eyebyesickle
10 kW
Nice. I have not tried heat sinks yet but have been experimenting and am about settled with thermal pads around the motor like that!
New "TSDZ2 Torque Sensor Central Motor"
- Thread starter larsottar
- Start date
ezrider1199 said:
Sounds strange. Never seen this. When off there is a small floating voltage, nothing that would power a light. When on just over 6 volts. On your bench power supply what is the current and voltage you are using to drive the lights? This circuit only supports a few watts.
I have never tripped the light circuit, but from previous comments it appears to cut out and then reset when the load is removed.
ezrider1199
10 W
- Joined
- Jul 2, 2019
- Messages
- 77
It's definitely strange.. initially, in series it was powering the first light but the second light was dim or dimly flickering. In retrospect, I'm thinking that the first light was taking the whole voltage (that's how series works [?]) so that behavior makes sense.
Then i tried it in parallel to get them to light up equally, and that's when i witnessed the weirdness i described in my previous post. I really hope i didnt mess up that part of the motor controller.
The reason i thought it was safe to plug it in was because my bench power supply was running at 6v and 350mah (it's just a simple power brick with adjustable voltage - up to 12v). I thought it was low power enough to attempt and plug it into the TSDZ2. These 2 lights i was using are from 185 lumen AA lights, so not that powerful.
I also have a janky old multimeter from radioshack and just ordered a new one in order to troubleshoot the problem better because im losing trust in the old one - it's hard to make contact and get readings, etc.
Then i tried it in parallel to get them to light up equally, and that's when i witnessed the weirdness i described in my previous post. I really hope i didnt mess up that part of the motor controller.
The reason i thought it was safe to plug it in was because my bench power supply was running at 6v and 350mah (it's just a simple power brick with adjustable voltage - up to 12v). I thought it was low power enough to attempt and plug it into the TSDZ2. These 2 lights i was using are from 185 lumen AA lights, so not that powerful.
I also have a janky old multimeter from radioshack and just ordered a new one in order to troubleshoot the problem better because im losing trust in the old one - it's hard to make contact and get readings, etc.
mctubster said:
ezrider1199
10 W
- Joined
- Jul 2, 2019
- Messages
- 77
Awesome 
That sounds like an effective mod. Will have to do it on mine
That sounds like an effective mod. Will have to do it on mine QuirkyOrk said:
casainho
10 GW
- Joined
- Feb 14, 2011
- Messages
- 6,058
I am just sharing here because this is a very important break through:
quote=Rydon post_id=1487352 time=1565328969 user_id=50225
Indeed it is! We just installed v0.20.0 alpha 3 on a friends coaster brake handcycle and it works!!! This is truly a turning point for powered handcycles. After celebratory yells and high fives, I took this picture of Jeff with the handcycle we installed it on. It's his bike.
.
.
We only tested it with 5 levels of power assist with the multiplier being simply 1 to 5. EMTB mode was set to 5. We will tweak it later but the initial test ride was awesome. We have yet to set up the advanced cadence sensor mode so there was a very slight delay for the motor to stop that was only noticeable in the higher levels. It was overall very smooth and responsive.
Thanks to Buba for a truly awesome release and to Casainho for providing the foundation. You guys are helping more people than you realize that will be able to once again do what they love. Also, a special thanks to Buba for supporting the coaster brake. Jeff can only ride a coaster brake since he is a quadriplegic with only partial use of his harms and no use of his hands. He pedals, steers and brakes with his arms all at the same time. He is super excited to ride this in an 80 mile mountain race in Park City, Utah next week.
/quote
quote=Rydon post_id=1487352 time=1565328969 user_id=50225
buba said:
Indeed it is! We just installed v0.20.0 alpha 3 on a friends coaster brake handcycle and it works!!! This is truly a turning point for powered handcycles. After celebratory yells and high fives, I took this picture of Jeff with the handcycle we installed it on. It's his bike.
.
.
We only tested it with 5 levels of power assist with the multiplier being simply 1 to 5. EMTB mode was set to 5. We will tweak it later but the initial test ride was awesome. We have yet to set up the advanced cadence sensor mode so there was a very slight delay for the motor to stop that was only noticeable in the higher levels. It was overall very smooth and responsive.
Thanks to Buba for a truly awesome release and to Casainho for providing the foundation. You guys are helping more people than you realize that will be able to once again do what they love. Also, a special thanks to Buba for supporting the coaster brake. Jeff can only ride a coaster brake since he is a quadriplegic with only partial use of his harms and no use of his hands. He pedals, steers and brakes with his arms all at the same time. He is super excited to ride this in an 80 mile mountain race in Park City, Utah next week.
/quote
QuirkyOrk said:
Nice to see these solutions seems to have a positive result.eyebyesickle said:
I wonder if it is sufficient enough to stick this thermal silicone pad only into the casing on the flat side, so after tightening the screws of the casing, the metal of the motor-topside makes thermal contact with the casing. (no thermal pads around the motor)
In that case it is easier to service the motor afterwards.
Retrorockit
1 kW
I'm in South FL so I need to do something with my TSDZ2. It's nice to know this simple mod works. I was planning something more elaborate, but if this gets the job done then why not. Thanks for posting your results.QuirkyOrk said:
Elinx said:
Thanks for the input!
I don't think servicing the motor will be too much of an issue. The silicone pads are slightly sticky/tacky and can be taken on and off pretty easily. That stickiness and the case tightened up against them are what hold them in place.
(They are also pretty cheap at $1-3 USD on ebay and could be replaced if they got messed up while you were taking them off. I just ordered a bunch more so I can experiment with different thickness in different places)
Retrorockit said:
Thanks for the kind words. Yeah I tried to keep it as simple as possible. I wanted to keep the sealed casing and weatherproofing of the motor intact because using the bike rain or shine is very important to me. I also realized that passive cooling would totally get the job done in addition to being nice and simple.
I Think i have broken my controller or motor after greasing the bleu gear…. 
After greasing i build it all together and started up the TSDZ2.
Everything looked oke, but when i peddal there was no support anymore en get E004 error on display...
Very strange numbers in display like 97 km/ hour till zero km / hour.
I checked the speedsensor and everything looks oke.
I opened the motor and the only thing i noticed was that i forgot to put the motor cables down in the clip.
Some cables looks a kind of flat. They are not broken, i checked them and are oke….
I put new software on the controller and didn`t gave any results.
Now i don`t know how to search for the problem. I think myself there is a problem with the controllor or motor.
Is there a way how i can check the motor or controller to check if they are oke?
Is there anyone who will help me to try to find the problem on my 36V 350W motor ? :wink:
After greasing i build it all together and started up the TSDZ2.
Everything looked oke, but when i peddal there was no support anymore en get E004 error on display...
Very strange numbers in display like 97 km/ hour till zero km / hour.
I checked the speedsensor and everything looks oke.
I opened the motor and the only thing i noticed was that i forgot to put the motor cables down in the clip.
Some cables looks a kind of flat. They are not broken, i checked them and are oke….
I put new software on the controller and didn`t gave any results.
Now i don`t know how to search for the problem. I think myself there is a problem with the controllor or motor.
Is there a way how i can check the motor or controller to check if they are oke?
Is there anyone who will help me to try to find the problem on my 36V 350W motor ? :wink:
Attachments
Retrorockit
1 kW
The tight fit between the pads and case is important. The more pressure the better the heat transfer. It also helps that they're malleable and can conform to the uneven surface of the motor.QuirkyOrk said:
I was wanting to do something in coppper or aluminum for bettter heat transfer. But if what you did solves the problem that's all that really matters.
ezrider1199
10 W
- Joined
- Jul 2, 2019
- Messages
- 77
QuirkyOrk,
What thermal pad thickness do you recommend? i saw they start at 0.5mm and go up by 0.5mm increments. Thanks
What thermal pad thickness do you recommend? i saw they start at 0.5mm and go up by 0.5mm increments. Thanks
casainho
10 GW
- Joined
- Feb 14, 2011
- Messages
- 6,058
If you use our OpenSource firmware, we can help as it provides technical real time data of motor and sensors.cejean said:I Think i have broken my controller or motor after greasing the bleu gear….
After greasing i build it all together and started up the TSDZ2.
Everything looked oke, but when i peddal there was no support anymore en get E004 error on display...
Very strange numbers in display like 97 km/ hour till zero km / hour.
I checked the speedsensor and everything looks oke.
I opened the motor and the only thing i noticed was that i forgot to put the motor cables down in the clip.
Some cables looks a kind of flat. They are not broken, i checked them and are oke….
I put new software on the controller and didn`t gave any results.
Now i don`t know how to search for the problem. I think myself there is a problem with the controllor or motor.
Is there a way how i can check the motor or controller to check if they are oke?
Is there anyone who will help me to try to find the problem on my 36V 350W motor ? :wink:
ezrider1199 said:
I used these 2mm pads (The only reason I chose these was it shipped from the US and I am inpatient
)https://www.ebay.com/itm/Grey-GPU-CPU-Heatsink-Thermal-Conductive-Silicone-Pad-100mm-x-100mm-x-2mm-US/254000030793?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649
The 2mm seems to be a good thickness so that one layer of it will just contact the outer shell and still allow the casing to be slotted over. I tried 2 layers of 2mm sheets stacked and I could not get the shell on. On the top of the motor the patches are 2 layers 2mm pads stacked on top of each other. I ordered a couple 3mm pads today and I am going to experiment with those as well.
I actually currently only have about 1/3 of the circumference of the motor, where the gap is the thinnest, covered in thermal padding. There is definitely room to make the contact patch bigger and increase performance even more.
Retrorockit
1 kW
I got my Trek Navigator 400 going again. The Nexus 7 hub simply wasn't up to 750W power. Skipping in 7th gear. Probably a roller clutch not holding the power. So it now has a 6001 series Nexus 8 speed. I reused the older IM 70 roller brake because the cable setup was different for the newer IM 80. The Shimano Nexus 8 speed shifters were all too big and wouldn't allow room for my horn button so I used a Microshift brand Nexus shifter. It's now a legitimate 25mph bike, and 1st gear is low enough for riding around pedestrians. 44tX19T cogs and 26X2" tires. No complaints from the 8 speed at this power level.
https://electricbike.com/forum/forum/main-forum/diy-discussion/82828-trek-navigator-400-igh-tsdz2-project/page4
https://electricbike.com/forum/forum/main-forum/diy-discussion/82828-trek-navigator-400-igh-tsdz2-project/page4
Retrorockit
1 kW
On the topic of thermal pads.
The best themal transfer material is diamond, then graphite. Both of which are way beyond metals.
Diamond is obviously not practical, and grahite forms in flakes or nano tubes that that conduct heat only lengthwise.
Panasonic seems to have made a thermal pad with graphite fibers aligned vertically. It's soft, flexible and comes in the thickeness we need. It also tranfers heat lengthwise so could be wrapped all the way around the motor.
https://media.digikey.com/pdf/Data%20Sheets/Panasonic%20Semiconductors%20ICs%20PDFs/EYG-T-Series_Graphite-PAD_NPI.pdf
It comes with adhesive on both sides, so that would be an issue. But it looks really good otherwise.
It might be a little expensive for experimentation, but if we get the sizes nailed down where you can get the right thickness the first time it should be good.
The best themal transfer material is diamond, then graphite. Both of which are way beyond metals.
Diamond is obviously not practical, and grahite forms in flakes or nano tubes that that conduct heat only lengthwise.
Panasonic seems to have made a thermal pad with graphite fibers aligned vertically. It's soft, flexible and comes in the thickeness we need. It also tranfers heat lengthwise so could be wrapped all the way around the motor.
https://media.digikey.com/pdf/Data%20Sheets/Panasonic%20Semiconductors%20ICs%20PDFs/EYG-T-Series_Graphite-PAD_NPI.pdf
It comes with adhesive on both sides, so that would be an issue. But it looks really good otherwise.
It might be a little expensive for experimentation, but if we get the sizes nailed down where you can get the right thickness the first time it should be good.
In the link you gave I saw that these pads have a conductivity of 13 W/m-K, but for a priceRetrorockit said:
The cheap silicon pads about 2 W/m-K, that is a lot of a difference, but always better than air with 0,03 W/m-K
3M pads have a conductivity of about 5 W/m-K
Searching the web I found this and that material with a (given) conductivity of 12 and 12,8 W/m-K, but I don't know if it is useable for our function.
Elinx said:
It's so cool that you guys are looking into better performing thermal pads. I am curious how much of a difference they make in cooling performance compared to the cheap $1-3 pads. I like the cheap easily replaceable nature of the ones I am using now, but if there were a solution that dropped temps even more for less than $20 or so it might be worth it to me.
Also if you were interested in keeping the motor as stealthy/inconspicuous as possible you might be able to cool it with just high performance thermal interface pads and not have the aluminum thins on the outside.
Retrorockit
1 kW
They're expensive in the pre packed 35x35mm individual size for electronics, but possibly much cheaper in the bulk 140x140mm sheets we would want. I haven't priced it yet. I will probably be buying this for my computer overclocking projects anyway. In the US 750W is a legal E bike rating. So 750W continuos is a legitimate target for us. I have an inquiry in to Panasonic tech support about this material, but they're on vacation until 8/20. They may not bother to reply to an individual anyway. I had been looking at copper foam, carbon foam, and solid metals like Copper. But the success of the DIY TIM pads got me looking onto this option. I may still look into doing a Copper solution. I haven't taken the cover off of my motor yet to measure things. I've been busy with a 7 to 8 speed IGH upgrade.Elinx said:
Retrorockit
1 kW
Retrorockit said:
Retrorockit
1 kW
Here is a price sheet for the Panasonic TIM pads.
https://www.digikey.com/products/en?FV=ffece3c0
Basically the thinner the pad the less expensive. Thermal conductivity is useful for comparing one material to another, but thermal resistivity takes into account the thickness of each part. 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, and thermal grease to join to the inside of the waterproof case. Maybe an affordable solution will be found. This will be true of all pads, and the 3M part could turn out to be the most cost effective.
Themal conductivity of Copper is 400. So thin TIM is better than one big thick pad. But a $3 solution is hard to beat!
https://www.digikey.com/products/en?FV=ffece3c0
Basically the thinner the pad the less expensive. Thermal conductivity is useful for comparing one material to another, but thermal resistivity takes into account the thickness of each part. 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, and thermal grease to join to the inside of the waterproof case. Maybe an affordable solution will be found. This will be true of all pads, and the 3M part could turn out to be the most cost effective.
Themal conductivity of Copper is 400. So thin TIM is better than one big thick pad. But a $3 solution is hard to beat!
nukezero
10 kW
I have a Felt-brand Cruiser coaster bicycle and it uses a ~74mm bottom bracket that is 35mm in diameter. The diameter seems a bit wide than usual. WIll the TSDZ2 fit properly and not have any "shimmy" inside the BB? As far as I can tell, the existing bottom bracket is not a sealed cartridge unit. The whole bearing will fall out of it if I take it apart.
Similar threads
