mwkeefer
1 MW
Maybe wider 
DaVinci drive.....
- Thread starter recumpence
- Start date
recumpence
1 GW
Here are some numbers for you;
I mentioned, earlier, that the 3220 runs warmer at the rear of the can, where is protrudes from the motor cradle, than it does at the front (the smaller motors do not because their windings are fully covered by the motor cradle). Well, I installed two of my heatsinks at the rear of my 3220 and ran the daylights out of the trike. Here are the results;
After repeated drag launches and full throttle blasts, same as I always run for product testing, I tested the motor temp. I ran the trike so hard THE CLUTCH reached 170 degrees and this was with very little clutch slip, but the motor was MUCH cooler. I actually had wheel-spin, I was running the clutch so tight. Anyway, the clutch reached 170 degrees, but the hottest the motor ever got was 121 degrees F. That is the back of the motor can at the hottest point, not the heatsink (which would be cooler). I was always able to hold my hand on the motor. No matter how hard I ran it, I could not get more heat than that. Now, consistant running up a long hill may give more heat. But, this is 50 degrees cooler than I saw previously. The entire drive was quite warm. So, it is sinking heat all the way into the beefy part of the drive unit. Also, the whole unit cools relatively quickly. All this was without a fan, just the new drive with heat sinks. A 3210 runs similarly without any heatsinks because all windings are within the motor cradle, therefore passing all heat into the drive unit itself.
Matt
I mentioned, earlier, that the 3220 runs warmer at the rear of the can, where is protrudes from the motor cradle, than it does at the front (the smaller motors do not because their windings are fully covered by the motor cradle). Well, I installed two of my heatsinks at the rear of my 3220 and ran the daylights out of the trike. Here are the results;
After repeated drag launches and full throttle blasts, same as I always run for product testing, I tested the motor temp. I ran the trike so hard THE CLUTCH reached 170 degrees and this was with very little clutch slip, but the motor was MUCH cooler. I actually had wheel-spin, I was running the clutch so tight. Anyway, the clutch reached 170 degrees, but the hottest the motor ever got was 121 degrees F. That is the back of the motor can at the hottest point, not the heatsink (which would be cooler). I was always able to hold my hand on the motor. No matter how hard I ran it, I could not get more heat than that. Now, consistant running up a long hill may give more heat. But, this is 50 degrees cooler than I saw previously. The entire drive was quite warm. So, it is sinking heat all the way into the beefy part of the drive unit. Also, the whole unit cools relatively quickly. All this was without a fan, just the new drive with heat sinks. A 3210 runs similarly without any heatsinks because all windings are within the motor cradle, therefore passing all heat into the drive unit itself.
Matt
wojtek
100 kW
it is only getting better and better! Matt - you are an Artist!
Im left with no choice- i am buying one!
Im left with no choice- i am buying one!
Green Machine
100 kW
This assembly came out very nice Matt...beautiful. This could solve the heat problem i have been having hill climbing.
MitchJi
10 MW
Hi Matt,
Stating the obvious, beautiful work! Excellent features, primary 5:1 belt drive for 3220's, easier belt adjustment, cooler running...
Stating the obvious, beautiful work! Excellent features, primary 5:1 belt drive for 3220's, easier belt adjustment, cooler running...
How difficult would more primary reduction be to implement, like 6:1 to 7:1?recumpence said:
recumpence
1 GW
The only way to get more reduction is a larger pulley. That is possible, but expensive because the 80 tooth pulley is just barely under 5 inches diameter and can use 5inch round stock. A bigger pulley requires buying a larger round bar. It is doable, though.
Matt
Matt
mdd0127
100 kW
- Joined
- Sep 30, 2008
- Messages
- 1,084
Pfeifer has kart racing pulleys that custom centers could be made for. Also, I had good success with a company called hm manufacturing. They will sell pulley stock with the teeth hobbed at reasonable prices. You just cut it into slices and machine out the centers.
The new drive is beautiful BTW.
The new drive is beautiful BTW.
recumpence
1 GW
Precision machining is another source.
I have looked a bit.
I have looked a bit.
rkosiorek
100 kW
Matt,
are the parts for your limited slip available separatetly?
rick
are the parts for your limited slip available separatetly?
rick
recumpence
1 GW
Yes, actually, I am having a large batch of custom aluminum limiters made very soon as well. That will be a stock item on the DaVinci to keep weight down.
Matt
Matt
mwkeefer
1 MW
Really Matt - 120F?
You know I think torque limiting should be electrical (and can be with sensors) to save weight and Q factor but if you telling me that your motor (3220 4T I presume) maxxed out at 120F (you did wait 25/50 seconds before taking can reading right) then Delrin becomes a viable sensor ring material which would work wonderfully and could reduce production costs of Hall Timing Systems for Astro and other Inrunners as needed down into the sub 5.00 (minus labor) point - According to the manufacturer rep, delrin will support 160F and I can do just as I do now and embed for an addtional 2.00 cost, the NTC heat sensor calibrated to trigger an isolated interrupt if the temp exceeds a preset value (resistor sets the value).
What is your current gearing on this DaVinci drive, I know your looking at 4:1 stage #1 (currently) and I presume 100t 5mm is within the realm of possible for some of your builds, with the new belt tensioning system - you could perhaps even reduce the primary drive sprocket from 20T down to 17t or even smaller due to additional belt wrap and proper tension. That could in theory put you at 5:1 or even as high as 5.8824:1 for your first stage.
I know you gear out for 40ish mph most of the time (especially on the 3220s) so I would assume the 12S gearing is likely about (presuming 26" rear wheel) 13.5 or 14:1... assuming the highest stage #1 reduction is 4:1 that would mean a final drive ratio of 3.5 : 1.
If you were to jump up to 16S and sensors (done properly) - the temp should stay lower (by producing more torque sooner and for a longer pull to provide faster time to efficient zone, also by running in the optimal power / efficiency range of the motor) so the new reduction you would want (presuming you had 14:1 before for 40mph) 19.3545 : 1 total which means without changing the pulleys you would need to acheive 4.8386 : 1 in stage #2 - could use #25 chain to do this, 80t rear sprocket with 18t drive #25 sprocket - or you could have a custom rear 130mm BCD chainring made, using 13t odessy freewheels you would need a 63t rear chainring to get close but wouldn't that put you into a much more powerful and efficient motor speed?
Sorry to go OT - had some successful results with the aluminum sensor rings lately and considering producing them on a more mass scale - don't worry Matt a free 9FET infineon controller modded for 60v nominal operation to be run with 16S packs (it will work from 12S to 20S) at 100A or 6KW continuous, phase current is set to 2x or 200A but reducing this if there is belt slippage will reduce startup torque - block time would be one other item to play with - we can discuss this at more length once you receive your rings and your controller... I will include two wired and ready to go rings in aluminum and 1/2" thick, you already know where they go and between the hint you mentioned Bob gave you about a year ago with regards to ABC locations, it seems wiring them 90% of the time will be step by step easy (call it 10 minutes + perhaps drying time, depending on how you mount).... I will add two small dimple screws (pointed set screws) you can use to gently lock the timing rings in place - normal 5 conductor wire can be used as extension if needed but I will also include 2 10cm 5S JST-HXT Male/Female Extension Cables so you can wire up plug and play.... Even if one of your motors turns out to be difficult to calibrate, it's about 15 minutes total with the watt meter in place and the bike suspended or rear/drive tire lifted to get lowest no load currents.
Back OT: Three Questions:
1.) So when are you going to be ready to ship these / my NuVinci drive??
2.) When do you expect to be able to offer 1 - 1.5" mounts again??
3.) How many AH/WH did you pull durring your testing, what are your stats (what happened to the days we posted our Eagle Tree or HV ICE logs?)
Matt - PM or call me about the drive pulleys we spoke of, there may be an issue (or may not be) depends how fast you need em?
I know to many 585 or more is alot for a "drive mount" as they see it, but let me aliken it to my Generator (somthing I just discovered in the Hurricane recently past) - the unit is 199.99 new (great deal right), for 1000w (really 800w constant, just enough for the fridge) and I only paid 50.00 for it but the freaking 2 stroke **(yep I made a foobar and bought 2 stroke gene, it's going don't worry) is rated as 2HP... so it take 2HP of ICE (internal combustion engine) to provide just a touch above 1HP of continuous power, 1.25 hp max for a few seconds? Thats worse than 50% efficiency.
Add to that a full Astro 3215 or 20 will set you back 560.00 or 695.00 and even if you add hall sensors plus the few additional bits of chain and sprocket (sans power / battery) your looking at worst about 1500.00 with an HV160 setup about 1800.00 still without power so is it for everyone, no... but run properly it's 90-93% efficient. A good 3KW gas generator is about 85% efficient, a decent hub motor (run at the proper design voltage / current for the wind and copper weight) is at best 80%... in price comparison, a good 3kw generator will run you several thousand dollars (1-3K) and a hub motor kit of a higher end (figure capable of 3KW) would run you in the neighborhood of 450.00 without power for a 26" rear but you would also need to fully rewire the system, replace the controller with a 200-250.00 aftermarket setup, shorten the leads, replace the phase and power wires and that all adds up to about 800 - 1000.00 if done right, including the torque plates you would have made to street that power without breaking your frame - oh yea, at an additional weight offset of approx 16 lbs, again without power.
Now does the 1500-1800 for the mechanicals seem so expensive now? Not really, especially if it's setup properly - done right the Astro Solution (even a lowly 428.00 3210) will provide a more efficient, more able ride with ample power in reserves and the ability to climb hills like a billy goat (if geared low enough or shiftable)?
IMHO (as Matt can tell you) for the time he puts into it, the wonderful machine work and polished or annodized finish, the constant re-evaluation of the status quoue to improve it either in thermal, power output, ergonomics, strength, or any other enhancements... That's why I ordered one and once the 1.5" mounts are available again (because they are cleaner than anything I could make for tube mounts) I will be ordering at least 2-3 more for customer builds
Oh yea - for the record, though I speak with Matt often... I don't work with him, for him, receive a discount for anthing I say or don't say - heck I have to pay for my review units with him (limited production runs, it's fully understandable) and not at dealer cost but at retail... so there is no incentive but love (in that respect the man kinda way) for my words.
You know I think torque limiting should be electrical (and can be with sensors) to save weight and Q factor but if you telling me that your motor (3220 4T I presume) maxxed out at 120F (you did wait 25/50 seconds before taking can reading right) then Delrin becomes a viable sensor ring material which would work wonderfully and could reduce production costs of Hall Timing Systems for Astro and other Inrunners as needed down into the sub 5.00 (minus labor) point - According to the manufacturer rep, delrin will support 160F and I can do just as I do now and embed for an addtional 2.00 cost, the NTC heat sensor calibrated to trigger an isolated interrupt if the temp exceeds a preset value (resistor sets the value).
What is your current gearing on this DaVinci drive, I know your looking at 4:1 stage #1 (currently) and I presume 100t 5mm is within the realm of possible for some of your builds, with the new belt tensioning system - you could perhaps even reduce the primary drive sprocket from 20T down to 17t or even smaller due to additional belt wrap and proper tension. That could in theory put you at 5:1 or even as high as 5.8824:1 for your first stage.
I know you gear out for 40ish mph most of the time (especially on the 3220s) so I would assume the 12S gearing is likely about (presuming 26" rear wheel) 13.5 or 14:1... assuming the highest stage #1 reduction is 4:1 that would mean a final drive ratio of 3.5 : 1.
If you were to jump up to 16S and sensors (done properly) - the temp should stay lower (by producing more torque sooner and for a longer pull to provide faster time to efficient zone, also by running in the optimal power / efficiency range of the motor) so the new reduction you would want (presuming you had 14:1 before for 40mph) 19.3545 : 1 total which means without changing the pulleys you would need to acheive 4.8386 : 1 in stage #2 - could use #25 chain to do this, 80t rear sprocket with 18t drive #25 sprocket - or you could have a custom rear 130mm BCD chainring made, using 13t odessy freewheels you would need a 63t rear chainring to get close but wouldn't that put you into a much more powerful and efficient motor speed?
Sorry to go OT - had some successful results with the aluminum sensor rings lately and considering producing them on a more mass scale - don't worry Matt a free 9FET infineon controller modded for 60v nominal operation to be run with 16S packs (it will work from 12S to 20S) at 100A or 6KW continuous, phase current is set to 2x or 200A but reducing this if there is belt slippage will reduce startup torque - block time would be one other item to play with - we can discuss this at more length once you receive your rings and your controller... I will include two wired and ready to go rings in aluminum and 1/2" thick, you already know where they go and between the hint you mentioned Bob gave you about a year ago with regards to ABC locations, it seems wiring them 90% of the time will be step by step easy (call it 10 minutes + perhaps drying time, depending on how you mount).... I will add two small dimple screws (pointed set screws) you can use to gently lock the timing rings in place - normal 5 conductor wire can be used as extension if needed but I will also include 2 10cm 5S JST-HXT Male/Female Extension Cables so you can wire up plug and play.... Even if one of your motors turns out to be difficult to calibrate, it's about 15 minutes total with the watt meter in place and the bike suspended or rear/drive tire lifted to get lowest no load currents.
Back OT: Three Questions:
1.) So when are you going to be ready to ship these / my NuVinci drive??
2.) When do you expect to be able to offer 1 - 1.5" mounts again??
3.) How many AH/WH did you pull durring your testing, what are your stats (what happened to the days we posted our Eagle Tree or HV ICE logs?)
Matt - PM or call me about the drive pulleys we spoke of, there may be an issue (or may not be) depends how fast you need em?
I know to many 585 or more is alot for a "drive mount" as they see it, but let me aliken it to my Generator (somthing I just discovered in the Hurricane recently past) - the unit is 199.99 new (great deal right), for 1000w (really 800w constant, just enough for the fridge) and I only paid 50.00 for it but the freaking 2 stroke **(yep I made a foobar and bought 2 stroke gene, it's going don't worry) is rated as 2HP... so it take 2HP of ICE (internal combustion engine) to provide just a touch above 1HP of continuous power, 1.25 hp max for a few seconds? Thats worse than 50% efficiency.
Add to that a full Astro 3215 or 20 will set you back 560.00 or 695.00 and even if you add hall sensors plus the few additional bits of chain and sprocket (sans power / battery) your looking at worst about 1500.00 with an HV160 setup about 1800.00 still without power so is it for everyone, no... but run properly it's 90-93% efficient. A good 3KW gas generator is about 85% efficient, a decent hub motor (run at the proper design voltage / current for the wind and copper weight) is at best 80%... in price comparison, a good 3kw generator will run you several thousand dollars (1-3K) and a hub motor kit of a higher end (figure capable of 3KW) would run you in the neighborhood of 450.00 without power for a 26" rear but you would also need to fully rewire the system, replace the controller with a 200-250.00 aftermarket setup, shorten the leads, replace the phase and power wires and that all adds up to about 800 - 1000.00 if done right, including the torque plates you would have made to street that power without breaking your frame - oh yea, at an additional weight offset of approx 16 lbs, again without power.
Now does the 1500-1800 for the mechanicals seem so expensive now? Not really, especially if it's setup properly - done right the Astro Solution (even a lowly 428.00 3210) will provide a more efficient, more able ride with ample power in reserves and the ability to climb hills like a billy goat (if geared low enough or shiftable)?
IMHO (as Matt can tell you) for the time he puts into it, the wonderful machine work and polished or annodized finish, the constant re-evaluation of the status quoue to improve it either in thermal, power output, ergonomics, strength, or any other enhancements... That's why I ordered one and once the 1.5" mounts are available again (because they are cleaner than anything I could make for tube mounts) I will be ordering at least 2-3 more for customer builds
Oh yea - for the record, though I speak with Matt often... I don't work with him, for him, receive a discount for anthing I say or don't say - heck I have to pay for my review units with him (limited production runs, it's fully understandable) and not at dealer cost but at retail... so there is no incentive but love (in that respect the man kinda way) for my words.
recumpence
1 GW
Mike,
Yes, 6 to 1 is doable with this drive. I need to let my wallet catch it breath before I machine more large pulleys to achieve that, though. I can even go up to 7 to 1 (or very close to it). But, again, I need to get some cash flowing first. These are possibilities I have looked into.
The temp on my trike was taken right after 3 miles of back and forth full throttle blasts up a 1/8 mile road. The motor was, innitially, reading 117f at the rear of the can (the only exposed part of the can I could shoot the temp laser at). After 60 seconds it settled up at 121f. I could hold my hand on the hand indefinately. The clutch, was HOT, though and it was only slipping for 2 to 3 seconds with each full throttle blast before it was locked. So, it was not set to blow all power out through the clutch.
Oh, it was 88f ambient outside that day.
Matt
Yes, 6 to 1 is doable with this drive. I need to let my wallet catch it breath before I machine more large pulleys to achieve that, though. I can even go up to 7 to 1 (or very close to it). But, again, I need to get some cash flowing first. These are possibilities I have looked into.
The temp on my trike was taken right after 3 miles of back and forth full throttle blasts up a 1/8 mile road. The motor was, innitially, reading 117f at the rear of the can (the only exposed part of the can I could shoot the temp laser at). After 60 seconds it settled up at 121f. I could hold my hand on the hand indefinately. The clutch, was HOT, though and it was only slipping for 2 to 3 seconds with each full throttle blast before it was locked. So, it was not set to blow all power out through the clutch.
Oh, it was 88f ambient outside that day.
Matt
-Bugsy-
100 mW
What is the total RPM reduction at the freewheel output?
Beautiful design Matt. Hats off...
Beautiful design Matt. Hats off...
recumpence
1 GW
The drive is available in 4 to 1 and 5 to 1 ratio. If running a 3220, 4 to 1 may be required because of the number of motor teeth engaged. With a 5 to 1 ratio, the motor pulley is only 16 teeth with only 5 teeth engaged. The 4 to 1 ratio setup has 6 motor teeth engaged.
Matt
Matt
liveforphysics
100 TW
recumpence said:
What sort of bicycle brake are you running that actually works my friend?
I tried 4 of the biggest baddest types I could buy, with 220mm rotors, and 240mm rotors, and everything turns blue, warps, and the calipers themselves boil the fluid and fade to nothing, literally destroying a rotor and pads in like 20mins of aggressive riding. No brand was able to escape the thermal dissipation physics of light parts trying to eat a ton of energy in a short amount of time. If I go do repeated 1/8th mile rips on my bike, even my motorcycle brake starts to discolor (but never fades and the pads don't melt or boil). Gotta share the matt-fu-knowledge that makes a bicycle brake rapidly absorb ten's of thousands of watts of energy (just 3kw of stored kinetic energy over 10sec of acceleration and braking down to turning speed in 3 sec makes ~10kw braking) and not turn into a warped faded paper weight.
recumpence
1 GW
Remember, I am only running 40mph.
This is my yellow trike. The front wheels are 16 inch and each has a 203mm floating rotor with Hope 4 piston calipers. The rotors blue, but do not warp.
I am saved by the fact that the wheels are so small and I have two complete brake setups to stop the beast.
Matt
This is my yellow trike. The front wheels are 16 inch and each has a 203mm floating rotor with Hope 4 piston calipers. The rotors blue, but do not warp.
I am saved by the fact that the wheels are so small and I have two complete brake setups to stop the beast.
Matt
liveforphysics
100 TW
Ahh, a pair of fronts would help a ton. I didn't realize this was on a trike. I thought maybe you had a magic-bullet figured out that I had missed.
I didn't realize this was on a trike. I thought maybe you had a magic-bullet figured out that I had missed.
recumpence
1 GW
liveforphysics said:Ahh, a pair of fronts would help a ton.
I think if I were to build a pure race bike with one front brake assembly, I would use a 60cc dirtbike disc brake.
Matt
wojtek
100 kW
do you have any dirt bkes disc brakes in mind or tested some?
recumpence
1 GW
wojtek said:
Nope, I have never had the need to yet.
Matt
AussieJester
1 TW
recumpence said:
Same...then we aren't stopping anywhere near the weight of Luke and his motorcycle with pedals.
Judging by pics i have seen of our lad he would weigh well over 210pound not sure the weight of his bike
but i'm sure it wouldnt be alot of difference between it and your trike.... What would you weigh Matt 170-175pound? Either way must be a fair bit of weight difference...
KiM
Green Machine
100 kW
From looking at the pics, and having a lot of recent experience with a matt drive this devinci looks like it is big design improvement over older drive units.
It looks slicks as hell to start with. Absolutely beautiful.
Also it gets rid of the belt tensioner roller and uses a new tensioning system, which seems like it will be less prone to belt slippage and the belt will last longer. Also looks like it will be easy to tension belt.
Also the heat sink features of this new drive system is key for me.
I am one of the guys who have one on order and am excited.
It looks slicks as hell to start with. Absolutely beautiful.
Also it gets rid of the belt tensioner roller and uses a new tensioning system, which seems like it will be less prone to belt slippage and the belt will last longer. Also looks like it will be easy to tension belt.
Also the heat sink features of this new drive system is key for me.
I am one of the guys who have one on order and am excited.
sk8norcal
1 MW
cant see the pics ???
johnrobholmes
10 MW
I put in a request to the mods about the pics. Looks like we may have had another attachment loss.
liveforphysics
100 TW
AussieJester said:
I'm a full 100kg right now KiM, with less muscle mass than anytime in the last ~12 years, and continuing to spiral downhill in my physical conditioning state.
Soon they will come to harvest me for my blubber to make lamp-oil, and I will be too weak to put up a defense. 
I need to get back into Jiu Jitsu and/or swimming when I'm not working all >12hr work days. The biggest difference though is going to be the speeds not the weight. The kenetic energy the brake has to absorb is equal to mass x velocity ^2 /2, so if Matt is going 40mph when he brakes, and I'm going 60mph when I brake, I have double the energy my brake has to absorb. If he is working with two brakes and half the energy, then it's only 1/4 the brake heat, so it seems totally reasonable that bicycle brakes could survive and function well in his application.
Similar threads
