New "TSDZ2 Torque Sensor Central Motor"

[Waynemarlow]

Just as an aside have a look at the Perforamce Bearings of a strip down and upgrade to a Brose motor and you can see a lot of what we would need to do as far as waterproofing and fitting a belt drive.

https://www.emtbforums.com/community/th ... ors.19102/

[jbalat]

Do you have anything on how the Bosch Gen 4 works (Bosch CX 2020) since this totally disconnects and gives zero drag, or the motor used in the specialised Levo SL ?

Actually there are some pics here of the SL motor but hard to see whats going on
https://flowmountainbike.com/tests/firs ... tain-bike/
https://bikerumor.com/2020/02/04/first- ... ight-emtb/

And I found some on the Bosch
https://www.e-bikeshop.co.uk/blog/post/ ... bike-2020/
https://www.bosch-ebike.com/en/products ... e-line-cx/

[Waynemarlow]

The Bosch motor is the only one of the two I have ridden with and it has a truely irritating rattle from the engine when it freewheels, I’m not sure it actually fully disengages, but uses a sprag clutch like the Brose and it’s that we can hear much like the rear hubs when not under load.

Whatever, it is almost academic with the TSDZ2 motor, it is what it is and has two sprag bearings, yes they are going to have some drag when the motor is off, but do we really care. From experience I have had only ever had to cycle home twice with a dead motor ( both times water ingress related ) and it is more the fact that we use a much larger front gear than an analogue bike, that makes it feel as though you are riding an overweight over geared tank.

[ebikeman123]

Here is my review of the TSDZ2 750W:

It worked fine in the beginning, but the power from throttle only was very underwelming.
A big problem soon came up too: Chains and cassettes get worn out surprisingly fast. Maybe every 300km, the chain starts jumping on the cassette on the smallest cog first, then on larger and larger cogs. Extremely frustrating and expensive to have to change these components every 2 months.

Then, after maybe 600km, the speed reading stopped working. No idea why.

Shortly after that, the thumb throttle cracked (never used excessive force)

A couple months later the sprag clutch (one-way bearing) started slipping. So I had to use only the throttle, even if it was uncomfortable and slow.

Then, finally the blue plastic gear gave in after having the motor for only about 9 months.

Luckily I had bought an extra blue plastic gear. Unfortunately, when I tried to open the motor, it was impossible because the screws stripped their heads when I used a screwdriver(maybe they used loctite or something, but the metal extremely weak). Have still not been able to open the motor, and it is just lying around as junk.

Will never buy this motor again or a mid-drive because of the maintenance nightmare. The quality is incredibly low. I would say 70% of the things I have bought from Aliexpress/dealextreme/banggood etc have been pure trash.

After this I have tried a geared rear hub motor (mac). It worked really nice in the beginning, but it only lasted 6000km (7 months), before everything just stopped working. It shorted itself because it apparently traps and condenses water (when I opened it lots of water was trapped for some reason) Also the axle bent by itself, causing the cogs to disintegrate because of misalignment. More Chinese crap. They use the cheapest steel possible.

Will maybe try a direct drive next time. Are there any high quality motors out there that can last a long time with little maintenance?

[jbalat]

Hi ebikeman thanks for your thoughts I hear what you are saying and its unfortunate.
I think the tsdz2 is the wrong motor choice if you want throttle and the fact that you were even considering throttle tells me that you probably run very high assist levels in which case you should have just bought a bafang750w motor. The TSDZ2 is not designed for continuous high power output despite the ratings they are sold with.

Stay away from geared hub motors unless you get something big like a Direct Drive crystalite motor which will last forever, however the extra weight on the back and the problems with fixing punctures means I will just keep away.

I have more than 20k on my tsdz2, yes I changed the sprag clutch and blue gear early on and only gone through 1 rear cassette and 3 chains maybe but I don't expect much from the motor and never stress it out. I usually ride with about 200w of assist and only exceed 300w up hills and against strong winds, make sure chain line is good and keep chain well lubricated.

You should be able to drill through the tops of the screws and try using easy outs or even better file some flats on the sides of the heads and use some pliers to take the old screws out then replace them with good ones. Yes they are crap.. the trick is to use sharp machined hex drivers and not cheap alan keys

[Wapous]

Subject: Study for the replacement of the gear mechanism by a belt drive mechanism.

Specifications:
1- A mechanism easy to implement without major modifications.
2- The mechanical efficiency (torque) must be similar.
3- The noise of the mechanism must be less being the desired goal.
4- Establish the costs of this modification.

There is still a lot of research to be done. Check with suppliers for available parts and custom parts cost.

Please note that the belt teeth have not yet been drawn. In this design, the pinion gear respects the minimum number of 6 traction teeth recommended for timing belts.

It will not be done for tomorrow but at least we have the general idea.

If you have any comments ....

[casainho]

1- A mechanism easy to implement without major modifications.

More modifications will mean an expensive modification. See the TSDZ2 is very special motor because is very cheap and very flexible - other commercial mid drive motors are much more expensive, because you have to consider price for the batteries and displays that are usually proprietary (you do not have any alternative), also the price for maintenance / repair parts that in some case the final users simple can not buy them!! - and also some motors are very hard or expensive to unlock the power and speed limit -- TSDZ2 is very special when running our developed OpenSource firmware, we have even now have the TSDZ2 wireless that have similar functions as this BOSCH advanced display with GPS navigation and user heart rate:



For TSDZ2 we have our mobile app and we integrate with Garmin wireless GPS displays and so user have the top quality GPS navigation from Garmin as also user heart rate, including the EBike TSDZ2 data -- here a ride I did with my TSDZ2 EBike where I am monitoring my heart rate for fitness purposes:



What I want to say is that TSDZ2 is very cheap and even if this modification is a bit expensive, will still for sure bring a good value for users - and they are free to do it or not, will be an optional cost.

I will be able to help make a custom firmware that will rotate the motor on the inverse direction.

About the main gear plastic cover, I am pretty sure it can be 3D printed as I think I already saw someone doing it.

Can you do it all by yourself? design the gears, machine them, install and test? this would be great, you would me able to move quick!! and simplify, do not try to make everything perfect / follow the 4 specifications you wrote in the first iteration.

[Waynemarlow]

About the main gear plastic cover, I am pretty sure it can be 3D printed as I think I already saw someone doing it.

It’s those two covers on that side of the motor which give me most grief with water ingress, I seem to have tried a dozen ways to stop water getting behind that cover, I’ve slowed it down but never fully stopped it. Perhaps we could have a total redesign to integrate the two covers into just a single cover.

[James Broadhurst]

A big problem soon came up too: Chains and cassettes get worn out surprisingly fast. Maybe every 300km, the chain starts jumping on the cassette on the smallest cog first, then on larger and larger cogs. Extremely frustrating and expensive to have to change these .........
Luckily I had bought an extra blue plastic gear. Unfortunately, when I tried to open the motor, it was impossible because the screws stripped their heads when I used a screwdriver(maybe they used loctite or something, but the metal extremely weak). Have still not been able to open the motor, and it is just lying around as junk.

Sorry to read of your troubles. You’ve nearly done all the things that I’ve done but at least you must have got the bottom bracket of to fit it which is more than I did on one bike.

I use a Shimano rear derailleur with a clutch which applies tension to the chain and I’ve done thousands of miles without having to change either cassette or chain. Whether that is down to the clutch I don’t know but I suspect it is. The only issue is that these clutches are only fitted to >9 speed, or so I believe, so your bike needs to have a cassette on its rear wheel. The presumption might be that 7 or 8 speed tourney gears have such a small difference that the chain never loses tension but with electric bikes I’ve seen one or two where the chain has become so slack with hooked gear teeth that the chain jumps off easily.

I stripped the heads on the motor cover M4 screws (non-drive side) and took an angle grinder to it in a bit of a tantrum. Subsequently and entirely caused by old age and arthritis I bought a cheap-ish battery drill with decent speed control and a good set of metric Allen head bits. Pressing hard into the head and operating the drill works every time. Applying pressure on the Allen key is something that’s difficult to do by hand.

[Wapous]

Can you do it all by yourself? design the gears, machine them, install and test? this would be great, you would me able to move quick!! and simplify, do not try to make everything perfect / follow the 4 specifications you wrote in the first iteration.

[/quote]

This design shown here is just a concept.
For example if the ideal width of the belt versus the torque applied is unrealistic according to the manufacturers, then a new design will have to be imagined. The belt is GT3 type which ensures maximum grip and minimum noise. But this type of belt must suit our technical specifications and dimensions and have durability.
The diameter of the pinion gear is at its maximum and engages at least six of its teeth with the belt to avoid the ratchet effect. It would be a challenge if we had to change the diameter of the pinion gear.
So, I still have to refine this design before sending the technical file to two or three manufacturers. They need the 3D views as well as the technical drawings.
Regarding the production by myself of this concept I must decline because I live in a condo without a mechanical workshop. I am very interested in purchasing a 3D printer from Prusa Company this spring.
Having said that, if I can get the prototype parts cheaply from the manufacturers, I will spend the money. For assembly and testing there is no problem if I have the components in hand.
As this concept seems to interest you, I will finish the drawings presented here and send them to different suppliers and manufacturers for quotations and feasibility.

[manoz]

A big problem soon came up too: Chains and cassettes get worn out surprisingly fast. Maybe every 300km, the chain starts jumping on the cassette on the smallest cog first, then on larger and larger cogs. Extremely frustrating and expensive to have to change these .........
Luckily I had bought an extra blue plastic gear. Unfortunately, when I tried to open the motor, it was impossible because the screws stripped their heads when I used a screwdriver(maybe they used loctite or something, but the metal extremely weak). Have still not been able to open the motor, and it is just lying around as junk.

Sorry to read of your troubles. You’ve nearly done all the things that I’ve done but at least you must have got the bottom bracket of to fit it which is more than I did on one bike.

I use a Shimano rear derailleur with a clutch which applies tension to the chain and I’ve done thousands of miles without having to change either cassette or chain. Whether that is down to the clutch I don’t know but I suspect it is. The only issue is that these clutches are only fitted to >9 speed, or so I believe, so your bike needs to have a cassette on its rear wheel. The presumption might be that 7 or 8 speed tourney gears have such a small difference that the chain never loses tension but with electric bikes I’ve seen one or two where the chain has become so slack with hooked gear teeth that the chain jumps off easily.

I stripped the heads on the motor cover M4 screws (non-drive side) and took an angle grinder to it in a bit of a tantrum. Subsequently and entirely caused by old age and arthritis I bought a cheap-ish battery drill with decent speed control and a good set of metric Allen head bits. Pressing hard into the head and operating the drill works every time. Applying pressure on the Allen key is something that’s difficult to do by hand.

I replaced those screws with socket head screws so require a 3mm allen key to remove them and use grease on them since the motor takes a pounding with the British weather. I also use purple extreme on my chains and coat the teeth on the 4 smallest cogs with the same. Purple extreme is the best stuff imo.
My problem I'm having now is that there is stickiness in the rotation of the cranks when I initially start off but it isn't there when I'm going faster. Dunno if this is down to the blue gear wearing out but I would rather know before I open up the motor and replace it. Perhaps it needs fresh grease?

[manoz]

About the main gear plastic cover, I am pretty sure it can be 3D printed as I think I already saw someone doing it.

It’s those two covers on that side of the motor which give me most grief with water ingress, I seem to have tried a dozen ways to stop water getting behind that cover, I’ve slowed it down but never fully stopped it. Perhaps we could have a total redesign to integrate the two covers into just a single cover.

File or cut some notches on the bottom of the motor casing, just under the controller. That way water can go straight out. Use a bit of duct tape to make a sort of valve so water can't be kicked up inside but can be expelled. When I come home after a rainy day of work I'm left with a puddle on the floor but would rather have the water come out then stay in. I've had no corrosion or controller failure since I did that.

What I was finding was that my controller was stuttering or failing due to water and when I would get home and open up the casing a whole load of water would come out and I would have to leave the motor cover off and let the whole thing dry overnight and it would work next day. Cutting notches has stopped that faff.

[Waynemarlow]

Interesting. Sadly I ride at this time of year in mud and large puddles enough to worry about forcing water up into motor area if I drilled a hole in the bottom. To be honest I think I have resolved that particular problem, the water was getting in through the silicone that packs around the wires coming through case to the outer compartment, before leading off in either of two directions. I simply packed over and around the wires with water proof grease with enough height that when the cover went on it compressed the grease. Do bare in mind though there is a hole high up where the torque sensor wires run through the case to the controller. I think if water is getting throughhere you would needed to have totally submerged the motor ( which one of lads did, missed a bridge ), I have found no real way to stop that as the wires have to allow for adjustment for wear on the faces of the torque sensor and some way of allowing pressure to eqaulise in that inner compartment from temperature changes.

The big one now is to stop water getting into where the big gear and pinion gear is. The X style seal doesn't seem to fully stop water or so I thought. Having watched the Brose video it looks like they have been having exactly the same problem and its not the large X seal but in and around the bearings themselves. It seems that they have started fitting X seals between the peddle and the bearing face. I need to look at that in more detail.

[manoz]

Interesting. Sadly I ride at this time of year in mud and large puddles enough to worry about forcing water up into motor area if I drilled a hole in the bottom. To be honest I think I have resolved that particular problem, the water was getting in through the silicone that packs around the wires coming through case to the outer compartment, before leading off in either of two directions. I simply packed over and around the wires with water proof grease with enough height that when the cover went on it compressed the grease. Do bare in mind though there is a hole high up where the torque sensor wires run through the case to the controller. I think if water is getting throughhere you would needed to have totally submerged the motor ( which one of lads did, missed a bridge ), I have found no real way to stop that as the wires have to allow for adjustment for wear on the faces of the torque sensor and some way of allowing pressure to eqaulise in that inner compartment from temperature changes.

The big one now is to stop water getting into where the big gear and pinion gear is. The X style seal doesn't seem to fully stop water or so I thought. Having watched the Brose video it looks like they have been having exactly the same problem and its not the large X seal but in and around the bearings themselves. It seems that they have started fitting X seals between the peddle and the bearing face. I need to look at that in more detail.

What about tapping a bolt hole underneath, in the gap between the controller and the actual motor compartment. You can plug it with a bolt and at the end of the day unscrew it to let the water out? Only thing with that is the water will stay in there until you undo the bolt. I use the motor for work so use it in the rain and go through lots of puddles in the city, I've not had a problem with my method tbh.I gave this advice to a colleague who had the same problem with the controller, his wasn't reading the battery level proparly

[Waynemarlow]

Mmm I'm calling puddles 100 - 200mm deep and 50m long full of mud and water, such is the long period of wet weather we have had this winter, probably pushing the worst condition on the trails I've ever seen it in England.

I cannot fault the TSDZ2 in this regard, the motor was never ever designed for this sort of abuse.

[manoz]

Aah ok I get you now. Was it easy to fit to that?

[Waynemarlow]

No and yes, took a lot of working out the adaptor to fit the Bafang mounts, once I had a design and made the prototype, if it was copied now onto a CNC mill or it would be a doddle.

Normally though I have the engines in the more normal location

[Waynemarlow]

Here's a photo of the TSDZ2 mounted into the Bafang mounts.

[frenchie]

Here's a photo of the TSDZ2 mounted into the Bafang mounts

wow impressive engineering there
did you originally have a Bafang motor in there? How does the TSDZ2 compare?
Cheers

[Waynemarlow]

No I have never had a Bafang engine in there, when I bought the Dengfu E10 frame the M500 motor I wanted, was not available. What I did have was a TSDZ2 engine so built an adaptor. Would I recommend doing so, sorry it just takes too long from scratch and now it would be far better to buy the Bafang as its availible.

However the Bafang has all the usual problem of no spares and the dreaded EU speed limits, which off road are a real pain.

[Waynemarlow]

Forgot this one

[Waynemarlow]

And one with the size of the Brose belt. Do bare inmind that we would have a much lower gear ratio and possibly we wouldn't need such a wide belt.

[Wapous]

I have received information from a supplier regarding the replacement of the helical gear mechanism by a belt drive. The belt system would not work as they say a ratio exceeding 4: 1 is not recommended. Let's not forget that the system ratio I presented is 6: 1.
As it is impossible to modify the diameter of the driving gear and that of the driven gear, I stop this development.
I include the technical drawing that the suppliers received.


Here the message from the first supplier:

"While you can achieve a great transmission ratios on steel spur gears, you cannot do it with timing belts.
Timing belts approaching ratio 4:1 will have problems:
They will wear fast especially at start and stop, even if you tension the belt to the max.The small pulley may simply jump the teeth and run in one belt spot.
 Some racers use this ratios for competitions- but they change belts every day, or even more often.
This ration will work well if you use a gearbox."

 Thanks for contacting MisumiUSA.

[Waynemarlow]

Well that sorts that idea, pity as a quiet motor would be great.

Thanks for doing all the enquiries Wapous.

[casainho]

As it is impossible to modify the diameter of the driving gear and that of the driven gear, I stop this development.
I include the technical drawing that the suppliers received.

Thanks!!