New "TSDZ2 Torque Sensor Central Motor"

Huguenot said:
izeman said:
Huguenot said:
Back to the loose and or wobbling crank axle problems, and thanks to safeaschuck for posting the detailed pics (way) above.
I have TSDZ2 on my wife's cargo bike, and other than some initialization problems seemingly fixed, if not totally eliminated, by a firmware upgrade (stellar customer service from EcoCycles, BTW) -she loves it. ....

So I would like to discuss some alternatives.

.....
This is a discussion I'd like to have as well. As I heard of broken axles, bad fits, loose bearings, and stuff like that I guess making a new axle would be the best way to do it. Using proper steel would help a lot. But I haven't opened the motor or removed the axle so I have no idea how hard it would be. Lathe and mill are ready and waiting :)

Hello izeman and others.
Regarding the broken shafts, I believe the steel used is less of a problem than the snap ring groves, and the fact that the crank mounting bolt holes are deeper than they need to be, at least on some of them. I am most concerned about the wobble/clunk and the wear it is causing on other parts, like the disc portion of the sensor. I was quite surprised at how soon this problem showed up on our bike, as some people report far more miles/km before looseness.
Given the design of the torque sensor there does not appear to be much that can be done from a redesign standpoint, but I am open to ideas.
The hardest thing to duplicate would be the splines, but they are short, few in number and stand proud of the shaft itself so should be less difficult than they might be. IMO, just controlling the seat diameters so the result is closer to a press fit would suffice for many people.
I just ordered a replacement axle from Eco Cycles ($25) to experiment with. Once I figure out just what parts of the shaft are hardened I may try building up the bearing seats somehow, likely with just solder, or perhaps by turning the shaft down a bit and pressing on a thin bushing, then turning down to correct diameter.
Again, the loose fit is probably done for assembly purposes as it saves the manufacturer exactly nothing to make it too small at the piece part level.
While the axle is no doubt made on CNC equipment, final assembly is no doubt done by hand.

According to the link below it must be a functional play.

https://www.emtbforums.com/community/threads/e8000-bottom-bracket-bearing-play-after-1400km.643/

This can't be right since it's a completely different motor.
 
rosenbaaron said:
... The issue I'm having is that if i charge my battery past full then it won't engage when I turn it on. ...

This is a frequent issue with the TSDZ2. The thought is that the part within the controller that reports the voltage has a very loose tolerance, and sometimes it's fine, and, sometimes it's not. With the factory firmware, my 48V unit works fine as long as the battery is only charged to <= 54.0V (full charge ~54.6V). This is one of the issues that the Open Source Firmware gets past, and I have tested it to do so, but I presently prefer the factory firmware for other reasons. The only recourse, without swapping parts yourself, is to return it for adjustment/replacement, and I have heard of a few cases (pre-pandemic) where this was successfully done, but, it takes lots of time for the round trip. My controller reports 1.5 volts higher than actual. Sorry the news isn't better, but I am now well adjusted to charging less than 100%, which I am told provides a longer life from the battery pack anyway.
 
Just snapped an axle at 540kms.
Anyone know where to buy just an axle by itself or the cheapest place to get a torque sensor unit?

received_271223474239613.jpeg
 
Chewy said:
Just snapped an axle at 540kms.
Anyone know where to buy just an axle by itself or the cheapest place to get a torque sensor unit?

https://www.eco-ebike.com/collections/tsdz2-accessories-parts/products/axle-replacement-for-tsdz2-stock-for-68-73mm-bottom-brackets

Is that a chain guide on your bike? Im having some issues with chain drop on my FS bike during the downhill parts, so I've been thinking about making some sort of guide. Could you share some details?
 
Huguenot said:
The hardest thing to duplicate would be the splines, but they are short, few in number and stand proud of the shaft itself so should be less difficult than they might be. IMO, just controlling the seat diameters so the result is closer to a press fit would suffice for many people.
I now checked the picture from the eco-cycles webshop and see what you mean. Those splines will be hard to replicate without the right tools or a cnc lathe/mill.
 
Huguenot said:
...I am most concerned about the wobble/clunk and the wear it is causing on other parts, like the disc portion of the sensor. I was quite surprised at how soon this problem showed up on our bike, as some people report far more miles/km before looseness...
That's my main concern as well. Fixing the inappropriate tolerances so the wobble goes away. Little wobble leads to more wobble on the long run. That's inevitable. And making the whole thing from proper steel so it won't break is good for your health. A broken axle can lead to really bad injuries, as it brakes in moments of highest stress - when you stand on the pedal. :shock:
 
Chewy said:
Just snapped an axle at 540kms.
Anyone know where to buy just an axle by itself or the cheapest place to get a torque sensor unit?
As far as I'm aware the axle comes as part of the torque sensor assembly:

http://www.pswpower.com/ven.php?cargo.2016-7n-n002

Can you share any further info on where the axle snapped? Was it at the circlip groove? Also, had you fitted the additional bearing on the right hand end of the axle behind the dust seal?

As spares seem quite expensive to source at the moment with shipping costs etc, I took the decision recently to buy a whole new motor to replace one of mine with the most miles on it, then I have the old motor as a complete set of spares when needed, not the cheapest solution agreed, but will probably pay off for me in the long run, plus no waiting the get the bike back up and running when something fails.
 
As posted above, you can buy the axle alone from eco-bike. Only makes sense when you're based in the US. P&p double the price for shipping to Europe. And you still get the same junk as was installed originally. Same issues.
I once had an axle of a MAC hub motor break (it has this large axle bore for the phase wires). Made a new one myself and never had an issue. Maybe this is a winter's project. Remove the motor, remove the axle and see if it can be replicated.
 
HrKlev said:
Chewy said:
Just snapped an axle at 540kms.
Anyone know where to buy just an axle by itself or the cheapest place to get a torque sensor unit?

https://www.eco-ebike.com/collections/tsdz2-accessories-parts/products/axle-replacement-for-tsdz2-stock-for-68-73mm-bottom-brackets

Is that a chain guide on your bike? Im having some issues with chain drop on my FS bike during the downhill parts, so I've been thinking about making some sort of guide. Could you share some details?

Its an old Roox chain
dog lower roller. I've just cut down the mount and bolted it on to the lower shock mount.
It keeps the chain on most of the time, but ive ordered a 104bcd adapter and am going to run a narrow wide ring and hopefully that stops all dropped chains.


Dug a little deeper and found my sprag bearing is stuffed, and the brass gear is wearing at an alarming rate. I've put the blue gear back in.

received_2673628242881301.jpeg

received_694455788052001.jpeg
 
devboy-greg said:
Chewy said:
Just snapped an axle at 540kms.
Anyone know where to buy just an axle by itself or the cheapest place to get a torque sensor unit?
As far as I'm aware the axle comes as part of the torque sensor assembly:

http://www.pswpower.com/ven.php?cargo.2016-7n-n002

Can you share any further info on where the axle snapped? Was it at the circlip groove? Also, had you fitted the additional bearing on the right hand end of the axle behind the dust seal?

It snapped on the drive side at the circlip groove.
The crack appears to have been propagating for a while looking at the fracture surface.
The axle snapped landing a decent sized drop on a downhill track.
I have not fitted the extra bearing and do not believe it would have any effect on this failure as it broke on the opposite end of the axle.

received_341607270143101.jpeg
 
Chewy said:
devboy-greg said:
Chewy said:
Just snapped an axle at 540kms.
Anyone know where to buy just an axle by itself or the cheapest place to get a torque sensor unit?
As far as I'm aware the axle comes as part of the torque sensor assembly:

http://www.pswpower.com/ven.php?cargo.2016-7n-n002

Can you share any further info on where the axle snapped? Was it at the circlip groove? Also, had you fitted the additional bearing on the right hand end of the axle behind the dust seal?
It snapped on the drive side at the circlip groove.
The crack appears to have been propagating for a while looking at the fracture surface.
The axle snapped landing a decent sized drop on a downhill track.
I have not fitted the extra bearing and do not believe it would have any effect on this failure as it broke on the opposite end of the axle.

Thanks for the extra info and pictures.

I have my TSDZ2 fitted to an enduro bike and I'm fearful of the axle snapping on drops, but assumed it would be the none drive side that would go first as it has the longer segment past the circlip and therefore more leverage, hence why I have the extra bearing fitted offering support past the circlip.

No such remedy for the drive side though other than better quality steel, I guess we are pushing the motor beyond what it is designed for.

Others on this forum are talking about machining a better quality axle, I would be interested in purchasing one if it can be produced at a reasonable price.

Regarding the brass gear, I had the same problems as you and switched back to the blue gear some time ago and have not managed to trash it yet! I do have brake sensors installed though.
 
izeman said:
Huguenot said:
The hardest thing to duplicate would be the splines, but they are short, few in number and stand proud of the shaft itself so should be less difficult than they might be. IMO, just controlling the seat diameters so the result is closer to a press fit would suffice for many people.
I now checked the picture from the eco-cycles webshop and see what you mean. Those splines will be hard to replicate without the right tools or a cnc lathe/mill.

CNC equipment is not needed for a one-off of the round parts of the axle, and I believe the end tapers and spacing of the splines can be handled on a mill with a collet indexer. The actual profile of the splines may require a custom cutter however.
All doable if one is determined enough, I will report back once my replacement axle arrives.
 
One thing that might help prevent shaft breakage is to shot peen the shaft. Cracks start at the outer surface and move in. If you compress the surface by shot peening then the surface is compressed and doesn't go into tension as the shaft flexes slightly during use. This is standard procedure for parts like connecting rods in gasoline engines or axle shafts.
Another option for building up a bearing journal is the spray molten metal onto it. This is done to build up worn crankshaft journal in automotive machine shops. Proably not DIY, but commercialy available.

Luna Cycles in the US is offering a nice 36V. 500W belt drive bike which is more like a European bike than the BBSHDs they're famous for.
The TSDZ2 never caught on here. I thought you guys might like this one even though it's not TS.
https://lunacycle.com/luna-fixed-stealth-ebike/
 
devboy-greg said:
No such remedy for the drive side though other than better quality steel, I guess we are pushing the motor beyond what it is designed for.
That is an understatement Indeed.
This motor is very good for light city use but totally inappropriate for MTB.
I'd even say that the max power should remain below 350w to be sure to have a reasonable life span.
I really like these little motors but using it out of its capabilities results in autodestruction.
 
One more pic. The biggest problem is the circlip groove. It is a massive stress riser and is where the crack starts to form. I can't imagine shot preening working here.
One thought I have had is to tig weld up the circlip groove and to find another way of locating the axle. Or at least tig the top side of the grove that is under tensile force when you have your pedals in the level position when landing jumps etc. and then use a modified circlip.

I ride enduro style terrain with the TSDZ on a 6" travel bike and have always been very mindful not to overstress the axle but they are just very weak.

received_366296464345731.jpeg
 
The circlip groove could be radiused at the bottom to remove the sharp corners, or a 1/2 round groove cut and a round wire snap ring used instead. I didn't notice the snap ring groove in the photo just the fatigue fracture pattern.
https://www.arconring.com/wire-snap-rings/open-closed-gap/

Then next question made me revisit this problem. What about drilling in from the end through the bolt hole, well past the snap ring groove, and press fitting a piece of steel rod in there so the crack can't go all the way through. It may still break but it would be a soft failure. Just thinking out loud here.
 
izeman said:
[/quote
Another thing that makes me worry: The supplied screws for the cranks are like butter. Normally those screws need to be screwed down pretty hard to prevent the cranks from wobbling. And I once had to add some more turn after a short ride and it stripped all threads from the screw. Now I'm afraid the inner thread my be damaged as well - any idea what to do?
Next thing will be heating the crank and then quickly pressing it on. Maybe that helps. What the recommended torque for that screw?
Better bolts should be readily available at any decent harware or bike shop. They come in different strength ratings. 10.9 is a good metric rating. 8.8 is common, and unmarked bolts are crap.
If the internal threads are damaged, a bike shop or auto mechanic should be able to drill out the damaged threads, tap the hole larger and install either a Helicoil spring insert, or a threaded repair bushing. The Helicoil method removes less metal ,and works quite well.
 
Retrorockit said:
izeman said:
[/quote
Another thing that makes me worry: The supplied screws for the cranks are like butter. Normally those screws need to be screwed down pretty hard to prevent the cranks from wobbling. And I once had to add some more turn after a short ride and it stripped all threads from the screw. Now I'm afraid the inner thread my be damaged as well - any idea what to do?
Next thing will be heating the crank and then quickly pressing it on. Maybe that helps. What the recommended torque for that screw?
Better bolts should be readily available at any decent harware or bike shop. They come in different strength ratings. 10.9 is a good metric rating. 8.8 is common, and unmarked bolts are crap.
If the internal threads are damaged, a bike shop or auto mechanic should be able to drill out the damaged threads, tap the hole larger and install either a Helicoil spring insert, or a threaded repair bushing. The Helicoil method removes less metal ,and works quite well.
In addition, remember to grease the Spindle and crank. This allows them to slide easily to increase the pressure on the taper while reducing the load on the bolt and the threads.
 
Huguenot said:
izeman said:
Huguenot said:
The hardest thing to duplicate would be the splines, but they are short, few in number and stand proud of the shaft itself so should be less difficult than they might be. IMO, just controlling the seat diameters so the result is closer to a press fit would suffice for many people.
I now checked the picture from the eco-cycles webshop and see what you mean. Those splines will be hard to replicate without the right tools or a cnc lathe/mill.
CNC equipment is not needed for a one-off of the round parts of the axle, and I believe the end tapers and spacing of the splines can be handled on a mill with a collet indexer. The actual profile of the splines may require a custom cutter however.
All doable if one is determined enough, I will report back once my replacement axle arrives.
If you say so I have no doubt to not believe you, and hope you are right. The pictures I've seen are all too small or blurry to really see what those splines look like. If someone could make some high res full size pics of the details it would be easier to say. I have a regular lathe and mill only (and indexing table etc ...) but NOW it's time to ride the bike and not take it apart.
If someone has a broken axle and is willing to ship it to me I can have a look at it and try to replicate it.
 
Retrorockit said:
Better bolts should be readily available at any decent harware or bike shop. They come in different strength ratings. 10.9 is a good metric rating. 8.8 is common, and unmarked bolts are crap.
If the internal threads are damaged, a bike shop or auto mechanic should be able to drill out the damaged threads, tap the hole larger and install either a Helicoil spring insert, or a threaded repair bushing. The Helicoil method removes less metal ,and works quite well.
Thanks for the hint. I know that I could just buy better quality screws. It's a regular fine pitch thread M8x1.0. Those Chinesium grade screws supplied with the kit may serve a purpose: Better strip the (cheap) screws before you damage the (expensive) axle? If I use 10.9 screws then probably the axle would fail before the screw?! Taking into account that the tap hole is already weakening the axle a lot, I think Helicoil inserts would worsen this issue even more, don't you think so?

My question should have been more precise: Have others also experienced screw stripping and falling off cranks? Or am i the only one?

IMHO the only proper way is to make a new axle from ground up. Circlip grooves DO add a lot of stress, but with an axle made of proper steel it still should work pretty well. Round ring clips are a bit of an issue though as they have only a limited amount of "hold force" to the side.
 
RicMcK said:
In addition, remember to grease the Spindle and crank. This allows them to slide easily to increase the pressure on the taper while reducing the load on the bolt and the threads.
Thanks. I didn't grease them. Good point. I think I remember someone mentioned to heat the crank up with a propane flame (or eq) before mounting. This will result in a press fit. Any experience doing such?
 
izeman said:
Retrorockit said:
Better bolts should be readily available at any decent harware or bike shop. They come in different strength ratings. 10.9 is a good metric rating. 8.8 is common, and unmarked bolts are crap.
If the internal threads are damaged, a bike shop or auto mechanic should be able to drill out the damaged threads, tap the hole larger and install either a Helicoil spring insert, or a threaded repair bushing. The Helicoil method removes less metal ,and works quite well.
Thanks for the hint. I know that I could just buy better quality screws. It's a regular fine pitch thread M8x1.0. Those Chinesium grade screws supplied with the kit may serve a purpose: Better strip the (cheap) screws before you damage the (expensive) axle? If I use 10.9 screws then probably the axle would fail before the screw?! Taking into account that the tap hole is already weakening the axle a lot, I think Helicoil inserts would worsen this issue even more, don't you think so?

My question should have been more precise: Have others also experienced screw stripping and falling off cranks? Or am i the only one?

IMHO the only proper way is to make a new axle from ground up. Circlip grooves DO add a lot of stress, but with an axle made of proper steel it still should work pretty well. Round ring clips are a bit of an issue though as they have only a limited amount of "hold force" to the side.

A better axle would be better. Or a new replacement. But if you use the cheap bolts again ??? As far as overtightening the better bolts. They don't tighten themselves. Use a torque wrench. That may be the original problem anyway. The fine thread will produce more clamping pressure but in a soft material will strip easier. The grease is also a good idea to keep moisture out of an Aluminum joint. Aluminum oxide is harder than steel. A little salt water up in there and the bolts will be the least of your problems.

Another approach to the snap ring stress riser problem would be to use a longer bolt and move the threads in just behind the snap ring groove so the tension on the bolt loads the snap ring groove in compression. Whether this will be enough to eliminate the bending (tension) load from the crank arm is a mathematical question, but it should reduce it, and again the failure mode would not be a full break. I don't think the amount of metal is the problem. All bikes have a shaft that size, and many are drilled all the way through. If you can keep the crack from starting it should be good. For most people manufacturing a new axle is not a practical option.
If the part being retained by a wire circlip has a bevel or radius this drives the wireclip down into it's groove and they tend to stay put. I'm not aware of any great thrust load in these motors ( no tapered roller bearing to be found). A displaced wire clip is a lot easier to fix than a broken axle (or leg). Technically if the wire groove was rolled into the shaft instead of cut it would actually increase the strength of that area. But not DIY for most people ( or Tong Sheng apparently).
 
That's why I asked for the correct torque the first time, but haven't received an answer yet. I can't find a number.
Of course a proper built axle can be used with proper screws as well without risking stripping the gears. It may be that my internal threads are already somehow damaged, which leads to easily stripped threads of the screw now. Inserting a taper is quite a wobbly thing instead of a proper tight fit :(. And yes, the whole set is more or less brand new. Replacing the blue gear led to removing the crank three times now and therefore making the situation worse.

When you say that making a new axle is no alternative for most people i have to agree. But for me it's a good way to find out where the problems are. Maybe this will find it's way into production of future axles by Tongsheng, or someone with industrial equipment will take our knowledge and produce better products for "the masses".
 
izeman said:
That's why I asked for the correct torque the first time, but haven't received an answer yet. I can't find a number.

My understanding is that these bolts should be torqued to 38-42nm (or 28-31 ft lb.). The taper should be free of burrs and lightly greased.
The hardware that comes with the kit actually has this value marked in small print right on the black plastic collar that surrounds the bolt head.
(Indeed, many bicycle parts seem to have torque values indicated right on the item these days which is a nice feature, especially when dealing with a carbon frame.)
When installing, at least new parts for the first time, this can seem like a lot of torque, but riding for even a short while with a loose crank arm will make it unusable very quickly, so worth paying attention to.
Bafang crank bolts should be torqued to the same value, BTW as they carry identical markings.
 

Attachments

  • IMG_4442.jpg
    IMG_4442.jpg
    64.4 KB · Views: 1,204
There are torque standards for all grades and threads of bolts, and charts that list them.
https://www.fastenermart.com/files/metric_tighten_torques.pdf
The suggestions I'm making can not only be applied to a replacemnet part, but incorporated into a new product. I'm a mechanic, not a machinist, or engineer. I've seen a lot of broken parts. I've also seen what doesn't break. I've also done a lot of high performance modifications. Brakes, steering, suspension and engines, all things you don't want to make worse, especially at high speeds. I have read some engineering text book, or books written by engineers for guys like me, to see what the designers had in mind before I modify things.
It really shouldn't be necessary to do anything but grease the parts and properly torque the fastener for the crank arms to stay put. If they don't then there is a parts problem. A Helicoil will be stronger than the original thread. A larger diameter in the part and the clamping load is distributed the full length instead of the first 3 threads. This is your best chance for taking advantage of a better bolt. But if there's something wrong with the fit of the taper then it wont fix that. You could get some machinist blue dye and paint the inside of the crank arms, R&R them and see whats going on there. I think some of the motors had a shorter taper area to make the axle narrower. These may be the single bearing versions. It's possible you need to do something to get the crank to seat fully on the taper. Make sure the bolt isn't bottoming out in or against the axle. If you think drilling a larger hole is a problem, then go deeper, rethread and get a longer bolt.
 
Back
Top