[200mph]

FYI. Belt is grabbed by the sides as the pulley moves inward. After engaging and returning back to idle the moveable pulley should not be touching the side of belt. If you watch belt closely when shutting engine off, you may see the pulley move ever so slightly away from belt. Using slow motion video on you phone can help detect the motion.
It’s also possible for the ID of the belt to be dragging: basically acting like the belt is too short. If the engine is separately mounted from transaxle there may be some adjustment in the motor mount to change the distance between primary and secondary. If the transaxle is directly mounted to engine this distance is fixed. Or the secondary pulleys may be positioned too close to each other. Some secondaries can be adjusted using shims.

Suggest focusing on the primary first.

many videos available showing how CVTs work. Their design may be slightly different across manufacturers, but concept is the same.

 

[Bradhill]

If I’m not mistaken the AMT 626 utilizes a Primary Clutch manufactured by Salsbury.

This type of clutch is still used today in the go-cart industry, however I‘m not certain of the Salsbury part number for your clutch? My guess, if you contact Salsbury and tell them you have a John Deere AMT, they will refer you back to John Deere.

This clutch operates on torsional springs that move the cam arms, this is very different than modern clutching used in most Utility Vehicles today.

 

[jansmeets33]

Thank you very much! I will try to get more knowledge.

 

[MBDiagMan]

When you have it up and can get to the primary WITH ENGINE OFF can you move the belt easily on the primary clutch? Does the belt have a little play side to side against the sheaves?

 

[jansmeets33]

When you have it up and can get to the primary WITH ENGINE OFF can you move the belt easily on the primary clutch? Does the belt have a little play side to side against the sheaves?

I can move the belt, but not easy, and sits against the outer sheave. When running and rave up the speed, then the belt on the primary moves to the upper side while on the secondary, the belt moves from the top going down. There is no spillage. The belt is at all times engaged, (also when starting). So the outer shave is engaged, I think?. Would that point the problem toward the transmission (gear) box? Moving into gear by force causes damage, I think. What could be the reason? (something inside dislodged or wear?)

 

[Bradhill]

I can move the belt, but not easy, and sits against the outer sheave. When running and rave up the speed, then the belt on the primary moves to the upper side while on the secondary, the belt moves from the top going down. There is no spillage. The belt is at all times engaged, (also when starting). So the outer shave is engaged, I think?. Would that point the problem toward the transmission (gear) box? Moving into gear by force causes damage, I think. What could be the reason? (something inside dislodged or wear?)

Start with a through cleaning of both clutch’s with hot soapy water and or non chlorinated brake cleaner followed by compressed air. You will need to remove the dome cover on the Drive clutch ( Primary) to clean properly. Use nylon brushes including a toothbrush to access the tight areas.

Cleaning is the first step in troubleshooting your clutch’s and will give you an opportunity to look for excessively worn or missing components.

 

[MBDiagMan]

It sure sounds like the belt spacing is wrong or the primary clutch needs attention. What I would need to get it right is both the primary clutch AND the belt so it can be properly set up.

 

[jansmeets33]

Just to be clear, in case the wording in my previous post was not correct: When Idle, the belt is at the top of the secondary and at the bottom of the primary but still with running belt and is running near the outer sheave while the gap of say 5 mm is on the engine side. But when raving up the engine, then it does what it should do. Is it so, that the outer sheave should not run at idle? It runs when idle and so engages the secondary clutch at idle; is that wrong?

 

[jansmeets33]

When you have it up and can get to the primary WITH ENGINE OFF can you move the belt easily on the primary clutch? Does the belt have a little play side to side against the sheaves?

Just to make sure: At the primary, when engine is off, the belts sits against, or at least very close to the outer sheave, but on the other side it has 5mm room. there is no slippage. This is the same when running at idle, but the outer sheave does run also and that way engaging with the belt the secondary clutch. Would that indicate that the primary is faulty or stuck or blocked?

 

[MBDiagMan]

Okay great! I misunderstood. Sorry.

I expect that your problem is not the shift problem if I understand correctly. Best of luck with it.

 

[jansmeets33]

Thank you very much! I have still the shift problem. My question is now: When the engine is running, say at idle speed, and in my case, while the secondary clutch is engaged and makes the transmission (gearbox) running, would that running stop me from shifting gear? So, my main core question is now: Is changing gear only supposed to do so, when the gearbox is not engaged? If that is the case, then I think that I understand that indeed the problem is with the primary clutch, which confirms your answer. So, is this scenario correct?

 

[jansmeets33]

It sure sounds like the belt spacing is wrong or the primary clutch needs attention. What I would need to get it right is both the primary clutch AND the belt so it can be properly set up.

Tank you! That is appreciated! The belt is 30 mm wide and the tightness is that it between the pulleys been pushed in, approx. 20 mm (max. 25 mm). I have cleaned with brake cleaner the belt & pulleys + inside the primary clutch, and it can now slip by hand a little, which was not possible before. Primary still engages belt at idle. Still feels tight.
When I move the belt to engine side and start the engine, then the belt walks to the outer side, but before it reaches there, then it has already engaged the belt. Previous owner has moved the gearbox at the top 4 mm forward (bolt-holes widened), but should I widen it a bit further? (move the 2 welding places) ? (and watching of course the alignment)? Or is there a better way of stopping the primary of taking the belt already at idle revs? The belt is original Deere and looks like unused but has some cracks.

 

[jansmeets33]

Start with a through cleaning of both clutch’s with hot soapy water and or non chlorinated brake cleaner followed by compressed air. You will need to remove the dome cover on the Drive clutch ( Primary) to clean properly. Use nylon brushes including a toothbrush to access the tight areas.

Cleaning is the first step in troubleshooting your clutch’s and will give you an opportunity to look for excessively worn or missing components.

Thank you! It's very apprciated! have cleaned the belt with Brake/Carb cleaner incl. the Primary inside and took the ramp plate off. Inside it looks all correct, apart from the 3 rollers have some spell on the axle, but that could be designed this way. (The clutch works when revs go up). It looks like the belt is too tight? See my comment just above this at MBDdiagMan

 

[200mph]

Are you certain the correct belt is installed?

Primary clutch works on centripetal force. If it’s engaging at idle:
1. The idle speed is too high and/or
2. The force balance between weights and spring is not right. Perhaps the spring is not generating the correct amount of preload to counteract the weight force…. Weak or fatigued spring? Any chance a shim can be added to spring to see if this would alter the clutch engagement rpm?
The fact the transaxle has been moved already suggests belt length is wrong.

 

[jansmeets33]

Tank you! That is appreciated! The belt is 30 mm wide and the tightness is that it between the pulleys been pushed in, approx. 20 mm (max. 25 mm). I have cleaned with brake cleaner the belt & pulleys + inside the primary clutch, and it can now slip by hand a little, which was not possible before. Primary still engages belt at idle. Still feels tight.
When I move the belt to engine side and start the engine, then the belt walks to the outer side, but before it reaches there, then it has already engaged the belt. Previous owner has moved the gearbox at the top 4 mm forward (bolt-holes widened), but should I widen it a bit further? (move the 2 welding places) ? (and watching of course the alignment)? Or is there a better way of stopping the primary of taking the belt already at idle revs? The belt is original Deere and looks like unused but has some cracks.

Are you certain the correct belt is installed?

Primary clutch works on centripetal force. If it’s engaging at idle:
1. The idle speed is too high and/or
2. The force balance between weights and spring is not right. Perhaps the spring is not generating the correct amount of preload to counteract the weight force…. Weak or fatigued spring? Any chance a shim can be added to spring to see if this would alter the clutch engagement rpm?
The fact the transaxle has been moved already suggests belt length is wrong.

Thank you! That is very appreciated!
1) The speed is low. The belt tension was 2 cm and was thus too tight. By trying 4 cm; it made no difference. I made it 5 and still no difference. When I turn by hand the primary (engine is off) then the primary moves the belt and engage the secondary, which should not happen, is my understanding?. When I turn the secondary by hand, which moves the belt, but the primary clutch does not move, and it slips there as I expected. I took the belt off and pushed the secondary wheel and there was no resistance, it kept running. I found rust on the primary axle. But initially I thought it was a brown coating. (possible that the previous owner had oiled it but is was hard and difficult to remove so I used sandpaper first 240 and after 1500 grid. But the rust pits are still visible. See attachment.

 

[jansmeets33]

Are you certain the correct belt is installed?

Primary clutch works on centripetal force. If it’s engaging at idle:
1. The idle speed is too high and/or
2. The force balance between weights and spring is not right. Perhaps the spring is not generating the correct amount of preload to counteract the weight force…. Weak or fatigued spring? Any chance a shim can be added to spring to see if this would alter the clutch engagement rpm?
The fact the transaxle has been moved already suggests belt length is wrong.

2) I have learned a lot, but I still do not know what the force balance between weights and spring really is?
With the engine off, and the gear in neutral I feel no resistance both ways when pushing the secondary wheel to turn in regard to that the primary wants to engage even by moving it 1 cm. by hand, while the belt is now in the middle and does not touch the side walls. The axle after cleaning is reasonably smooth as you can see in the picture. I also turned the belt around the other way, but it made no difference, and moved the engine to test the drive belt tensions. Hopefully that there is a solution somehow?

 

[jansmeets33]

Are you certain the correct belt is installed?

Primary clutch works on centripetal force. If it’s engaging at idle:
1. The idle speed is too high and/or
2. The force balance between weights and spring is not right. Perhaps the spring is not generating the correct amount of preload to counteract the weight force…. Weak or fatigued spring? Any chance a shim can be added to spring to see if this would alter the clutch engagement rpm?
The fact the transaxle has been moved already suggests belt length is wrong.

The belt installed is no, JD RE2872.

 

[200mph]

The forces I’m talking about are the ones acting within the primary itself. Under the outer cover is a spring that forces the pulley away from the belt. Sometimes friction prevents the pulley moving to its full open position. (Return spring)

For belt engagement to occur at speed, weights are used to create centripetal force to push the pulley half toward the other. This force is greater than the return spring so engagement occurs when engine Rpms increase. The higher the rpm the greater the force pushing the movable pulley towards the other squeezing the belt towards the top of the pulley.

If the spring is broken or weak the pulley will contact the belt at too low of an rpm and cause drag.

There are some good videos on YouTube showing how a CVT works (same type used on Gator, snowmobile, go carts, etc. )

Once you understand how they work everyone’s comments will make more sense.

The transaxle function is to select F/R/N. It does no shifting like an auto or manual transmission. Ratio changes are done by the clutch (CVT) including it to slip while at a stop and in gear.

You can check belt number by looking up John Deere illustrated parts list on line. www.jdparts.com

Note any PIN breaks if specified in the pn description or notes.

Hope this helps.

 

[jansmeets33]

When you have it up and can get to the primary WITH ENGINE OFF can you move the belt easily on the primary clutch? Does the belt have a little play side to side against the sheaves?

Now, after adjustments, the belt moves easily and has little play against the sheaves. But still engages the belt with low raves? The secondary clutch has no resistance and turns over very easily, with the result that I cannot change into gear properly. With the ENGINE OFF, and moving the secondary clutch by hand forward or backward, then the belt moves (slip) easily over the primary clutch. But if I move the sheaves of the primary, then the primary moves the belt and engages the secondary clutch. The primary does this without touching the side sheaves. The belt sits only on the axle. (It does this with a belt tension of 4 cm. and also with 5 cm tension) Hard to believe, but the secondary sheaves have total NO RESISTANCE, hence that it so easily turns, Is that normal ??? That way, I cannot change in to gear, because the primary turns. When running idle, I can easily stop the secondary from running, which I did with a piece of timer over the edge of the secondary sheaves. So what is happening here? How can that be solved. The axel of the primary is now reasonable clean and shiny, and the side walls are like polished. Any help is very appreciated. Thanks!

 

[jansmeets33]

The forces I’m talking about are the ones acting within the primary itself. Under the outer cover is a spring that forces the pulley away from the belt. Sometimes friction prevents the pulley moving to its full open position. (Return spring)

For belt engagement to occur at speed, weights are used to create centripetal force to push the pulley half toward the other. This force is greater than the return spring so engagement occurs when engine Rpms increase. The higher the rpm the greater the force pushing the movable pulley towards the other squeezing the belt towards the top of the pulley.

If the spring is broken or weak the pulley will contact the belt at too low of an rpm and cause drag.

There are some good videos on YouTube showing how a CVT works (same type used on Gator, snowmobile, go carts, etc. )

Once you understand how they work everyone’s comments will make more sense.

The transaxle function is to select F/R/N. It does no shifting like an auto or manual transmission. Ratio changes are done by the clutch (CVT) including it to slip while at a stop and in gear.

You can check belt number by looking up John Deere illustrated parts list on line. www.jdparts.com

Note any PIN breaks if specified in the pn description or notes.

Hope this helps.

Thank you! That did really help my understanding. Great! Also, I looked up the belt, I have the correct RE28721 belt according to the JD Manual, you referred to.
The Ramp Plate has some wear from the rollers but did not look excessive and while raving the engine up to full speed; the belt came up on the primary and went down on the secondary, Did you see the attached picture of the rusted axle after cleaning and wonder if that could be the problem or that the belt is the problem or that the secondary is missing resistance during idle revs? Hopefully that you see what the real problem is here and why the secondary has no resistance while idling, just in case that this is a needed feature?