DaVinci drive.....

recumpence

recumpence

1 GW
Joined
Apr 19, 2008
Messages
5,304
Location
On Earth right now. That can change at any time, t
Hey Guys,

What you are looking at is the culmination of over a year of planning (in my head), hundreds of hours of design time, and production cost in the 5 digits.

This is my newest drive. I am calling it the DaVinci drive due to its somewhat artistic appearance.

The foundation of this drive is the motor cradle and carriage. These two parts are machined from 1.5 inch 6061 plate. The motor slides through the cradle, which clamps to the motor can. This sinks heat from the motor directly into the drive unit. Belt tension is accomplished by one simple screw that pushes on the curved tab in the curved slot. That screw (accessable from the top of the drive) is simply turned clockwise to add belt tension. SUPER simple. This drive is EXTREMELY rigid and strong. The improvements over my standard V4 drive are as follows;

-Stiffer for better belt tracking
-Quieter due to the can being wrapped in the cradle and the increased rigidity of the drive itself.
-Super simple belt adjustability
-Easier service (if ever required)
-Lower Q factor (1/2 inch narrower than my V4 drive)
-More efficient (this drive arrangement eliminates the cantelever bearing on the V4, thus reducing drag)
-FAR better heat dissipation
-More power from the motor. This is due to being able to run the motor harder from the improved cooling. This should allow the user to run one size smaller motor with the same output power, or at least close to it.
-Greater power handling from the belt due to more rigidity and greater belt tension possibility
-Greater flexibility in mounting due to the offset mount foot. This foot can be mounted forward or backward, thus allowing more mounting options.

Bear in mind, this new drive is not replacing the V4. It is in addition to it. The V4 is still a good option for many applications especially where cost is concerned. However, the added cost of this new drive would be offset if the customer purchased heat sinks. So, my V4 plus 3 heatsinks (for a 3220), is very close to the cost of the DaVinci drive.

This drive was developed for a number of reasons (all of which I cannot get into here). The main reason is maximizing performance as well as useability, reliability, and adjsutability. Besides, a little Bling is not a bad thing. :wink:

I am looking into laser engraving for the DaVinci name on the drive.

We are puting the finishing touches on the first run of these drives. I am out of motor pullies right now and I need shims and other bits. But, the drives will be ready for sale within a couple weeks. Oh, I also have 3 Astro motors in stock as well if anyone wants one. :)

Lastly, this drive will be available in 4 to 1 or 5 to 1 ratio with the torque limiter.

The drive is $575.

I know it is not for everyone. But, I think, for the truely performance minded builders, this drive is the best way to go.

Matt
 
Wheazel said:
Looks nice, whats is the weight of the unit? (Everything except the motor)
Click to expand...

That is a good question. I do not have an answer because I am finalizing a few things. Innitially, it looked like it was going to be 15 to 16 ounces heavier. However, I found a few ways to shave roughly that much weight from the drive (some of the components on the drive, the drives themselves are already machined). Basically, if I went with the heaviest components, it weighs the same as my V4 drive with 3 heatsinks on the motor. However, again, it looks like it will end up being nearly identical to my V4.

I need to get individual weights of each component so I can compare everything. Basically, I weighed the motor cradle, carriage, and mount foot. Then I weighed the V4 parts and came up with a difference in weight. That weight difference was, then, my target to eliminate. I am looking into a run of aluminum clutches right now as well as hollow jackshafts. Those two changes will pretty much cancel out the added weight and make this drive a pound lighter than my V4 with heatsinks.

Matt
 
Nice work Matt, unfortunately one will need deep pockets to buy not only the drive but an astro to go with it as well :-( that makes it over a thousand dollars for single speed drive/motor combo doesn't it? :-| definitely not for everyone, sure is nice though Matt, as are all your drives. ;)

KiM
 
The average sale for a system from me is $1,400 to $1,500 for motor, controller, reduction, throttle, mounts, etc. So, the added $190 is not a huge thing. As I said, the V4 is still available. So, no worries there. :)

Matt
 
Great work Matt as per usual, another awesome option to have. :mrgreen:

What is the overall width on the new unit? Is it any narrower than the previous offerings?

Nice to see those 2 new trikes sporting your running gear, again, very slick.
 
Absolutely astounding Matt. I love the engineering and finished look. You are really creating a spectacular line of products and each one gets better. You have addressed so many issues that make this system easy to install, maintain, and use.

Love it Matt.

Clay
 
comradegerry said:
Great work Matt as per usual, another awesome option to have. :mrgreen:

What is the overall width on the new unit? Is it any narrower than the previous offerings?

Nice to see those 2 new trikes sporting your running gear, again, very slick.
Click to expand...
It is 4 inches wide at the motor, 5 inches wide at the jackshaft (from edge of clutch to edge of freewheel) and 5 inches tall at the tallest point. That is 5 inches pulley diameter and 5 inches tall at the motor cradle.

Matt
 
Beautiful and it looks like the design changes improve the function and reliability. Add me to the growing list of individuals interested in the new drive. Looking forward to seeing this installed on a bike, hopefully an upright bike as well as a recumbent.

Rich
 
I have been running the drive this afternoon and I have some impressions for you;

#1 It runs VERY smoooooth. I am not sure how it could have a different feel than the V4, but it does. It is noticeably smoother.

#2 It sounds different. It is not only quieter, but sounds different overall. I think that is because you can hear the belt over the motor whine now.

#3 The motor does run cooler. A 3210 runs very cool. A 3215 is relatively cool, but not as cool as a 3210. A 3220 runs cooler than on the V4, but the rear half of the windings are not in the motor cradle. So, the rear of the motor runs warmer than the front. However, that was only after torture testing it. I ran full throttle back and forth up and down the street untill I could smell the clutch cooking (and it was set to grab, not slip much, but I still ran it untill the clutch was very hot). Even then, I was able to hold my hand on the rear of the motor. I would say for racing with a 3220, one of my 3/4 inch thick heatsinks would be reliable, super high performance setup. For a 3210 or 3215, the drive unit pulls all of the heat from the motor into the cradle and carriage of the drive.

#4 The belt tension adjustment is AWESOME to use! It is super easy to experiment with belt tension looking for the perfect ballance between efficiency and torque holding. Outstanding.........

Anyway, that is my [totally unbiased :wink: ] operating experience with the new drive.

Matt
 
Perfectly undestandable Matt,
all that extra mass around the motor will cancel any ringing & defo make a more "solid" sound to it.
 
I am not out :) But mounts in smaller diameter tubes sizes, larger #25 sprockets with 2 3/8" ID and 4 bolt pattern to fit with Ggoodrums dual freewheel adapter would be nice, even just if you could stock the largest normal rear sprockets 130mm BCD (56t?, biggest find).. would go a long way..

The new drives look sweet, I've got all sorts of ideas already... wonder if you have considered designing one aroudn the little brother version (2" Diameter, 2" deep ... it seems to be good to > 1000w and could really hit a target market).

Cant wait to get mine :)

BTW: What Astro's Do you have Available - Turns, Version, Used / Not Used?

Regards,
Mike
 
I also jumped on the Nuvinci bandwagon - a tad concerned as noone seems to know if the maximum power is input or output on the torque side of things and it's not clear from the docs I have yet - yelp that's a power door actuator, 12.00 from some catalog I foget the name of now (top of head never works this time of day)... In the end, I will likely use the Nuvinci with a 3215 6t or 3220 4t but at about 1/2 maximum continuous as the maximum to keep the heat down, should be ample torque and with 350 degrees of shifting automatically (will be replaced with my own system in weeks time, using high current / torque servos off 700 class heli's.

This is going to be fun - I didn't see any mention yet, and I hope matt doens't get angry with me but I believe there may even be a 100t version of the system available soon, with the stock 20T drive pulley that would be 5:1 primary and in theory with the new tensioning system, I suppose a maximum (for a medium <= 4kw) powered system) would be 17 or even 16t driving the 100t for a 5.8 : 1 or 6.25 : 1 belt primary, feeding hopefully an aluminum #25 90t or 100t sprocket I'm trying to make today at the machine shop :)

Hopefully I get to the machine shop early enough to have my shafts ground and some other bits and pieces (odds and ends done).

Matt.... - when you get time, give me a call, I got another email from our new friends at Astro and a phone call trying to push as many 32XX series as they could.... so maybe it's time to do another group buy, force the hall sensors to be internally installed (they are good to 150C or 302F well beyond the Astro Bob specified limit of 212F or 100C, the SS441As) I can (any of us can by this point) provide them an internal mechanical drawing of their motors and how to implement halls but it would be nice if they could clarify their wiring methods (may already be done somewhere and I've missed it) but what I mean is that I am certain they have an ABC sequence just not sure which one leads off the A phase but that's where Hall Sensor #1 should go, dead center of slot #1 then six slots later another in dead center and finally 6 slots further yet another sensor... it could also be done by using Slot #1, Slot #3 and Slot #5 of any of the 4 electrical revolutions to reduce the wiring mess of distributing 5v and 5v gnd across the backend.

Regards
Mike
 
mwkeefer said:
I also jumped on the Nuvinci bandwagon - a tad concerned as noone seems to know if the maximum power is input or output on the torque side of things and it's not clear from the docs I have yet - yelp that's a power door actuator, 12.00 from some catalog I foget the name of now (top of head never works this time of day)... In the end, I will likely use the Nuvinci with a 3215 6t or 3220 4t but at about 1/2 maximum continuous as the maximum to keep the heat down, should be ample torque and with 350 degrees of shifting automatically (will be replaced with my own system in weeks time, using high current / torque servos off 700 class heli's.

This is going to be fun - I didn't see any mention yet, and I hope matt doens't get angry with me but I believe there may even be a 100t version of the system available soon, with the stock 20T drive pulley that would be 5:1 primary and in theory with the new tensioning system, I suppose a maximum (for a medium <= 4kw) powered system) would be 17 or even 16t driving the 100t for a 5.8 : 1 or 6.25 : 1 belt primary, feeding hopefully an aluminum #25 90t or 100t sprocket I'm trying to make today at the machine shop :)

Hopefully I get to the machine shop early enough to have my shafts ground and some other bits and pieces (odds and ends done).

Matt.... - when you get time, give me a call, I got another email from our new friends at Astro and a phone call trying to push as many 32XX series as they could.... so maybe it's time to do another group buy, force the hall sensors to be internally installed (they are good to 150C or 302F well beyond the Astro Bob specified limit of 212F or 100C, the SS441As) I can (any of us can by this point) provide them an internal mechanical drawing of their motors and how to implement halls but it would be nice if they could clarify their wiring methods (may already be done somewhere and I've missed it) but what I mean is that I am certain they have an ABC sequence just not sure which one leads off the A phase but that's where Hall Sensor #1 should go, dead center of slot #1 then six slots later another in dead center and finally 6 slots further yet another sensor... it could also be done by using Slot #1, Slot #3 and Slot #5 of any of the 4 electrical revolutions to reduce the wiring mess of distributing 5v and 5v gnd across the backend.

Regards
Mike
Click to expand...
Too funny. As you were typing, I was typing a group buy thread. :)

Problem is, I am not sure if it is a good idea to push the sensors with this group buy and possibly muddy the waters........

Matt
 
They want to make a batch, if they are calling and emailing me follow up... then they want to sell some.... I don't think were too far off to ask them to insert the SS441A into the stator and bring the wires out + 5v, -5v, A, B, C phase hall sensors at 120 degrees.

I'll make it simple as can be:

[120 degree electronic timing - works fine]

Astro 3210, 15 and 20 all share 24 stator slots... therefore, the distance in degrees between each stator slot (physical degrees now)

360 degrees / 24 slots = 15 degrees physical between center of each stator gap in an astro (all the same)
360 degrees / 4 pole pairs = 90 physical degrees per electrical RPM (since there are 4 per mechanical shaft revolution)
90 / 3 = 30 degrees from center of each hall to the center of the next

Slot #1 (A Winding I suspect) Insert Hall Sensor - for the controller this is your -120 degree sensor or A sensor - sensor must be dead center within the slot - distance to next slot 15 degrees
Slot #2 skip - distance to next slot 15 degrees, total 30 degrees
Slot #3 Hall B - Again centered exactly - this is your 0 degree sensor or Hall B - distance to next slot 15 physical degrees
Slot #4 skip - distance to next slot 15 degrees
Slot #5 Hall C - Final Hall Sensor - this is + 120 degrees electrical...

A simpler method (more complicated, longer wiring):

Since each slot is 15 degrees, 120 / 15 = 8 slots, so a hall every 8 slots would give the same 120 degree timing as above but in physical not electrical, to the controller it's all the same :)

Regards,
Mike
 
Mike:

The nuvinci will take 7hp continuous and 96 ft/lbs of torque on the burst.



Matt:


You are really tempting me with these drives. I need to replace my old v1 since it is being used as an articulated arm motor winder, but I don't know what I would install the 4v on :lol:
 
JRH,

It may not be ideal for the highest performance category unless used as a mid transmission, if it can handle high as 1000 RPM input and output (it can) then the torque could be limited below 96 ft lbs burst but for instance on my 3220 4T I run an HV110 *(same one still) and see 130A peaks max at 44.4v nominal 12S... my bike is geared for 35 mph top end and with a 169 kv and 8 in oz per A, the shaft itself produces peaks in excess of 1040 in oz (5.4167) but by the time my reduction is accounted for:

4:1 primary
4.0679 : 1 secondary
total reduction: 16.3076 : 1 to the rear
factor the increase in torque
becomes 88 ft lbs of torque so your right it would just fit but what about all the ICE HV 160s, at 12S now your over the limit by a bit, maxing out at 115.5122 ft lbs of torque to the rear.

Assume I want to maximize the performance and power of a given setup, if I switch to my 3220 6T with a much higher kT and lower kV, I believe kV @ 113 and kT @ 12 in oz per Amp and run on 20S for 74v nominal and 83v hot off the charger (segway voltages people) then reduce to the same 35mph top speed it would look like this:

Ice HV 160
Maximum RPM: 9153
Max Shaft Torque: 1920 in oz / 10 ft lbs
Nominal RPM: 8362 RPM (well in the 90%-93% powerband)
Required Reduction: 18.4874 : 1 for 35 mph
Nominal Torque: (94A Maximum Continuous) 108.6158 ft lbs

Dont get me wrong, I'm not complaining just pointing out that for some... using these upstream of the rear will be the solution to reducing the input torque (or is the limit on output torque too).

One question oh wheel wizard, how could one fit this into a Schwin Stingray Lightning (2008 I think) largest 4.5" wide rear tire (3" rim minimum)

Regards,
Mike
 
You must log in or register to reply here.

Similar threads

A
    • Like
Nirve Switchblade - Belt Drive Torque Sensing Mid Drive Beach Cruiser
Replies
4
Views
1,181
Admiralrofl
A
S
    • Like
Xtracycle Edgerunner electric conversion with Grin All Axle v3 and SuperHarness
Replies
2
Views
4,694
smitty2k1
S
R
    • Like
A critique of mid-drive riding recommendations (and hub drive criticisms)
Replies
22
Views
13,139
ebike4healthandfitness
ebike4healthandfitness
R
    • Like
    • Brainy
Really good article comparing Bafang vs. Tongsheng vs. CYC
Replies
14
Views
25,661
Chalo
Chalo
F
Direct/Fixed Mid-Drive Gearing Calculator for Tangent, LR, & CYC
Replies
1
Views
1,067
DogDipstick
DogDipstick
Back
Top