Project 49

[90GTVert]

Timing was only increased for the test runs.

I think I've got my work cut out for me getting this in tune. I haven't spent a lot of time on it yet, but so far... no good. Started out with the CVT setup as it was, 5g/1K contra, and an 80 main in the carb. Slow. 7000RPM max, 36MPH, running cool. Tried both 75 and 85 mains, 80 seemed like the best. The running cool part may just be that we've been having 95-100 degree temps and then the last 2 days have been 80s. Not to worry, back to 95 tomorrow. :doh:

I then decided that maybe it just needed the revs to increase. I went from 5g to 4g sliders. 8000RPM, 38MPH. Then I wanted to try a little more RPM still. I have no lighter weights so I used a 2000RPM contra. 9000RPM, slow, not sounding happy. That's when I quit for tonight.

[reveeen]

Clutch springs..... try heavier clutch springs. I have to run the heavy ones as I have virtually no bottom end. Then tune the CVT to remain at the maximum mechanical advantage until the clutch "hits". (if you don't it will bog) Set up right it should pull the front wheel when it "HITS". I'm "hitting" about 6000 rpm with 3g rollers and pull to around 10,000.

[90GTVert]

I really don't think it's clutch springs. I have no trouble getting off the line. It's just that it doesn't speed up. I just made 2 more quick runs. One peaking mid 8,000RPM range. The other in the upper 7,000RPM range. It felt like there's power there in the high 7s at least... but it doesn't speed up. Something's gotta be up with the CVT. No way it feels like power is there but I go 36-38 on every run and no matter what RPM if the vari and TD are working right. If anything it would feel like the clutch is slipping... but I can tell it's not. It slips and takes off, grabs in, pulls, and then feels like it's slipping again.

[90GTVert]

I took a ride this morning and it does not shift, or I can't hear it. I rode for 13 miles, and I never heard a shift. It flashes past the clutch and picks up and just steadily climbs. A few times I slowed down from a cruise and the RPM dropped to low 6,000RPM range. Then it really struggled and dropped more. Once in the 5,000RPM range it's completely dead and goes just over 20MPH. I'd have to let off and get back in it to get RPM to start rising again.

I took the CVT apart and it all looks OK. I had marked the variator before that ride and I had 7.5mm left on the face with a max speed of 40MPH (hill). I was getting 5-6mm with 5-6MPH more, so it doesn't seem that the travel is really off. I do see that the 40MPH groove I had in the last torque driver is starting to form on this one. It's very slight though, and wouldn't explain the lack of a shift.

[90GTVert]

First, I forgot to say welcome last time. Sorry it's 1 post late, but :welcome: to the forum. ;D


It's the stock pipe. I'll try others, but I wanna see if I can get this to work out first. This same stock pipe has seen 9,000RPM+ on a 90cc with more duration, and I see no reason it should cause the transmission to stop shifting. There's power in there at certain points. I can feel it. The CVT just doesn't want to cooperate right now.

[90GTVert]

I've tried all sorts of weight/spring combos. Some work better than others, none work well. Swapped to a different stock vari and different stock TD. No improvement. I'm lost as to why it no longer shifts like it did. Almost acts like it gets into 1 gear and stays there. There's some change in RPM, but it's a lot different than it was before all of this. Since the CVT has all been apart, cleaned, replaced, etc... I think I'm gonna check for air leaks and maybe take the top end off just to be sure I didn't do anything dumb on reassembly. I'm starting to wonder if I put the piston in upside down or something crazy. Anyway, here's a vid. I've had better results with a more RPM, but you can see it sucks big time.

[90GTVert]

I checked compression and it was 142psi. I had 166psi when I assembled it. I removed the cylinder and checked everything over. I found that ring end gaps were .031" and .035". Service limit is .030". I replaced the rings with a new set. Those have .011" and .012". I don't have my notes from it now, but that's about where the first set was when the engine was assembled to start the project. Both kits from ScrappyDog. I like them, but this is why I don't usually buy engine parts from them. Seems like they have the worst tolerances. I can usually get a 90cc piston kit from PFS or eBay and have the same or less end gap.

Otherwise it all seemed OK. I did extend what were prob 1mm chamfers to 2mm chamfers as suggested just to be safe. I can't get a decent pic of it to save my life. My cam now has a couple of spots and I think they're screwing up the focus sometimes. Tried cleaning the lens, but they may be inside. I dunno. Anyway, it's got a good full chamfer all around, because I wanted to be sure that wasn't causing a problem.

Put it all back together and did a compression test. 147psi, up 5psi. My only thought is that I had more oil in the cylinder last time when I did a compression test and that could affect it.

[palezu]

My suggestions(probably not the case but maybe worth checking):
-since you widened the port, the piston skirt might not cover the exhaust port completely at TDC(from the picture the pistons exhaust side skirt looks really narrow)
-or the port is now too wide, there is not enough clearance to the transfers next to it so that you're getting mixture short circuit

those would also explain why it had ran cool on your first test as you wrote.

[90GTVert]

I checked and it does cover the whole port at TDC.

Not sure just how narrow is too narrow for clearance between the ex and trans, but I only moved them about 0.5mm each way. They're something like 4mm apart.

[90GTVert]

I wanted to get some numbers for corrected and uncorrected compression after the piston and port mods to see if they seemed like I should really be so far off on cranking compression.

The modified piston crown now displaces only 0.9cc, from 1.1cc stock.

With the head gasket :

CCR = 6.3:1 (Previously 7.1)
UCCR = 10.7:1 (Previously 11.1)

Without the gasket :

CCR = 6.8:1 (Previously 7.7)
UCCR = 11.5:1 (Previously 12.0)


Makes quite a difference. Just for giggles, I did the math if the piston crown were to remain unchanged (keeping transfer duration at 125° instead of 128°). UCCR does not change, since all volumes would remain the same.

With head gasket :

CCR = 6.5:1
UCCR = 11.1:1

Without head gasket :

CCR = 7.0:1
UCCR = 12.0:1


Then, I wondered what it would take to get corrected compression back to where it was at 180psi with the head gasket before all of this (7.1:1). I would need to get trapped volume down to 4.4-4.5cc. It's 5.1cc w/gasket and 4.7cc w/o gasket. I'd need to remove about 0.2mm from the head to get 4.4 - 4.5cc and 7 - 7.1:1 CCR. Using the 2nd stock head, seen before in my compression tests, with a head gasket would give me 4.5cc, so that would be the simple way for me.




Now, since I had the piston and cylinder out, I wanted to check out the possibility of an exhaust port coverage issue at TDC as palezu mentioned. I could see that it is fully covered, but I wondered just how close the port would be to the edge of the skirt. I used a Sharpie to mark the port's location on the piston skirt.

It's covered, but not by a lot.




This side is really bad. The marker is smeared a bit, but it's still very close. This seems like it could be a problem.




I was really surprised that .5mm or so on each side of the port just to remove some imperfections there would push the limits so far. I got out the new piston, same thing as this one, to see how that looked.

Oh, it appears I took off way more than I realized when smoothing out the edges.




You can see the stock piston comes to a thin and jagged point. I suppose since it's so thin the material went faster than I noticed.




Even though I don't like all those rough edges, it appears that the smart move on the exhaust side would be to leave the width of the skirt alone. If anything, I may just quicky run sandpaper across it so it's not rough. I think with the stock skirt, the small extra width of the new port should be OK.

I believe what I'm going to do is work on this new piston to round it out and get the durations back, just leaving the exhaust side alone. Then see how it runs. I think I will end up trying to match corrected compression if it runs better after the new piston, and see how that does.

[90GTVert]

I smoothed out the new piston and filed the crown for transfer duration as I did the last one... only this time I left the skirt by the exhaust port alone. Reassembled and now it has 153psi cranking compression. Piston crown displaces the same as far as I can tell with my syringe, but I assume it's got just a tad more material left than the last one.

It's definitely feeling better than before. I have not done any tuning. This is just like the last video posted here, but with the new piston. Lacking a little from before the porting I think, but much better than before the piston. Again, no tuning. It may help to watch the previous vid just to refresh your memory and see the difference more clearly.



While it's showing promise, I'd really like to get it back to the same corrected compression by using the other head, then see how it does. As long as there isn't excessive heat or anything weird, then I'll tune it. I really hope to see gains here rather than just being able to get it back to the same level.

[190mech]

Brent,Do you remember the pic I posted why the nice CR125 cylinder couldnt be used with the BWS100 piston on Skeedr?

I like to use a sharp dental scribe to mark the piston thru the Ex port to see how much(if any)wider a port can go..
Lessons learned the hard way are never forgotten!!

[90GTVert]

I do remember that... now that you've reminded me. ;D I have a horrible, horrible, memory for some reason though. I'll prob forget this now that I've screwed it up even. lol


I was gonna swap heads to get the corrected compression back to where it was pre-porting, but I don't like how the other stock had has less cooling fins.




I removed enough material from the head using sandpaper over glass and then cc'd it to be sure I had it right.




Re-assembled and checked compression. 153psi... same as before.




While I was working, I knocked a socket off the scoot onto the compression tester and busted the glass. Can't see why that would change anything. Figured I'd try the Actron tester I got a while ago because it has the adapter for GY6s. It sucks. lol




Even though the testers didn't think so, I could tell compression was better. There is power there. It feels pretty strong at times. It feels stronger from just under 7,000RPM up to 8,000RPM anyway. I still think there's a problem with the CVT. It will not go faster, or as fast as before, even feeling like it makes more power. It seems to me that something is slipping, no matter what I do, or it's travel is limited. I'm not sure how this is still possible with a new belt, variator, and rear pulley.

So good news, I think the cylinder has potential. Bad news, I gotta figure out why it's not using it.

[90GTVert]

I pulled the entire CVT apart once again this morning, determined to find out what's up. The fact that it feels like it may be slipping a little and doesn't want to speed up after as certain point made me think about the spacer washer that's stock and I have left in. The problem there being I have been faster with it in place, and at less RPM. I can't see why that would suddenly change. I did verify that it had no burrs or anything like that. This also made me realize I have not replaced the fixed drive face or done more than run my fingers over it to see if it felt like a groove forming. Seemed fine from those checks.

I put a straight edge across the face of the fixed half though, and could see just a tiny bit of light. It seems to me this would have to be somewhere in the 30MPH range. I couldn't even get a .002" feeler gauge under there, so I don't really think it's a big issue, but to be safe I replaced the fixed half with another that passed the straight edge test.




Then I wanted to do the same for the rear pulley. It definitely drops off toward the center, where it would be at speed. It's not worn, just made that way.



I have replaced the torque driver a couple of times now for grooves forming in it's paths, but I can't remember 100% if I replaced the fixed half of the rear pulley or when. I checked others I had around. As it turns out, the ones with Jog90 markings on them drop off in the center. The others are flat. During Project 90 I had also noticed that the Jog90 torque drivers had longer grooves. Perhaps it's a bad idea to pair a drop off ramp with a short groove torque driver? Seemed promising to me anyway.




Just when I thought all may soon be right in the world... the crank stripped.




This was one of the larger thread, M12 x 1.25, cranks at least. Some are M10 (not sure of pitch off the top of my head). I don't like re-threading the smaller ones, since they seem too small to begin with. The M12 has more room for new threads of a smaller size.

[90GTVert]

Got it re-threaded to 7/16"-14. None of the stuff I found mattered. Still doesn't want to speed up, most comfortable in the mid to upper 30s. The only thing I've really got left is to remove the spacer from the variator and/or swap to the modded drive face for possibly a better result. Again, not sure why it would require that all of the sudden though.