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

The frame is a BB92, this adapter was a PF30 to threaded so the frame now accepts a 1.375x24tpi threaded BB. I chose the FSA Platinum 100x148mm. Same adapters I think someone posted here before, coincidentally I had ordered them before that post. It took some time but they arrived no prob. and look to function well.

https://www.ebay.com/itm/J-L-BB30-BSA-Bottom-Bracket-Conversion-Adapter-Convertor/301825751629?hash=item4646377a4d:m:mGzn1-tgncXw7tnBk7EtrMQ
 

Attachments

  • s-l1600.jpg
    s-l1600.jpg
    42.3 KB · Views: 4,308
Dave, looking forward to the vid with the new Peregrine frameset. Trying to find a pedal friendly frame with minimal flex for my next Tangent build later on this year. No mountains where I live so a DH frame with slack geometry isn't going to cut it for me in forested singletrack.
 
by Ecyclist » Mon Jan 22, 2018 4:20 am
I don't have Cyclone or Bafang kit and I don't know if that will help but I measured my setup with my Hypocycloid powerd by 6374 motor (3200W and 125Kv).
Motor only 45W
Motor with transmission 65W
Motor with transmission running rear wheel off the ground 85W

Ecyclist in your picture it shows you are running the motor at 37v and it has 85w full throttle no load. also are you running grease inside your gearbox or oil ? as only 20w to run the gearbox at full rpm is very low ? and a lot less than daves 200w so something seems wrong.

For that 6374 motor to get 3200w what battery voltage do you run is it 37 volts ? or 52v ?

Full throttle no load is a function of battery voltage and motor size . So to compare apples to apples to the 6kw tangent 3220 that does 6kw peak you would have to compare it only at the same battery voltage u use to get full power. I.e 52v and the 6374 motor is a smaller motor so yes it would have less full throttle no load. since your motor is only 3200w peak then it also most likely would have HALF the full throttle no load of the 6kw 3220.
 
hey Dave look like you found a other place to find hight point frame . copy of the Brooklyn Machine Works hitman frame which Very rare less then 10

https://www.pinkbike.com/forum/listcomments/?threadid=176492

www.pinkbike.com/video/430700/

https://youtu.be/CFztApN-vQU
they ride good

they one with gearbox

https://www.pinkbike.com/photo/13936284/

https://www.youtube.com/watch?v=Ca79xcV_3nA

how long. did take to find one ? was it new or 2 hand frame


sorry I not touch my frame over 1 year. my. life been hard I have made all the parts. to run. :evil:


look like you upgrade gearbox gears again, dose it change the sound. best of lucky. 8)
 
It is a new Peregrine frame, his first model with the split downtube. It's super wet up here right now so no trail time yet. On pavement I've seen 7500W and noticed no frame wiggle, can't wait to compare this to the Zerode's high pivot. Peregrine is a bit smaller, sits lower, the BB is a couple inches lower than the Z, steel vs aluminum. I've even moved the same suspension components over for a fair comparison.
 
hi I seen others beside me ask for water cool gearbox , have you made one yet if should when will you put out test data

did You think about inject in oil in side the gearbox with a pump plus water cooling

I happy to a beta tester. I move back to dh World Cup tracks soon


your Love ride new bike should be like a cloud ride it, 8) 8) 8) 8)


I would make a floating brake. I have 3 bikes with them ride a lot better. ,, you not racing. they feel better

have you try new vesc http://www.trampaboards.com/vesc-6-complete--vedder-electronic-speed-controller-trampa-exclusive-p-24166.html

I got 1 year the vesc4.12 I never got to use it.


greatYou not stop the unreal work thank you , it for the love which it rare for ebike . :shock: all the best
 
aCeMadMod said:
hi I seen others beside me ask for water cool gearbox , have you made one yet if should when will you put out test data

did You think about inject in oil in side the gearbox with a pump plus water cooling

I'm thinking of doing this as well using PC water cooling components. I want to combine an ESC waterblock as well, motor really doesn't need liquid cooling as it can handle 200C (not to mention it takes a lot of flogging to get the motor anywhere near that hot) but it would be great for the ESC. If I was buying now I think Dave has improved the heat transfer enough to invalidate any liquid cooling of the motor due to the stator/housing bonding.

Anyone have photos of the PCB of the talon? I've been searching but come up blank. Most interested in knowing if it easy to get a flat block on top of the fets or if I will have to machine clearances in for other components. Rather not take apart mine at the moment as Dave has already waterproofed it. The water block could just go on the flat side of the talon but if I'm going to go to the effort why not go direct on the fets.

Dave any insight into your waterproofing method? I would guess silicon/acrylic conformal coating on the PCB and looks like a little epoxy where the wires and LEDs are. That correct? Possible to provide any pics of the internals?

If I can provide a little input, I think there are better options for sealing the wires into the ESC as epoxy gets too brittle. A silicone acetate should provide good flexibility and sealing.

I'm really debating if a large rad would have enough of a benefit to justify attempting LC on the motor. That's a lot of heat to shed and most of the time your building heat at low speed when your rad is going to be least effective. Fans are out of the question, passively I think the ESC cooling makes sense but the motor cooling might be more effort than its worth. Cooling the motor also brings in other fluid considerations that wouldn't be an issue just cooling the ESC.
 
The Talon is difficult to get apart, I've broken two prying them open. Several chips are glued to the black heat sink case. I fill it full of ACF-50 and let it drip out then seal the LED on the face with clear epoxy and cover the phase wire entrance for good measure. The Castle ESCs forgo active freewheeling but use very good cooling. I've had my eye on these for a while: http://www.yge.de/en/high-voltage-hv-2/

The active rectification will reduce the heat generated within the ESC under partial throttle probably 4x. We can really abuse the Talon tho, it's well built and performs great for about $200 ($300 is the equivalent YGE, and it's basically bare PCBs). Here's a HobbyKing the ripoff version: https://hobbyking.com/en_us/hobbyking-yep-180a-hv-4-14s-brushless-speed-controller-opto.html?___store=en_us.
 
Configuring the YGE/Hobbyking ones through that blinking card is a pretty bad experience, compared to the CC app, but on the other hand the card can be used in the field.
I didn't like all the beeping that the CC produces on start up but I've at least managed to set mine to only two quick beeps now and that's fine (it's not counting the battery series anymore, it can't be programmed for 14s so it gets it wrong anyway). The audible pop when connecting the battery is a bit annoying. This pop is less or non existent with other controllers I've tried, same battery and plugs.
 
The HV160's haven't proven to be super reliable. The RC forums talk about the Talon running cooler, I wonder if the rectification diode is cooled better in the Talon than HV160. I would say that most of the HV160's we've tried have blown after some time, the Talons have not.

The pop when attaching the battery leads to the Castle ESC's will always be there without an anti-spark connection.

There's not much to program on the ESC board itself, maybe motor timing and starting power, throttle end points; all stuff that only needs to be set once. I have the YEP 180A clone ESC, it needs a cooling solution since it's just a PCB with a ton of FETs. Haven't had the time to design an enclosure for it yet for testing.

-dave
 
12-C said:
Mugenski said:
Anyone have photos of the PCB of the talon? I've been searching but come up blank.

Hope this helps lol

post mortem (2).jpg

It's all potted up. Ugh the worst. The YGE or hobbyking controllers would probably be a better starting point for LC. I've always been happy with CC since I started using them over 10 years ago. Their startup algos are spot on and they take a ton of abuse, have yet to cook one.

I've looked at the phaserunner as well due to it's rather robust design but I think thermals would be an issue with the 6kw setup. I suspect it would suit the 3210 well but you're still going to pay more for it than a talon. VESC is nice but again it seems a little light on power for the 3220 and would also need a cooling solution.

At this point the talon seems best. Liquid cooling seems to be too much hassle if you can't get to the FETs easily. I think adding the larger heatsink is the best trade off all around for now.

I started off using the AS150 connectors but have since switched to the provided XT90AS as they are a little cleaner. They've been holding up fine to 150A and occasional 200A burst. I do notice a little warmth on the 8g wiring but nothing to be concerned about yet. Think the talon will trip before anything else goes.
 
Nope, the CA V3 DPS, it has all the options. The engine current doesn't run through the CA, it measures voltage/engine current and gets it's power by tapping into the ESC leads (shunt on the battery return side).
 
Hi Dave,

Couple more questions... I read this article on a build with your kit and two questions came up.

https://www.electricbike.com/tangent-ascent-6kw-the-ultimate-mid-drive/

1) Are the mounting templates still available for download? Before finalizing the kit I ordered I’d like to try a few to see which would fit best.

2) Is the ESC limited to 90% of a 14s? I’m building the pack so should I build 12s or 14s. I have enough HQ2’s for a 14s8p so this would be the time to know.

Thanks!
 
I’m building the pack so should I build 12s or 14s. I have enough HQ2’s for a 14s8p so this would be the time to know.
My opinion as a fellow Tangent (and Ducati?) rider: aim for 14s and as many parallel as you can. You will see some serious voltage sag (5-8 volts) on 18650's under these loads. I have fed mine Samsung 30q in 14s6p but recently switched to 16ah multistar lipos and I won't be coming back. But you'll do better than me if you plan on using 8p.
 
Dave,
Couple questions:

-As production ramps up, have you learned anything interesting resulting in design tweaks? Can you update us on the potential delivery times for those of us pretending to wait patiently for ordered units?

-Returning to an old exchange regarding the best medium to transfer heat from motor to case: It is clear that water is close to as good as it gets, but it is 17 degrees here this morning. Are you adding any antifreeze?

-As an industrial maintenance manager with an annoying safety preoccupation: What is the best solution so far for a complete battery disconnect or estop function? Does anyone have experience with the Lunacycle solid state switch? Alternatively, can anyone affirmatively state that there is no failure mode in the control system that would result in a full throttle runaway?

Chris
 
Alternatively, can anyone affirmatively state that there is no failure mode in the control system that would result in a full throttle runaway?
I have some experience here. I have broken a few controllers on a few different rigs, and these CC ones have only failed catastrophically- full shortcuts and very little drama. I’ve got limited ghost throttling when the exposed JST connector for the throttle got real wet and a little salty, but this disappeared with some ACF50. On bigger Infineon bike controllers I’ve had scarier fails with throttle malfunctions, due to water/salt interfering with parts of the much larger circuit board. A backpack xt90 pack plugs out real quick, fortunately... Maybe a big red kill switch would be nice addition. I’ve also had a moped throttle lock wide open on a busy street with pedestrians, that was all mechanical failure.
But if you want a totally “safe” vehicle maybe don’t build it yourself :) real motorbikes and cars need lots of certifications and need to comply with lots of laws. They are going to be safer in this regard. But never as much fun!
 
The Talon ESC doesn't like voltages above about 58.5, it gives a warning. Full charge on 14s is 58.8V so we're only talking about super fresh off the charger. This is at worst a minor annoyance for literally seconds, use 14s or you'll be wishing you had the first time you try it. Just charge the 14s pack fully and use the motor to drain the top 1% off the charge if needed, reset the ESC warning, resume your ride as normal.

I've attached the PDF plates, double check your printer isn't scaling the page or anything, can also actually measure the 211mm as marked to verify. I'm imagining there will be enough demand for a second design that curves the opposite direction to fit around suspension stuff. I'll make that as needed. I'm adding ISCG tabs to the mount also.

Production...about half of the parts are finished for this first batch of 20, about time to start the housings. Each needs about 2 hours of CNC spindle time. Most of the tweaks are small and related to manufacturing difficulty of the feature. The ring gear for instance- the oil channel features are very costly and I actually see not much benefit on the bike, so I had to omit them. Eccentric gears will always have the oil channel but I'll have to get them into the ring gear in a different manner. This is the biggest change to the design for increased production, other examples might be the shape of a pocket is altered a bit to match the cutting tool I want to use. I've made so many variations of this overall design (CAD->CAM->CNC->assemble/test) that I got most of the features right during the CAD portion of the 2nd gen process. In my head I don't design the gearbox anymore but layout CNC operations/tooling. Most of my time is spent making sure bolt holes line up and they're in the right spot, oh shit the ESC is too close to the crank gotta move it, oh shit moving the ESC makes it too difficult to connect the motor gotta move it again, oh shit now I've gotta make both mounts a little longer to fix this, oh shit I'm out of aluminum plate gotta go get more. The the cycle repeats for the next issue.

I'll have the first 5 or so 'upgrade' versions shipping next week I hope; I'll see if I can't talk somebody into posting their impressions since the gearbox half is identical in the new design. Astro is working on the motors and they should be finished in a week or so, about the time I'm ready for them.

Adding some anitfreeze to the water jacket is probably a good idea, it's only 10mL of fluid so you'd only need like, not much. The backpack battery is a pretty sweet failsafe disconnect, like rip cord, just grab and yank. Otherwise I'd say the solid state switch from Luna. Their site says 12gauge wires, I might consider stealing the PCB and soldering my own wires. The CA's throttle scheme I think is more foolproof than some of the Chinese controllers. Increasing voltage is read as increasing throttle and we set a over-voltage limit that indicates a short circuit and disables the CA. The digital throttle signal between the ESC and CA means that line is immune to short circuit causing motor runaway.

-dave
 

Attachments

  • Ascent standard mount, ESC.PDF
    22.4 KB · Views: 98
  • Ascent standard mount.PDF
    22.6 KB · Views: 94
Thank you Knurf and Dave,

Knuf – I don’t seek certified failure proof products. I raced motorcycles professionally, did technical rock climbing, time trials on tandem bikes, flew homebuilt aircraft and so on. I just try and anticipate failure modes and head them off. I expect I will break something as I push the bike a bit. Then you re-engineer to make it stronger, or learn not to push quite that hard in that way on that part. I will probably try the Luna switch. However I will plan for the possibility that it will fail and make sure I have enough slack in the power cable to jump around it. The point here is to see how outrageous we can get, not frighten the children or dogs, and not be too badly injured in the process.

Dave – Appreciate the update. I am familiar with the iterative design process as I also dream CNC sometimes. I work in wood, but the evolution of the design is similar – draw, jig, choose cutter, cut – repeat. I will see if I can dig up 3 cc of antifreeze.

Chris
 
Clyon2 said:
Dave,
Couple questions:


-Returning to an old exchange regarding the best medium to transfer heat from motor to case: It is clear that water is close to as good as it gets, but it is 17 degrees here this morning. Are you adding any antifreeze?

Try Evans waterless coolant, good for high/low temps and has no thermal expansion(apparently) which would be ideal for something like the tangent unit since expansion/contraction could lead to loss/vapor buildup, which would lead to further loss and vapor buildup as with cycles.


Chris
 
Clyon2 said:
However I will plan for the possibility that it will fail and make sure I have enough slack in the power cable to jump around it. The point here is to see how outrageous we can get, not frighten the children or dogs, and not be too badly injured in the process.
We're definitely on the same page there. Throttle failure are my main concern, followed by a wheel collapse or coming off.. But also I need to remember to respect the chains and mechanics of these things. It's so much easier with a combustion engine, you kinda know when to keep your fingers and hands away if it's idling. It sounds and even smells dangerous.. These electrical things are silent and seem dead but will cut fingers with just an accidental bump to the throttle handle. I'm sure all of you already do this, but I sometimes forget: Always unplug before touching drive trains or wheels!
 
3/8. motor shaft. or 1/2

3/8" motor shaft. The motor sees no shock or radial stress, the gearbox handles all of that so there's no need for a beefy motor shaft weighing us down. A 1/8" key is used between the motor shaft and input camshaft of the gearbox.
 
Clyon2 said:
Dave,
Couple questions:

-As production ramps up, have you learned anything interesting resulting in design tweaks? Can you update us on the potential delivery times for those of us pretending to wait patiently for ordered units?

-Returning to an old exchange regarding the best medium to transfer heat from motor to case: It is clear that water is close to as good as it gets, but it is 17 degrees here this morning. Are you adding any antifreeze?

-As an industrial maintenance manager with an annoying safety preoccupation: What is the best solution so far for a complete battery disconnect or estop function? Does anyone have experience with the Lunacycle solid state switch? Alternatively, can anyone affirmatively state that there is no failure mode in the control system that would result in a full throttle runaway?

Chris
Not sure if I'm on topic but I had so many (runaway) near misses from accidentally nudging the throttle (in the shop & while fixing my bike on the trail) I've now placed a Bafang brake sensor close to the end of my LHS handlebar grip, then I fix a round magnet onto my
LH glove. (in the crotch area between thumb & forefinger) in this way, the battery is only "live" when I have my left hand on the bar.
 
Back
Top