Either way, these upgrades should improve durability.

This’ll round out the transmission upgrades for now.
 

·
Super Moderator
Joined
·
9,840 Posts
Amazing stuff as usual. We all here hope it works out as planned that is for sure.
 
281 - 300 of 319 Posts
Top