New "TSDZ2 Torque Sensor Central Motor"

eyebyesickle said:
sysrq said:
"Never use thermal compounds and thermal pads together."
https://blog.arctic.ac/blog/2015/07/23/thermal-pads-or-thermal-paste/
Not sure if always true.

Forgot to take into account typical BLDC motor efficiency (85%), so the requered power handling capacity of the heat pipe would be only around 50-60W for a 350W motor. That would mean that wrapping one 5 mm pipe around the motor and attaching it with thermal epoxy in to the round corner next to the controller should be enough. This way there isn't any need to attach anything to the black low thermal mass external cover.

Another low key solution I can think of would be tying around the motor some 2-5mm thick OFHC copper L bends depending on measurements in between the screws around in conjunction with thermal pads using a jubilee clip. The other ends would press against the round protrusion of the external cover above the motor terminal plate making it possible to put some pressure on the 2 mm thick thermal pads while tightening the cower screws.

So, I don't know for sure, but this is what I am thinking...Usually thermal grease has way high Wm/K thermal transfer rating than the pads...so I think when you use both, the thermal grease is usually touching whatever you are mounting, and then the thermal pad.... I think if the thermal grease has a higher Wm/K rating than the pad, it makes it not work as good because it is pulling the heat from the pad, if that makes sense.... but the way we have it configured...the thermal paste is on the outside of it all, away from what we are trying to cool, so it is continuing to pull heat out, and transferring it to the motor housing cover... either way it is needed to lube up the sides, and it definitely is working well, so that is what I know for sure...

BTW I use 12 Wm/K GP extreme thermal pads... and GP extreme thermal grease too...they both seem to have the highest Wm/K rating for a readily available retail/product.

Just more detail, since im slow with the pics...

qty: 4 - GP extreme 40x80mmx3mm - one gets cut longways in half, and wrapped around middle of motor, then the next two stay whole and wrap around the motor. the ends dont touch - they dont fully encapsulate the motor, which is fine, because one side of the motor does not touch the motor housing side - it faces the controller. this is also where I install a temperature sensor.. the last one gets cut into strips and lays around the circular edge in the motor housing cover.

qty: 2 - GP extreme 40x80x2mm - one lays flat in the middle of the circle in the motor housing cover, the other is chopped into strips, and stacked on the 3mm strips around the edge of the motor housing cover, to make it 5mm.

qty: 1 - gp extreme thermal grease - luuuuuuuuube up the sides (doesn't take much)

total cost - approx $120 + shipping!!! YIKES!!!

...but still makes it a new motor IMO... worth every penny...I am open to something else, but this is super quick and simple... and works very good...

Copper oxide coated copper foam heat sinks could be another thing to look into, if one is interested in low weight and high surface area to air. In this case the most direct solution would be to modify the black cover to enclose the controller only while leaving motor open to air surrounded by copper foam heat sinks. That would probably require some cylindrical dust cover and water proofing though.
https://www.mouser.co.uk/new/versarien-technologies/versarien-lph-heat-sink/
Altough it could be another subtle marketing ploy depending on a specific set of applications so not to be taken seriously.
 
@eyebyesickle
Have you, besides the GP Extreme pads, also tried the cheap thermal pads?
So yes, have you measured the temperature differences too? With and without pads
I ask, because I like to know how much of a difference we could expect with the different thermal conductivity grade of the pads.
Also cheap pads have a better thermal conductivity than air.
 
knutselmaaster said:
samuelsson87 said:
andrea_104kg said:
I do not think it is possible. The torque sensor must feel a resistance, otherwise there is no torque.
On the descent, for example, the engine does not start by turning the pedals ... the wheel does not resist.

Ok, thanks. I've tried to 'hold' the chainring but I guess it's not an optimal setup ;)
If you have a solid bike stand that holds it firmly, you can adjust the brakes so that they give enough resistance to test the motor. Don't tighten too much though and only use short bursts or else you might damage the motor. I use this method to test "if it works", to really test the torque sensor you'll need to ride the bike.

Thanks but I would like to test it before even mounting it on the bike 8)
 
I'm looking to make a heat removal mod. I have seen a a few heat transfer mods utilizing the end of the motor where there is wiring and a large air space. Has anyone considered using thermal paste on the other side of the motor? It seems like there is a nice flat metal to metal interface where thermal paste could be applied to take heat from the motor and bring it to the case. CPU fins could then be attached out side the case in this area.

Thoughts?

QERGq5v.jpg
 
MathewAtYou said:
I'm looking to make a heat removal mod. I have seen a a few heat transfer mods utilizing the end of the motor where there is wiring and a large air space. Has anyone considered using thermal paste on the other side of the motor? It seems like there is a nice flat metal to metal interface where thermal paste could be applied to take heat from the motor and bring it to the case. CPU fins could then be attached out side the case in this area.

Thoughts?

QERGq5v.jpg
Looks like the end cap with terminals forms a fairly small contact surface area with the laminated core where the windings are, so there is a need to transfer the heat directly from the laminated core. Copper is best for heat transfer, while aluminium is better for heat dissipation in the air. The most professional looking solution could be a split two part copper ring machined to match the motor which would be tightened around the motor with screws threaded in to the ring, one peace L profile circular 1-2mm copper strip with 19x30mm indentions for pressure encrease against the laminations and with tightening screws on one side also could be tried enabling to transfer the heat to the outer casing. Thermal pads would still be preferable, although thermal grease also may work to some extent but would require too much precision for parts to align for the best heat transfer, otherwise it would put too much or not enough pressure on the casing screws, so the thermal pads would be still preferable, which might seem too expensive since they are marketed for computer upgrade. Interestingly whole new CPU heat sink with heat pipes will cost a lot less than a small peace of 40x80mm thermal pad.
As for applying thermal grease against the gearbox housing, the other end cap still hasn't got enough contact area with the laminated core for heat to be transferred. High thermal inertia of the gearbox housing theoretically might not be preferable for quick dissipation of heat to air.
 
Good info thank you. So seems like the laminated core area is the preferable place to remove heat from. I like the idea of building up that area and wrapping it with copper bar and having that copper bar go through the outer casing. The brainstorming continues.....
 
samuelsson87 said:
knutselmaaster said:
samuelsson87 said:
andrea_104kg said:
I do not think it is possible. The torque sensor must feel a resistance, otherwise there is no torque.
On the descent, for example, the engine does not start by turning the pedals ... the wheel does not resist.

Ok, thanks. I've tried to 'hold' the chainring but I guess it's not an optimal setup ;)
If you have a solid bike stand that holds it firmly, you can adjust the brakes so that they give enough resistance to test the motor. Don't tighten too much though and only use short bursts or else you might damage the motor. I use this method to test "if it works", to really test the torque sensor you'll need to ride the bike.

Thanks but I would like to test it before even mounting it on the bike 8)
I'm afraid that the only way to accomplish that is by asking someone else to ride it before you 😁
 
MathewAtYou said:
Good info thank you. So seems like the laminated core area is the preferable place to remove heat from. I like the idea of building up that area and wrapping it with copper bar and having that copper bar go through the outer casing. The brainstorming continues.....
Well yeah, without drawings there is no way to tell.
https://www.motioncontroltips.com/understanding-dc-motor-curves-temperature-part-2/
https://www.google.com/amp/s/www.researchgate.net/figure/Overview-of-select-electric-motor-cooling-strategies-a-Surface-air-cooling-with-a-fan_fig1_331417484/amp
 
SO I managed to flash the KT-LCD3 and motor with the opensource firmware. When I press the power button attached to the screen it turns on but just show a bunch of 8's. not sure why. Any help is appreciated. Thanks.
 
Is version 0.20.0 firmware available yet? On github I see only 0.19
 
Retrorockit said:
Elinx said:
Retrorockit said:
Here is a price sheet for the Panasonic TIM pads.
............... Lower # is better. You can see that a thinner pad has less resistance, and lower cost. If used inside a copper ring to tranfer heat from all the way around the motor.......
I understand your arguments for this choice. Copper and a thin pad, but I think it wil not be practical.
First the price of the Panasonic is extreme high in comparision with other thermal pads.
Second I think copper and aluminium is not a good combination because of the electrode potential of metals, one of the metals will be corrode. In this case the aluminium casing. Corrosion is an isolator, so the effect you want to achieve is lesser you want. If you use aluminium for that ring (conductivty of 250 W/m-K) then there will be no problem.
Last point is, you want contact with the casing as much as possible. Then you need some elasticity from the thermal pads, wich is less with thinner pads.

Imho, a better choice would be a half ring of aluminium, what has a snug fit inside the front of the casing, sticked to the laminated metal of the motor with elastic thermal pad and also stick inside the flat side of the cover casing an elastic thermal pad ring that rest a the top of the motor after fastening the cover.

Why a half ring. This because there must be some pressure to it if you mount the cover casing.
I think this is almost impossible with a sticked full ring.
That's the beauty of DIY. You can do whatever you want. The cost of the Panasonic pads is high, but the performance is very high also. I always look at the best material available, and work down to what is practical for me. The Cu,Al erosion issue would be true if there was an electrolyte such as water present ( as in CPU water cooling loop, or automotive cooling systems). Just about every computer heatpipe cooler has Cu pipes pressed into Al fins with no issues. But thermal grease will isolate the 2 metals both physically and electrically. The grease will also allow movement for expansion as will the themal pads. Cheaper pads are always an option and should perform just fine. The ring could have a gap in one place to allow for linear expansion that way also. This may be needed to adjust the diameter anyway. But you are looking at all the right issues to be considered. The choice between Cu and Al may come down to what dia. and wall thickness tubing is available to make the ring. Either one willl transfer heat much better than a single thick thermal pad. My thinking on the expansion of the Cu ring is that it will allow easy assembly when cold, but will expand create pressure to improve heat transfer when at working temperature.

Theoretically the best solution for average DIY user would be 30mm wide, 1-4mm thick strips of adhesive TIM pads with high Z axis thermal conductivity (1700W/m.k Panasonic has only 20W/m.k) with one end attached to laminations around the motor and the other end at the motor terminal end cap which would press against the external cover.

Flexible thermal straps with one end curved to 65mm radius could be another option for somewhat easier installation around the edge of the motor compared to bending 30mm wide solid copper plate around the circular edge.
https://thermal-space.com/wp-content/uploads/more-straps.png
https://thermal-space.com/thermal
DIY option using copper braid soldered to copper plates also exists. Not sure about thermal conductivity compared to solid copper though.

67mm copper pipe end feed stop ends could look a bit neater if successfully fitted by making the gaps for the motor screws around the laminations and probably a hole around the terminals, although it should be at least 69-71mm in diameter for a proper fit.
https://www.amazon.com/Libra-Supply-options-Pressure-Plumbing/dp/B01MT78Y5M
 
Hi,
i'm planning to add an ESP32 board between the motor controller and the display.
The board will act as "man in the middle" between the two UARTs and will be able to add blutooth and WiFi capability to the Controller.
I'm still really undecided about how to get power supply for the board. One option is to get the 5V from the controller board using the +5V for the the throtthle. Anyone knows how many mA is able to provide the 5V controller voltage regulator? The ESP32 will use up to 250/300mA when in trasmit mode.
Another option is to use the +Vbat coming from the display connected to a Step Down circuit, but this wil add another small board to inser into the motor case.
 
mspider65 said:
Hi,
i'm planning to add an ESP32 board between the motor controller and the display.
The board will act as "man in the middle" between the two UARTs and will be able to add blutooth and WiFi capability to the Controller.
I'm still really undecided about how to get power supply for the board. One option is to get the 5V from the controller board using the +5V for the the throtthle. Anyone knows how many mA is able to provide the 5V controller voltage regulator? The ESP32 will use up to 250/300mA when in trasmit mode.
Another option is to use the +Vbat coming from the display connected to a Step Down circuit, but this wil add another small board to inser into the motor case.
Are you aware of the SW102 display with Bluetooth? SW102 is very flexible and for instance we have now integration with Strava mobile app.
 
In my TSDZ2 I need more speed on flat sections so that I can catch up with my colleagues with Bafang drives :) On the back I have 11-42T. 110BCD DOUBLE CHAINRINGS seems to be the solution. I don't want to be slanting the chain and doing a problem with the front derailleur that won't reach that far. Will I not have a problem with adding to the original 42T sprocket another larger one for 5 arms and 110BCD preferably 50T?
For example like this:
https://allegro.pl/oferta/koronka-tarcza-stronglight-5083-50t-110mm-7679156601
Zrzut ekranu 2019-09-06 o 12.02.45.png
Zrzut ekranu 2019-09-06 o 12.05.22.png
This largest front sprocket can only be used with the 11T at the rear, for fast driving on asphalt. If this solution works, what else will I need besides the 50T sprocket shown?
 
Could the motor and various contoller devices be flashed through an OTG cable?

Specifically my little Windows 10/Android 5.1 dual boot tablet has only a micro usb port so I would have to use an OTG cable to connect the tablet to the standard sized usb cable end.

Thanks, for all of your work!

Longterm lurker
 
sysrq said:
Retrorockit said:
Elinx said:
Retrorockit said:
Here is a price sheet for the Panasonic TIM pads.
............... Lower # is better. You can see that a thinner pad has less resistance, and lower cost. If used inside a copper ring to tranfer heat from all the way around the motor.......
I understand your arguments for this choice. Copper and a thin pad, but I think it wil not be practical.
First the price of the Panasonic is extreme high in comparision with other thermal pads.
Second I think copper and aluminium is not a good combination because of the electrode potential of metals, one of the metals will be corrode. In this case the aluminium casing. Corrosion is an isolator, so the effect you want to achieve is lesser you want. If you use aluminium for that ring (conductivty of 250 W/m-K) then there will be no problem.
Last point is, you want contact with the casing as much as possible. Then you need some elasticity from the thermal pads, wich is less with thinner pads.

Imho, a better choice would be a half ring of aluminium, what has a snug fit inside the front of the casing, sticked to the laminated metal of the motor with elastic thermal pad and also stick inside the flat side of the cover casing an elastic thermal pad ring that rest a the top of the motor after fastening the cover.

Why a half ring. This because there must be some pressure to it if you mount the cover casing.
I think this is almost impossible with a sticked full ring.
That's the beauty of DIY. You can do whatever you want. The cost of the Panasonic pads is high, but the performance is very high also. I always look at the best material available, and work down to what is practical for me. The Cu,Al erosion issue would be true if there was an electrolyte such as water present ( as in CPU water cooling loop, or automotive cooling systems). Just about every computer heatpipe cooler has Cu pipes pressed into Al fins with no issues. But thermal grease will isolate the 2 metals both physically and electrically. The grease will also allow movement for expansion as will the themal pads. Cheaper pads are always an option and should perform just fine. The ring could have a gap in one place to allow for linear expansion that way also. This may be needed to adjust the diameter anyway. But you are looking at all the right issues to be considered. The choice between Cu and Al may come down to what dia. and wall thickness tubing is available to make the ring. Either one willl transfer heat much better than a single thick thermal pad. My thinking on the expansion of the Cu ring is that it will allow easy assembly when cold, but will expand create pressure to improve heat transfer when at working temperature.

Theoretically the best solution for average DIY user would be 30mm wide, 1-4mm thick strips of adhesive TIM pads with high Z axis thermal conductivity (1700W/m.k Panasonic has only 20W/m.k) with one end attached to laminations around the motor and the other end at the motor terminal end cap which would press against the external cover.

Flexible thermal straps with one end curved to 65mm radius could be another option for somewhat easier installation around the edge of the motor compared to bending 30mm wide solid copper plate around the circular edge.
https://thermal-space.com/wp-content/uploads/more-straps.png
https://thermal-space.com/thermal
DIY option using copper braid soldered to copper plates also exists. Not sure about thermal conductivity compared to solid copper though.

67mm copper pipe end feed stop ends could look a bit neater if successfully fitted by making the gaps for the motor screws around the laminations and probably a hole around the terminals, although it should be at least 69-71mm in diameter for a proper fit.
https://www.amazon.com/Libra-Supply-options-Pressure-Plumbing/dp/B01MT78Y5M
According to this datasheet Panasonic PGS sheets can be used for remote heat transfer.
 

Attachments

  • 1952012.pdf
    1.9 MB · Views: 81
New and improved :D
I changed out my modified inner chainring with the stock steel offset one, with an original spider and my trimmed 53 tooth chainring. I took off about 1.5mm, (see photo) and had to add 1 thin washer between the spider and drive hub. This change eliminated 2 offset washers, and drastically improved the chain line. I could reduce the derailleur offset my 5mm or more.
 

Attachments

  • 19-09-06_11.47.19AM-747x1328.jpg
    19-09-06_11.47.19AM-747x1328.jpg
    142.3 KB · Views: 1,719
  • 19-09-06_11.41.19AM-747x1328.jpg
    19-09-06_11.41.19AM-747x1328.jpg
    146.6 KB · Views: 1,719
  • 19-09-06_11.41.43AM-747x1328.jpg
    19-09-06_11.41.43AM-747x1328.jpg
    99.8 KB · Views: 1,719
  • 19-09-06_11.47.38AM-747x1328.jpg
    19-09-06_11.47.38AM-747x1328.jpg
    117.6 KB · Views: 1,719
RicMcK said:
New and improved :D
I changed out my modified inner chainring with the stock steel offset one, with an original spider and my trimmed 53 tooth chainring. I took off about 1.5mm, (see photo) and had to add 1 thin washer between the spider and drive hub. This change eliminated 2 offset washers, and drastically improved the chain line. I could reduce the derailleur offset my 5mm or more.

great, could I ask you to show more pictures especially on the side of the bike, so that I could understand how exactly you screwed it together? Did you have to modify the derailleur? How does it work for you?
 
I just fitted a new controller as I thought my non responding motor was a controller issue. I tried with open source on lcd3 and stock firmware on standard monitor still no joy. How can I raw test my bare motor to see if it’s dead?
6bb04242a17c9d03e0dd9ad37c2afacc.jpg



Sent from my iPhone using Tapatalk
 
adam B said:
I just fitted a new controller as I thought my non responding motor was a controller issue. I tried with open source on lcd3 and stock firmware on standard monitor still no joy. How can I raw test my bare motor to see if it’s dead?
6bb04242a17c9d03e0dd9ad37c2afacc.jpg



Sent from my iPhone using Tapatalk

I use one of these, with the ends switched....

You can also use a multimeter to test...
 

Attachments

  • mt3-1600-x-1200.jpg
    mt3-1600-x-1200.jpg
    177.4 KB · Views: 1,679
arka said:
RicMcK said:
New and improved :D
I changed out my modified inner chainring with the stock steel offset one, with an original spider and my trimmed 53 tooth chainring. I took off about 1.5mm, (see photo) and had to add 1 thin washer between the spider and drive hub. This change eliminated 2 offset washers, and drastically improved the chain line. I could reduce the derailleur offset my 5mm or more.

great, could I ask you to show more pictures especially on the side of the bike, so that I could understand how exactly you screwed it together? Did you have to modify the derailleur? How does it work for you?

Stock duel chainring derailleur (I am using a triple shift lever). I was able to eliminate the messy derailleur offset adapter {shown in photos), with much better chain line. I am not near my bike for a few days, but I now think that a steel large chainring might fit instead of the modified AL one without any problem. I did need to put a thin SS washer between the spider and drive hub to prevent the chainring nuts from hitting the plastic cover. I sealed the spider hub interface with silicone.

I am also messing with motor cooling. Once I get a good configuration I will post it.
 
eyebyesickle said:
adam B said:
I just fitted a new controller as I thought my non responding motor was a controller issue. I tried with open source on lcd3 and stock firmware on standard monitor still no joy. How can I raw test my bare motor to see if it’s dead?
6bb04242a17c9d03e0dd9ad37c2afacc.jpg



Sent from my iPhone using Tapatalk

I use one of these, with the ends switched....

You can also use a multimeter to test...
dc03f9a1e40f91af998a230181f8f161.jpg


Progress
I got the Walk assist and Throttle working with 0.19 + KT LCD3

Speed sensor not working

PAS only works for half a second and stops.

More work to do.

Will solder all my connections from higo plug stolen from stock display onto KT LCD 3 and Throttle.



Sent from my iPhone using Tapatalk
 
Hi! I bought a tsdz2 engine for a second bike. mounted correctly.I have a problem. After driving 60 km in a forest area with small hills, the engine began to emit noise at a higher load. on the film I do a gentle brake load to make that noise.
https://youtu.be/4wW5HgZSFxg
 
andrea_104kg said:
RicMcK said:
How about putting fine grit steel wool (000) pads between the motor and case? They are compressible, high conductivity.

Does the motor and case needs to be electrically isolated?

also exists fine copper wool, very good conductivity, i think that it work! it's suffit search on ebay...
There is also braided Copper grounding strap available.
https://www.amazon.com/Small-Parts-Flat-Copper-Braid/dp/B003R501JK
 
Back
Top