Trans mounted clutch... TRX90

To follow the power.

Gas air and spark explode in the combustion chamber, forcing the piston to move down with force. It pushes thru the rod and spins the heavy crankshaft whatchacallit "lobes" which pushes the piston back up to fire again. This spins the crankshaft.

When the crankshaft spins, fast enough, it engages the one-way clutch and small primary gear. The small primary gear spins the big primary gear/main shaft clutch basket, which spins the clutch center, which spins the trans main shaft (Enginedoctor would call it the trans input shaft), which spins the trans counter shaft (Edocs output shaft) if the bike is in gear, which spins the counter-sprocket (front chain sprocket)...to the rear wheel.

When at idle speed, the OW clutch center would be spinning, but the outer and small primary gear would not. Power would stop there. When rpm's go up, clutch engages and sends power on...

Clutch release mechanism...

There is a arm that sits on the end of the shift shaft...#4

That arm fits into the slot in the clutch cam plate...#5, which fits right into the bearing on the front of the main clutch.

The clutch cam plate remains still while the clutch spins.
It has a shape to it that has 3 low areas where the clutch lifter...#6 fits into. #6 is mounted to the inside of the clutch side engine cover by the clutch adjustment stud...#8.

When you push up or down on the shift lever with your toe, the shift shaft moves the arm, which spins the cam plate, which causes it to push in on the clutch center, which compresses the clutch springs and creates some space between the clutch discs, which disengages the clutch, which allows the clutch basket to spin independently of, the clutch center and trans main shaft. This disconnects the transmission from the crankshaft, even if the engine is revving high, so you can shift the transmission into gear. By the way, this DOES also simultaneously shift the transmission...into, out of, or between gears.

I should also show this...
This is the clutch sandwich. Fiber and steel discs, between the pressure plate and clutch center, with the springs bolted in...compressed. this will fall right out of the basket.

Good write up posts about the power delivery through the semi auto clutch systems. Helps us all understand what is going on inside the clutch housing. Your write up answers many questions but also creates a few more but I don't want to hijack your post more than I have already.

So relating to the exact clutch setup you have .......... I have seen a 114cc V2 head engine built out of a Honda TRX 90 motor that has the same crank centrifugal clutch and the transmission mounted "manual" clutch like you have posted. It was being raced in a TRX 90 quad chassis so the traction was high and his son was experiencing clutch slippage of the fiber disk manual clutch assembly. The first solution was to shim the stock "manual" clutch springs for more clamping pressure and eventually he used stronger CR80 clutch springs that stopped the clutch from slipping.

Then to gain a hole shot race advantage he converted the 90 motor to a full manual clutch. He modified the clutch cover and adapted a cable to the clutch cam plate allowing him to activate the clutch by a hand lever. Also removed the centrifugal clutch and it's rotating weight for better acceleration.


However what if you did the same thing (added clutch hand lever control) but also kept the centrifugal clutch still? Would you now have the ability to "pop the clutch" on a semi auto motor without holding the shift lever down with your foot?

I'm thinking this would work like a Rekluse clutch where you can feather or control the clutch with your hand while still retaining the centrifugal clutch that would prevent stalling the engine if you came to an abrupt stop without pulling in on the clutch lever. Thoughts?

There is no such thing as hijacking one of my threads. lol, I am the hijack king. I welcome any and ALL comments, questions, and criticisms. All.

Please, if I sparked a question, I definitely want to hear it. If I don't know the answer, then I want to learn it.

Your friend who's son is racing this setup might first try HD springs, they're available. Then, there is also a upgrade that uses a 5 disc clutch pack instead of the 4 that the stocker has. There is also the Takagawa special clutch system. It is a complete setup with the 5 disc clutch and manual conversion engine cover. Pricey sucker, but quality stuff. I've heard that it's not great for my style of riding, (the clutch pack I mean) but I think it might be the shix for racing...high revving.

Next up in this thread, I had planned to go into some of the Manual conversions/options. I have the Taky manual clutch, cover and bits only, on order. It shipped today in fact from Webike. I'm really liking the youtube that you posted too. I want to watch it a couple more times...I'm definitely interested in getting a closer look. With that setup, I think you could keep the auto clutch setup intact, and have a manual override. I think that's what you said...Rekluse.
I'm not familiar with Rekluse, but I might look into it.

I forgot to say that the Taky manual clutch kit would require a different crankshaft...one like the CT Z XR 50 70's have. Or usually a stroker crank for those bikes.

So my 124cc ST50 is running the TRX90 engine with to semi-auto clutch. But the stock clutch was machined to accept a extra fiber and steel disc...making it a 5 disc kind of like the taky and some of the Chinese engines. I ride that bike on the street and really have no complaints with the semi-auto clutch. The only bitch I could offer is that you can't just pull in a clutch lever and downshift to first like you could with a manual.

But since I've taking a big liking to the TRX motor, I'd like another, with a manual clutch.

My ST50 has a 54mm stroker crank that is a direct replacement for the TRX crank...long on the right side. It's made by Clipping Point. My motor also has a 54mm cylinder and a 4 valve head that are Clipping Point. I'm not sure if Takegawa or kitaco make a crank for the TRX90.

But if you want to convert to a manual clutch on a TRX, and you want a stroker crank, you can use the 12v CT Z type cranks in the TRX motor, but you have to sort out the clutch.

CHP used to have the taky clutch setup that is made for this job. The one pictured above. They also sold the cover only. Then they had the bits you would need to make the cover work with the TRX trans mounted clutch.
The bits... basically just a small primary drive gear and clutch nut made to mate up to the taky cover. The cover has to engage the crank shaft because oil is supplied to the crank thru the cover.

Another option allows you to keep the TRX side cover, eliminate the one way clutch, and I guess you figure out a way to disengage the trans mounted clutch...maybe the same way SS LS1's friend did it.
With this setup, you buy the small primary bits, plus a oiling tube that connects with the TRX's side cover to supply oil to the crank.

When I had my 124cc motor built, I asked for a manual clutch. Terry's plan was to just add a hydraulic clutch slave cylinder that would disengage the trans mounted clutch. The one way/centrifugal clutch would stay. It would still disengage at low rpm's and idle in gear etc. But at speed, you would ride and shift it like a manual clutch setup...you wouldn't have to allow the rpm's to drop before shifting, since you could disengage the main clutch with the hand lever.
This is the slave cylinder...it would go into the bearing on the clutch, and it mounts in the clutch adjustment stud hole.
This one would take quite a bit of work to get it properly mounted.

That clutchpack looks like the Wave 100/Nice 110 semi-auto.

Adding a lever-activated throwout mechanism would be the way to go, if you want full manual control. Don't forget, the rotating mass on the crank, becomes wildly different R-to-L once the centrifugal clutch has been removed. That's not a great idea, especially with an assembled crank. Next, the secondary (manual) clutch was not designed for full manual usage. There's less damping, in the basket (manuals have actual springs), the clutch discs have different friction material, less of it and 4 full-width discs, with no anti-chatter spring & seat. This clutch was meant to grab abruptly. IMO, that's a lot working against it, as a full manual conversion.

As for dropping rpms to shift...ummm...throttle modulation is part of normal shifting technique, if you want the tranny to last.

That Taky clutch conversion has a pleated element oil filter for a reason...no oil spinner. I actually prefer this type of oil filter. That said, any conversion that eliminates the centrifugal clutch, or oil spinner, or uses a lightened spinner, I'd pair with a lightened flywheel...for longevity's sake.

Very good points racerx. The main clutch does have rubber dampeners where the primary gear mounts to it. Not really adequate I don't think, but its something. When these clutches are run hard, especially when run as a full manual setup, those dampeners will be worn out. You can move the gear and the basket back and forth by hand power, in the worst cases...it'll be obvious.

When the crank mounted clutch is eliminated, I think most guys will switch to a inner rotor stator setup, a small, Daytona style flywheel/stator, or a lightened stock flywheel. The racers seem to be the ones who like that super light crank setup. Some guys don't like it at all and miss the weight, and rotational mass.

Eliminating the crank mounted clutch does eliminate the oil spinner. Like the doc has mentioned in another thread, you'd have to add a in line oil filter to the oil cooler setup, or even if you don't have a cooler.

Throttle modulation is definitely still a good idea for most. But a racer is going to be pushing the limits. Having a clutch lever allows them to speed shift. I was getting at the possibility of adding a hand clutch lever to replace the clutch lifter/toe lever setup, on the stock semi-auto clutch.

Here's what I once did...of course I did this before!

I had a honda wave crankshaft, which was the same as a trx90, except it was a 68mm stroke. I wanted to build a crazy interesting motor out of it, so I got a set of midblock chinese cases and stuffed it with a trx90 transmission and clutch. I added in a kickstart that matched the transmission and modified the cases so it would work. I used a set of Daytona TREX 150 cases, so it had the non-kick in gear setup. Once that was done, I took the 1-way clutch mechanism and modified it to be non-centrifugal by turning it all the way down and fill welding it solid and then turning it nice. Looked almost like a stock piece when I was done. Then I popped a normal chinese clutch cover on it and had a full blown manual clutch. This was 178cc when done.

When I built my "poor man's nice" engine, I was non-committal to that approach but it was effectively the same thing. I used an aasa 63mm crank, daytona trex cases, and a trx90 transmission, but this time I left the centrifugal mechanism intact. This gave me 157cc engine, semi auto. If I ever wanted to convert it, all i would need to do is swap the clutch cover and repeat the turning and fill welding as I did on the other build. If ever there was a regret, is that I sold this motor on, as I it was dead nuts reliable and fast.

Honestly, if I were looking to build a semi-auto motor, I'd built it damn near the exact same way I did the Poor Man's Nice except I'd consider adding a superhead to the top for even more power.

If you want to go 124 on a trx90 base, that's actually quite easy. Just swap the crank to the takegawa 54mm crank and switch out the crank gear from 18T standard on CT's to a 17T from an xr50 to match the clutch gear. Assemble as usual.

Also, a little tidbit of information. NEVER buy one of the aftermarket oil pumps on any 12v small block engine setup. Buy the TRX90 oil pump...same size as the Takegawa pump, and you can buy them new for about $7.

Been doing a little semi auto clutch research and found this video showing another idea to add hand clutch lever capabilities. Looks like they tried to make a full manual by also welding the slinger/centrifugal clutch assembly together.

I am thiking this hand clutch mod could be done while also keeping the centrifugal clutch assembly intact. This way the bike can be stopped without pulling in the clutch and not stall the engine due to the centrifugal one way setup that is still intact.

Personally I would not go through the trouble of the aftermarket clutch modification kits nor would I try to make a semi auto into a full manual as I don't really see the point. I would rather just use a dedicated manual clutch motor instead in which I have several already.

Please note this is for idea discussion only and not a direct recommendation from myself as each individuals needs might require someting different to meet their goals. I would never use something like plastic zip ties to make a modification like this on my personal engine but it's just to show an idea giving multiple clutch operation options on a semi auto motor.

Good stuff Fatcaaat. You're my hero. One of them for sure...mad scientist of mini bike engines. I come across your posts all the time when I research stuff like this. I've read about your Poor Man's Nice 3 or 4 times before, and I was around when you posted it for sale too.


lol...SS LS1 that vid looks like a advanced course in shadetree mekanix. I love it. Guy #2, lapping the crap out of the valves with a screw gun is a awesome backdrop. This is the way, I think, awesome mods begin. Idea...try it, with zip ties and duct tape...if it works, refine and improve from there.

That is really similar idea to how your friend did it too. Just connecting a cable that will spin the cam plate thingy, and disabling the centrifugal clutch.

This is the Takegawa clutch breakdown...I think it's the right one. The clutch pack looks very similar, just with a extra disc.

http://www.takegawa.co.jp/catalog_v...g=ALkJrhgtF1OVkBG2TZPEWHGzPNzBzoq2JA#page=606

Exciting!
I only have a few Takegawa parts. But the quality is evident when you hold them in your hands.

Beautiful cover for sure! Some nice motor eye candy for your ride.

Still not sure what auto clutch setup is best the split centrifugal/manual like the TRX 90......or the all in one centrifugal/manual mounted all on the crankshaft?

The split system appears to have more "manual style" clutch plates for more clutch holding capacity on higher HP motors. However seems like a lot of rotating mass to spin which might slow acceleration but would likely be very lugable once under way.


The centrifugal clutch has lots of surface area on the clutch shoes to grab the outer clutch drum for great clutch grip. Plus it looks like it would be rather easy to modify the spring strength or lighten the clutch shoes to increase the engagement rpm of the centrifugal clutch.



The all in one centrifugal/manual would likely be less mass to accelerate and less to decelerate which should help smooth down shifting engine braking surges. However there could be less "manual clutch plate" surface area compared the the split semi auto clutch.



However not sure how the centrifugal weights and center hub ramp clutch grip strength compares to the clutch shoes seen above?







Also not sure how you would modify the centrifugal clutch to raise the clutch engagement RPM with this setup. (7 centrifugal weights per side) You would likely have to just change out the clutch for one with fewer centrifugal weights like this one with only 4 weights per side (16 total) This is said to raise centrifugal clutch engagement RPM by 1500 RPM over the 7 weight clutch (28 total) but has the clutch grip strength been compromised?

Raising engagement speed will decrease clamping load and, with it, grip power. It's a question of how much higher rpm has to be attained before clutch grip is adequate. Sorry of that sounds kinda vague, lacking dyno data, it is vague...awaiting trial & error testing.

Whether a lightened rotating assembly and higher initial engagement (a.k.a. stall) speed is a good thing depends on whether you're racing or road riding. Shift smoothness is purely a matter of rider technique. Under competition conditions, the higher engagement rpm, reduced rotational mass and shifting technique thus required are all part of that balance equation. You will get increase clutch heating, wear debris and reduced service life; race engines don't wear out, they're run until they break. IDK that the stock clutch will live long when heated well beyond its intended heat range. On the street...whole other ballgame. A fraction of a second quicker 0-20mph time will be all but irrelevant, in the "stoplight grand prix." Reduced longevity doesn't play well if you want to get some real miles out of the bike. And, that racing shift technique can get old really fast when negotiating traffic, or going out for a long afternoon's touring. OTOH, if your usage plans are for low mileage and short rides...well...maybe a peaky & tweaky race setup is the way to go, if that's what suits your tastes...and you have the budget to absorb parts failures.

To be clear, mine is a skewed perspective...serious road use, i.e. long rides and over a number of years. IMO, the most critical engine power parameters are how well it performs up to 50-60mph and "NVH" (noise/vibration/harshness) as that has a huge impact on rider fatigue. Your priorities may be different.

This is all very good discussion IMO.

Honda must have 5 or 6 versions of that crank mounted semi-auto clutch. The ones pictured in post #37 must be the latest. XR CRF 50 & 70 types, or Chinese copies of them. Firepower357 Terry had told me there was a difference between the 50cc version and the 70cc version...those weights were one difference, and I thought there was one other difference. For bigger motors, he recommended the 70cc version of that xr/crf clutch, with TB's 3 disc clutch kit. Better holding power, and better operation than the earlier versions of the same clutch. As far a engagement speed and such, that's all over my head...outta my league at this point.
But I think the trx90 semi-auto with the trans mounted clutch is a better system yet. Honda designed it for the newer, bigger, 90-125cc engines that use a semi-auto. I know it works really nice on my 124cc, the only one I've ever ridden.

I might be moving a little too fast thru this thread...I'm passing over some of my own questions.

I'm planning 2 trx90 based engines.

1st one will be 114cc using the stock crank and clutch setup. With a 54mm cylinder and the stock trx90 head. I'm adding a kickstart. I'm not too concerned about this engine since it's just the stock setup, with a bigger bore. It's been done many times before...you can buy the BB kit everywhere. The lower end is mostly stock, and everything works together as Honda designed it...everything gets oiled. lol
I want to run this engine in a 90's CT, and/or a ZB50, and or a ATC70.

The 2nd one will be converted to a manual 124cc, using a bunch of good quality parts. I've tried to do my homework, but I'm still nervous about this one. 54mm Takegawa crank and cylinder. Eliminate the centrifugal clutch and run the stock 4 disc Honda clutch with the Taky clutch cover, small primary gear with extra bits. (The extra bits part is kinda tricky. But I think I know what I need.) TB V2 head and matching 54mm piston. Add kick shaft.
I want to run this motor in a ugly K0 CT70.

I was planning to run a 90's Honda stator and lightened flywheel.
Maybe that will still be too heavy??

I expect this engine to be similar to my 124cc semi-auto trx90 in my Dax. (My sons Dax now actually) But I expect it might rev a little quicker with less weight on the crank..?

I'm not a great engine builder, but I keep trying. You can research all you want and still come away a little confused. Someone can try to tell you haw to do it, but it's hard to get ALL of the details. You hear a lot of, "you might have to...to make it all work" Details are left up to the builder to figure out. I'll be gambling with some expensive parts.

This thread is to help me gather as much knowledge and info as possible, to give myself the best chance for a good result, on the first try.

I'll be posting my plans and progress as I build each motor too, to give myself the best chance...good result...first try.

kirrbby said:

You can research all you want and still come away a little confused. Someone can try to tell you haw to do it, but it's hard to get ALL of the details. You hear a lot of, "you might have to...to make it all work" Details are left up to the builder to figure out. I'll be gambling with some expensive parts.

Click to expand...

That is the heart of the matter when attempting to build something that hasn't been done before. Unknown combination = unknown outcome. Find a way to overturn that little fact of basic reality and you'll soon be rich.

IMHO, one needn't be an expert engine builder so much as a meticulous assembler. Machining can be legitimately farmed-out and, in most cases, should be...that can mean something as basic & routine as cylinder boring, or a valve job, not just exotic mods. follow that out by eliminating/avoiding obviously risky stuff...like dramatically increased stall speed (which dramatically increases heat generation) and unknown parts with no field service history...and you're well on your way to something that will at least stay together & running.

Beyond that, breakdown the engine build into lower & top ends. Power is made in the top end assembly. The lower end has to apply the power and survive it.

Thus far, you've got a Takegawa clutch conversion & crank, feeding into an OEM Honda tranny. IMHO, that's rock-solid quality stuffs. There aren't all that many piston, cylinder, head and cam choices available for any given engine. So, you choose the piston to fit the rod length (proper compression height) and combustion chamber (dome configuration), then the cam profile to suit your desired revband. Verify deck height (rotate the crank to see that it's no down inside, or stand proud of, the cylinder), that there's no valve-to-piston contact, check the ring endgaps...all that's left, then, is making sure that everything is immaculately clean & oiled for assembly. At worst, you decide that you don't like the way the engine behaves with the chosen cam; cams can be changed easily enough.

Yes, selecting parts from online catalogs can be confusing...and cause anxiety. I'm not making light of those concerns. Japanese transrates only so well in Engrish...ask me how I know. That said, I think you've got most of this pretty well planned-out. Going by what you've posted in this thread, so far, the biggest unknowns will be the clutch discs and which alternator to use with the AWOL centrifugal clutch. I can't see those things holding you back.