Dave's 80:1 RC mid drive kit build log

My bike is coming together well. Tweaks here and there. Keeping the wiring clean is probably the most time consuming thing to get right.
Pretty!

Reminds me I need to work more on protecting my cables and stuff.. But wow that fan would be dead in minutes if it was on my rig. My latest adventures were more like some form of boating or mud paddling.. Incredibly wet and dirty but so much fun! I have moved my ESC to a separate box, since I killed one or two (early revision ones before the seals got better) with road salt and slush. It won't overheat in there no matter what. Only time managed to get the ESC to overheat was when I tried changing the PWM to 24 kHz (while being sealed up). Faster switching increased ESC heat. It's a HV80 that I limit at 70 amps, 16 kHz. Always @52v.

Overall, the rig has been performing incredible the last few months. I did have problems initially but Dave tirelessly helped me out with all of them. Thanks again! Really looking forward to see the next thing.

IYqZWSZ.jpg
 
knurf said:
But wow that fan would be dead in minutes if it was on my rig.
No doubt! I don't expect it to last long at all. Only been using it for street riding thus far. First good hit of water and dirt will have it taken back off. More ideal would be making a larger heatsink or utilizing the flat surface on the back of the talon for some more mass. I've been beating on the talon pretty hard, I'm trying to get rid of an old chain/cassette so I can mount a wider range one but it just won't give up. Thinking about a steel narrow wide chainring. Been using an ebay alu one that seems to be wearing fast. Not sure if it's because my chain is worn or it's just to soft for the power.

Also working on a 3D printed TPU cover for the motor freewheel. I'm getting a lot of crap bouncing and sticking up in the motor chain. Probably worse due to over greasing. When it's done I'll make sure to post it up for anyone else. I know I have haphazardly greased my leg by bushing up against it.

Knurf, any reason you didn't mount in the triangle? That's the one big thing I wish I did differently.
 
I'm trying to get rid of an old chain/cassette so I can mount a wider range one but it just won't give up.

Hehe, I'm sort of doing the same with my whole bike (want something with more suspension travel) but it just won't quit. There are many small mysterious paint chips on the rear triangle and other odd places, I think they might be from the material flexing under load, but stuff still holds up...

any reason you didn't mount in the triangle?

Yeah I did that initially, but suffered problems with chain drops/twisting/chaos.. However, that was also before I started using narrow wide chain rings. They make a world of difference and let me use all gears. I think I had less problem with my chain lines and gears, even before narrow wide upgrade, by just moving the motor under the down tube. But yeah, narrow wide rings should be mandatory for this kit. I have the Hope Retainer ones, not worn out yet. 7075-T6 alu is pretty durable.

I might go back and install in the triangle sometime, but now I kinda like to have that water bottle. Backpack is heavy as it is and bouncing around in the forest does make u thirsty..

My chain/cassette wear was terrible (1000km / 600 miles) for the 11-tooth rear and commuting (another bike for that now), but run with just a few more cogs back there and stuff last twice as long or even more. As soon as the chain starts to randomly break I know it's time to throw it away for a new one - handy!
 
So for High Power & long range if you had to make a decision, what battery would be the best?

I still vote 14s over 12s; the drive works fine at 12s, lots of torque, but 14s is like electrons on amphetamines. Maybe you can order an additional 2s pack for each series string? Otherwise just gear the motor a little taller, use the torque to pull more chain. You have a 4turn 3220 which makes a load of torque in the 100A range.


Gen 2 Tangent serial 0001, ready for flogging (I mean testing of course)
20171123_122823.jpg


-dave
 
Gen 2 Tangent serial 0001, ready for flogging (I mean testing of course)
Yeah, cool! Now go be real mean to it and lets see what it can do..
 
dtx said:
Hows the cycle life of the rc cells been for you?
Me? I currently use three luna mighty mini's in parallel. Sags quite a lot under load but I don't think I've lost more than 5-10% cap after a year of regular usage. They we're expensive as @%&! to import here and I'm not sure they balance all that well. I've had to dissect one and help a cell pair back up to voltage, otherwise the BMS would cut out way early. Works now at least, but next pack will be rc lipo. 6s+4s+4s / 16ah could be to be the sweet spot for me.
 
Looks like my 3D printed cover isn't really needed on Gen2. Looks like the bearing is covered up now. Nice 8)

Just ordered a 2s pack to boost to 14s. Time to limit the wrist twisting a bit I suppose. Not going to over volt and over amp the controller, no more 200A holeshots haha. I am really liking the granularity and resolution passthru offers. Going to try Matt/CC advice on motor timing to 0 (stock was 4 advance) and drop the throttle ramp one point on the Talon for some heat reduction and controlability. Talon can come on a little harder than needed in passthru (but that's what RC hobbyists love about castle :twisted: ).

I haven't found settings I really like on the CA, it makes it feel numb all around. Maybe it's because I come from motorcycles but after running passthru going back to CA makes the bike feel like a scooter, great for reliability and lesser C batteries but not so great when I want power NOW.

Swapped 12g grin shunt wiring for 8g and a .5 mohm shunt. Someone needs to make a high capacity shunt with proper gauge wiring. Grins is great for commuters but when you're blurring the line between MTB and motos 12g wiring doesn't cut it for how I ride.
 
Hey Dave how is the HI-C battery doing compared to the multistars? I just sent Pat an email inquiring about a 20s7p pack for my mxus.
 
Just curious, how the Tangent setup compares to my Cyclone 3000w eCarbons full suspension fat bike.
I was running 20s HobbyKings 84V, and the CA3 allow fair acceptable throttle management. But on the tight trails 84v proved challenging at times so scaled back to 16s 10c HobbyKings @ 67.4V and now it is less exciting ride but surely much safer.
I am curious because I may like to try another Tangent and for obvious reason would like to stay the current battery setup of 16s- 10c - 67.4v HobbyKing Multi Rotor packs... would that work?
Tangent would replace the Cyclone 3000w
3000asm.jpg
 
Patrick's pack from Hi-C is high quality. Only a brief ride so far (testing gearboxes drag racing around the block), but the 14s8p of 30q pulls just as well as the Multistars at 5.5kW and doesn't get as warm. Top speed is a touch higher cause the 30q's seem to hold higher voltage at average power draws (1-3kW).

Elementary- you can run 16s through the Astro engines but not the Castle ESC's. 12s or 14s max. About any controller will work with the Astro if you wanted to source another, and Astro is more than happy to wind the motors to whatever voltage we ask.
 
Post
by elementary » Nov 29, 2017 11:40 pm
Just curious, how the Tangent setup compares to my Cyclone 3000w eCarbons full suspension fat bike.
I was running 20s HobbyKings 84V, and the CA3 allow fair acceptable throttle management. But on the tight trails 84v proved challenging at times so scaled back to 16s 10c HobbyKings @ 67.4V and now it is less exciting ride but surely much safer.
I am curious because I may like to try another Tangent and for obvious reason would like to stay the current battery setup of 16s- 10c - 67.4v HobbyKing Multi Rotor packs... would that work?
Tangent would replace the Cyclone 3000w

Dude I was reading your build thread didn't you already try the tangent ?
http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=83578&sid=1ecac4dc9ec042f3b61644f3cf8935df
And you preferred the more quiet and higher torque Cyclone 3000w , I cannot see what has changed from now to then ? This is a fundamental design difference between the two systems I can see small high RPM motors make more torque in the top part of the RPM curve vs larger motors have higher torque lower down the rpm curve. And also a plastic geared gearbox will always be quieter than a steel geared one. Unless you want something noisier and less torque down low ? but i think that is the opposite of what you want hill climbing or offroad with stop starts ?

Maybe on a race track where noise is not an issue or where you have long straits its beneficial to have torque rpm peak at higher RPM is an RC setup an option ?
 
I can see small high RPM motors make more torque in the top part of the RPM curve vs larger motors have higher torque lower down the rpm curve.

Torque is a function of how many amps are running through the engine, regardless of RPM. The 3210 easily makes 360in-oz of torque (that's only 60A phase), which is 1.875ft-lbs. Then multiply that by 50 or so get the torque input into the driveline. Then remember that the HV80 ESC can peak 140A....



Also, please keep your plastic drivetrain to yourself, the rest of us have some rippin' to do.


-dave
 
Ok Alex, you just talked me back to my senses... I was assuming the newer tangents would be quieter and lighter.
So Alex, thanks for saving me money and encouraged me to mod my controller to more amps an changing my throttle ramping.
You are right, I wouldn’t accomplish much by changing and I am already ripping faster than I can handle. It must be the cyclones are too dependable and I was just getting bored... guess I will go for a ride and break something.
 
Also, please keep your plastic drivetrain to yourself, the rest of us have some rippin' to do.


-dave

:lol: Hey Dave. Im really interested in purchasing your kit. Get ready for some stoopid questions. Ill re read this thread a couple of times, maximise my understanding of everything and grab the specs of my current bike and pack. Bloody excited though. Your web site states december for the new system but im happy to preorder depending on time frame.

ta
 
Questions are super useful.

13 days left on my December deadline.... Regardless, second gen is close. Building a new test stand so I can run the gearbox continuously and get some accelerated test results. One of the questions I want an answer to before shipping is a good estimate on the lifetime of the two seals. The gearbox is properly sealed now, with a seal on the output shaft and another on the motor shaft. These shouldn't see any pressure from the grease or oil inside (yet another test to run- grease vs oil) and so can be a little loose which will help extend their life and just maybe we won't need a steel sealing ring on the slow output shaft (I bet 50hrs before the seal wears a groove into the 7075, but let's see how the first unit fairs).

New enclosure for the CNC this weekend with a big coolant upgrade. Production runs will be much nicer and much more efficient now. Final 2 prototypes of the second gen to be cut this week, one for me, one for the test stand. These parts are proofing the CNC code for production, so we're ready to go when everything looks good.

Big post to follow soon.

-dave
 

Attachments

  • 20171217_164116.jpg
    20171217_164116.jpg
    118.5 KB · Views: 3,046
As far as batteries go I personally run 14s lipo in a back pack. I haven't added any caps to the hv160 and monitoring the ripple even with the longish cables there is minimal ripple. I attribute this to the very high quality high C packs I have. I have 20ah total 6p. The fact that I have 6p and very high C is what is likely keeping the ripple so low and the batteries have very low ir. I did some rides why 12s it really is notably weaker. At full charge the 14s is awesome, once down to nominal charge it's still great. 12s on a full charge feels more like a14s at half capacity honestly. I would stick to 14.

gearing 40:1 , I'm currently on a 9 cassette 11-28 with 36 up front and I'm upgrading to a SRAM pg990 which solid billet for the top 5 cogs so that you have more mating surface between the hub and cog spine. The stock 28 is actually sheared now and free wheels under power...oops. I'm actually going to gear it even lower and go with a11-34. Will be even better on some of the steep accents I tend to do. Just keeps the motor even higher in its rpm.

Mine is the gt fury on the previous page. Dave was a real gent and custom fabbed a mount bracket to the spec I sent him. This allowed me to also attach the e13 chain guides my bike came with once I had some other components spec'd and machined. It is very robust and now acts kind of like a bash guard too. If anyone is interested I can link the cad I made, I would all also advise a few small mods for improvement...

As far as throttle, the domino that he included with was a nice pieces of kit but I struggled with riding it as a bike since it was a full length twist. I like bunny hopping, mono etc and it was just too sketchy at times when I accidentally snap the throttle. I ended up swapping to a halled half length twist from grin. It works looks a charm and allows much lower and finer throttle inputs. And since its a half twist I don't wack the throttle with curb/log jumps, bunny hops etc. Now I feel like I have the power of the motto on a flickable bicycle. Real fun!
 
12-C said:
As far as batteries go I personally run 14s lipo in a back pack. I haven't added any caps to the hv160 and monitoring the ripple even with the longish cables there is minimal ripple. I attribute this to the very high quality high C packs I have. I have 20ah total 6p. The fact that I have 6p and very high C is what is likely keeping the ripple so low and the batteries have very low ir. I did some rides why 12s it really is notably weaker. At full charge the 14s is awesome, once down to nominal charge it's still great. 12s on a full charge feels more like a14s at half capacity honestly. I would stick to 14.

gearing 40:1 , I'm currently on a 9 cassette 11-28 with 36 up front and I'm upgrading to a SRAM pg990 which solid billet for the top 5 cogs so that you have more mating surface between the hub and cog spine. The stock 28 is actually sheared now and free wheels under power...oops. I'm actually going to gear it even lower and go with a11-34. Will be even better on some of the steep accents I tend to do. Just keeps the motor even higher in its rpm.

Mine is the gt fury on the previous page. Dave was a real gent and custom fabbed a mount bracket to the spec I sent him. This allowed me to also attach the e13 chain guides my bike came with once I had some other components spec'd and machined. It is very robust and now acts kind of like a bash guard too. If anyone is interested I can link the cad I made, I would all also advise a few small mods for improvement...

As far as throttle, the domino that he included with was a nice pieces of kit but I struggled with riding it as a bike since it was a full length twist. I like bunny hopping, mono etc and it was just too sketchy at times when I accidentally snap the throttle. I ended up swapping to a halled half length twist from grin. It works looks a charm and allows much lower and finer throttle inputs. And since its a half twist I don't wack the throttle with curb/log jumps, bunny hops etc. Now I feel like I have the power of the motto on a flickable bicycle. Real fun!

Thanks for the reply. Really useful info. especially the throttle and the custom mount which suits your bike perfectly aesthetically (deserves more photos please :wink: )

Is that mount shorter than the standard issue and potentially compatable with other bikes? Im also looking to go the downhill type bike route also....







tangentdave said:
Questions are super useful.

13 days left on my December deadline.... Regardless, second gen is close. Building a new test stand so I can run the gearbox continuously and get some accelerated test results. One of the questions I want an answer to before shipping is a good estimate on the lifetime of the two seals. The gearbox is properly sealed now, with a seal on the output shaft and another on the motor shaft. These shouldn't see any pressure from the grease or oil inside (yet another test to run- grease vs oil) and so can be a little loose which will help extend their life and just maybe we won't need a steel sealing ring on the slow output shaft (I bet 50hrs before the seal wears a groove into the 7075, but let's see how the first unit fairs).

New enclosure for the CNC this weekend with a big coolant upgrade. Production runs will be much nicer and much more efficient now. Final 2 prototypes of the second gen to be cut this week, one for me, one for the test stand. These parts are proofing the CNC code for production, so we're ready to go when everything looks good.

Big post to follow soon.

-dave

Excelent thanks man
 
Hey Dave I should be asking yourself about the custom mount bracket.... Do they come in different lengths? and if so compatable across different bikes? It seems that the standard one is sort of elongated to accomadate different mounting configurations.....

Just to wrap my head around this gt configuration. Seems to be a shorter bracket and visually perect.

DE83B185-42A9-4DC6-805D-ED0701CC0675.jpeg

Thanks
 
Compatible with other bikes, no problem. I'd need to know what angle the drive should be relative to the ISCG tabs.


ISCG 05.JPG


This mounting system makes it pretty quick to alter dimensions. I'm going to stick with one length each for the 3kW/6kW in the kits unless it's a special deal, there is some setup/computer time. If you want to mount the ESC in the side plate, then there is a minimum length of course.

-dave
 
You need to confirm in some manner what angle and clearance you will need for the motor to fit properly. Also I'm not 100% sure the the iscg tabs on my bike are actually true to the standard as set out by the white paper dimensions(the tab positions are true relative to one another but the relation of all 3 tabs to the horizontal/chain stay may be different from what I remember when I drew it up, the chain stay relation is also a component of the standard of I remember correctly) I will look up the info shortly and post the white paper and my cad drawings.


I just realized that Dave posted the diagram. On my bike the angle I needed I could not relate relative to the horizontal. I needed to determine the angle for the motor relative to the position of the tabs. Something about the suspension/bb/swing arm design really seemed to throw the horizontal relation way off.

Ive attached a few pics for interest.
GT fury bracket with roller and guide.jpg
GT Right plate disassembled top view.jpgGT Right plate disassembled.jpgpic 2.jpgpic 1.1.jpge13 bracket.jpgmorning ride comp.jpgmachined upper guide.jpgmachined roller front.jpgmachined roller back.jpge13 roller back.jpg
 

Attachments

  • e13 roller.jpg
    e13 roller.jpg
    199 KB · Views: 2,737
tangentdave said:
If you want to mount the ESC in the side plate, then there is a minimum length of course.

-dave

Dave, I still managed to fit the hv160 mounted to the side plate, it's just internally between them. If I recall this was a170mm length I had.

One thing that I thought would be really cool is to have side plates that are 2-piece each. The reason that this appeals to me is when I went to the mountains to do some downhill they didn't let me bring the bike because they do not allow electric bikes on the mountain. Thus I had to rent a bike, still had tons of fun...( Makes sense, with an eletic bike since you could be going 50km'h uphill while someone is flying down hill doing the same speeds, just too much liability for them I suppose. If everyone is forced to be on pedal power then everyone is forced to go in the same direction...)

The benefit of a 2-piece side plate is that you undo a couple bolts on either side and it's bacically a quick release system that allows you to drop the motor very quickly without the need to take the crank etc apart. And just as easy to replace. With a few bolts on either side it would still be just as rigid..... Allows one to ride electric in the single track and then drop the motor for a pedal only on the mountain bike parks.
 
Dave,

I checked out your site but I don't see any mention of the second gen specs?

I read a few threads up about grease seals? is that all?
 
I do like the idea of an easily removable drive. The bolted joint give me pause though, we pull on the chain pretty hard. What about the half of the mounting plate that would still be on the bike? I'd like it not to be there enough to pull the BB to remove the whole thing, but what do you guys think? Are there enough times you'd like to be able to unbolt the motor, even if it left mounting hardware on the frame?

Update on new design: These are entirely new parts, a complete redesign. In the pics is the housing for a 3220- this will be available to anybody with a 3220 or 3210 (the upgrade path). New units will have the stator bonded directly to the housing/heat sink.

20171224_102834.jpg20171224_102848.jpg20171224_102903.jpg


This is machined from solid, it starts out life as a 3.6kg piece of billet and becomes our sub 400g drive unit housing. This contains all but the last 10mm of the 3220 (where the end cap screws stick out the sides and the phase wires exit). The o-rings in the second pic retain 10mL of heat transfer fluid (water) to make up for not clamping the heat sink to the source. The cavity holds about 13mL so there is room for the water to slosh a bit; it's filled through a tapped hole in the side.

The gearbox is sealed from the motor at the motor shaft and sealed flathead screws. The outside of the gearbox is sealed against the output shaft with a commercially available rotary rubber seal by SKF. Another o-ring rides in a groove around the output cover, sealing the outer diameter.

20171224_103515.jpg20171224_112628.jpg

We'll still thread on left handed freewheels, they seem to be working fine. The motor chain is moving to the outer position, the derailleur chainring should be in a better position now for more frames. The 6kW kit uses a 16t motor freewheel to a 24t sprocket at the BB, 3kW will use the same gearbox with probably a 14/32 chain reduction.

Changes to the internals of the gearbox as well. We're using an even larger output bearing to accommodate a more advantageous positioning of the output pins. The eccentricity has been increased to move the tooth contact we're concerned with to the root of the teeth. A machined ring gear is used instead of pins- it's 30% lighter and requires less volume since the pins would need parts to secure them. The cuts remove weight and allow lube to circulate and lower pressures at high RPM. The eccentric gears and ring gear are machined from 4140 chromoly steel. The shape of the cycloidal profile has been chosen for strength, contact angle and lube retention. Machining tolerances have been optimized, this is the first copy of the production ready unit made using the production setups.



20171224_102953.jpg20171224_103029.jpg


Testing and such to follow asap, things should speed up now that the design and setups are about finished. Still need to assemble the test equipment (big ass fan) and torture both oil and grease filled units. I like oil, but we'll see. There's not enough inside the gearbox to be able to drain it effectively, but maybe a regular rinsing with fresh oil could do the same.


-dave
 
tangentdave said:
This is machined from solid, it starts out life as a 3.6kg piece of billet and becomes our sub 400g drive unit housing.

Stunning workmanship, but it must cost a fortune?
 
Back
Top