[pboglio]

Mine was a dog until I reflashed with the latest TCM flash, made a big difference in launching and rolling response. They aren't terrible off the line, but it isn't a drag racer anyway. It's much better from the roll and better on the CVT transmission as well too. Launching will always ruin a transmission.

The Juke forward clutches are like a fuse basically. When I go with these Raybestos upgrades I'm going to soften the turbo response a bit to take the stress off.. It makes a bigger difference if you turn the wick up to like 300 lb-ft, then it would probably shatter the forward drum and waste the belt too. I was always at 247 lb-ft (crank) right about where the stock clutches were just about slipping. My dyno run the clutches were slipping the entire time once I hit 210 lb-ft @ wheels, then below that they grabbed again. That little bit of slip looks like nothing but it wears them out something fierce......then say hello to a new CVT.

I got the Exedy clutches in a kit but the steel plates were Raybestos, no idea why they didn't use Raybestos on both frictions and steel friction plates. That was an oversight on my part for not checking. Those Master rebuild kits are mega mega tricky and dangerous to use. You could use that master kit and assemble everything perfectly and the CVT would fail to work properly. I have my sources for each individual sub-kit item and prefer some over others based on my initial bench tests and CVT rebuilders out in the field. I'll probably put together a list of P/N to build a high quality master kit from the sub-kits that are out there. That's on my to-do list for the forums.

Yep, clutches are the way to go. Stock torque converter is plenty good for any power actually, the Maxima converter is only a smidge bigger on the converter clutch and that thing put down 257 lb-ft stock.

 

[Macgyver]

Interesting video. Murano CVT.......

 

[pboglio]

That's a good one too, thanks for posting it. He does a very nice job quickly and concisely explaining each component function. They look simple compared to an auto trans. That planet carrier has (6) planet gears to the Juke's (4). Another reason why our little transmission is so fragile, there's only so much you can do to beef these things up

I've been tied up at work so not too much goin on with the transmission. I'll probably order up the Neway valve seat cutter for the input shaft pocket relief angle cut to clear space for the drum reinforcement insert. I'm looking at some ideas to either laser peen or micro-shot peen that 4340 drum insert piece to harden it. That's probably the last tricky detail I have left on this transmission.

This project is coming to an end and I'd like to put this transmission to good use. Probably I'll shift focus on the trans cooling mini-project at some point.

 

[Macgyver]

Yeah quick and dirty video. Literally Dirty.

 

[turbo noob]

I love videos like those! They open my mind up to other designs and they inspire builds like these where someone is hell bent on making it better.

 

[pboglio]

I've been watching a couple of videos on automatic transmission rebuilds, always something new to learn.

Below I show the datalog of when my CVT belt actually slipped. This is detected by a "gear ratio" change between measured and predicted. One column has "0.88" which is the actual gear ratio, another has "0.79" which is the commanded gear ratio. That is almost a 250-300 rpm slip which is very noticeable while driving. This is the belt slippage between primary and secondary pulleys. That's how I can tell the CVT is slipping the belt vs. the clutches. This is different than torque converter slippage. When the belt slips at WOT.....it's a hard jerk, then the computer catches it quickly and cranks up the pulley pressure. When the clutches slip you barely feel it and it's a smooth slip that keeps on slipping. The computer can't compensate because it's hitting the clutches with full available pressure the entire time anyway so there is nothing left to help.

That is the oil pump with the seized valve, this is how all the JF011E die. Below is the anodized stock valve. The Sonnax replacement is hard anodized which is like a ceramic coating in terms of hardness. That's been installed in a brand new pump I sourced from WIT. This is what pretty much controls the belt pressure.

 

[Macgyver]

Crank that pressure up.

 

[pboglio]

Got some quotes on the forward clutch drum reinforcement insert. Looking at 3dhubs, Protolabs, E-machines shop, and my local guy. Not cheap for a simple part. Probably averaging about $280 for (3) pieces, most of it in setup cost. I'll keep a couple extra in case I rebuild/upgrade any more Juke CVTs in the future.

The steel I'm using is a special 4140 ETD-150 alloy steel. What's pretty cool about this is it's high strength without needing a heat treatment, yet it's relatively easy to machine at C32 Rockwell hardness. It's up near 130,000 psi yield strength which is what a heat treated 4140 or 4340 alloy steel might get up to depending on the heat treat conditions. The is kind of critical since the part is so small, it needs to be as strong as possible in a compact size. I can also use 300M alloy which is a go-to material for input shafts that need mega-strength, but annealed it's still only 100,000 psi yield strength and C25 hardness so the 4140 is actually stronger. If my machine shop can't cut the 4140, then the 300M annealed is a good backup material to use.

The reason for this little part is because when the clutch drum get's hit with some big boost repeatedly, they tend to split clean in half. The Nissan Maxima run a reinforced stock clutch drum that can take the 257 lb-ft from the 3.5L engine. Unfortunately that drum won't fit our Juke as it's too tall, but this piece will help simulate that on the Juke. Anyway, when these parts break it typically destroys the stator housing, input shaft, and drum while the torque converter loses all hydraulic pressure as well. Basically you are DOA on the side of the road. I don't want to experience that again.

Normally a part like this would be press fit in place. In this case the forward clutch drum is some type of nitro-carburized 4140 or 4340 and it's so hard it might crack under the stress of press fitting this part over it. Most people don't understand that to press fit (2) parts together, one part has to be beefy and strong while the other part has to stretch a little bit like a rubber band. What I'm doing is under-sizing the insert bore I.D. Then I'm going to hone to size and match fit perfectly with the forward clutch drum hub spline O.D. There is no machining operation that can hold a +/-.001" tolerance without the cost getting out of hand, so under-sizing the part then repeatedly honing it I can inch up to the final slip fit. The C32 hardness isn't so hard that honing will take forever, it's just a matter of patience. I'll hone it to about a +.0001" (.025mm) gap clearance per side, then use some Loctite 603 bearing retainer anaerobic compound and I'll glue it in place. This'll hold up to 450*F so I think it'll be safe.

Here is the Loctite bearing compound I'll be using:

Loctite Bearing Retaining Compound, Oil Seal Retainer - Henkel

Opt for Loctite’s high strength retaining compounds that fill all microscopic gaps between cylindrical components to double the strength of traditional fittings. Visit for more information.

www.tryloctite.in

Anyhow, once I get the parts in a couple of weeks I'll show a before/after mounted on the drum showing how much beefier the design is.

 

[pboglio]

My machine shop finished up my custom parts, took about 3 weeks as scheduled.

It doesn't look like much, but I spent months designing/simulating it on the computer. Mainly because there isn't any actual space for it on a stock transmission.
I'm having to custom machine the input shaft to allow this thing to sit in the transmission with some decent clearances. It's boosting the clutch drum strength by about +70%.

The pictures show how the part gets assembled. Keep in mind that drum is a spare, not the actual one inside my transmission with the (4) clutches. I undersized the insert piece by about .002-.003" so I could match hone it to the existing drum for a slip fit. That is a little trick with machining, no one is going to nail the dimensions perfectly, so you undershoot then apply a 2nd precision operation to bring the dimension into perfect spec. It would have cost extra for the machine shop so I just do it myself with a rigid hone and some honing oil. Like making a custom watch, every piece is match fit since the transmission parts very somewhat.

The surface finish is beautiful, he used a very fine cut on the lathe. On the I.D. which is the critical finish he went so fine it looks like a brush finish, almost a precision ground finish. This is why he makes all my parts for work. Another point here, Pat's shop is local right here in Frankfurt and supporting American machine shops for me is important. I had some nice looking quotes from China, but knowing how they operate they would have substituted my alloy steel for something else anyway and royally screwed things up. The 4140 is a really attractive material, almost a silver color. It's what you'd call a Chromoly steel basically, but modified for higher strength in the raw purchased state.

This will be put on hold for a bit until I get my motor assembled and I can focus on it but here it is finally.