Dogati Electric Superleggera

[j3tch1u]

hey, who leaked our super secret photos?

the hot glue was just temporary (my buddy jumped the gun on the photos) until i found the approx position then i superglued and rtv'd the board to the cover. the screw holes on the can were slotted out to allow 6 deg of angle adjustment for the sensors. next i'll be hooking up burtie's TA (awesome tool btw) and try to squeeze a wee bit more performance out of this lil bad boy.

oh, i see what you mean by timing shift---good eyes! the encoder disk came unglued and i forgot to align it when i glued it back duh. no wonder my no load amps were off. thanks for catching that.

Nice!

It looks like there's a little timing shift between the rotor and the encoder disk. I guess it doesn't matter as long as the sensors line up at the right spot, which could be anywhere.

What holds the circuit board on? Just the glue?

 

[j3tch1u]

hey luke, the pcb traces are designed in a hexagonal chinese "bagua" to drive out the evil spirits haha.

btw, i got this 10 pc batch done for 20 bucks here:
http://imall.iteadstudio.com/open-pcb/pcb-prototyping.html

if you check the "open source" box, you get 2 free random boards from other ppl. maybe someone here will end up with some of mine.

Looks fantastic! Im very excited to see how it runs.

Great to see you Frank.

 

[j3tch1u]

the new dogati pack will use magnetic buss bars and cable ties. the magnets are press fit (with a dab of silver epoxy) into the buss bars. silver paste is also added to the cell terminals. a cable gun is used to tension the ties up nice and tight. the cable ties are heat stabilized and uv resistant. when all the ties are in place, the connection is very strong. the buss bars are epoxy coated on the edges and outer face to insulate against inadvertent short. pack assembly is super fast, low resistance and cells are easily replace/upgrade-able.


View attachment 1

 

[j3tch1u]

@ the frame builder /w my best friend Pudding

 

[fechter]

Wow, I've never seen magnetic buss bars. What if one cell is a little shorter than the others?

Do the magnets really give you enough pressure to prevent corrosion over time?

 

[j3tch1u]

i hope apple doesn't have a patent on that

the buss bars are 1100 electrical grade (soft). they flex and float a bit off the plastic cell holders so there is some tolerance for shorter cells. the positive disc magnets sit flush with the buss bar and the negative ring magnets protrude 0.6mm. some cells have a little nipple in the center of the negative terminal and i found rings sit more flush than discs. the 2 types of shapes help /w visual orientation. i also had to add a little offset for the holes every 3 cells to account for the "stack-up gap" of the 3-cell holders when they are connected together.

the magnets assist in positioning the buss bar and have some holding force but the cable ties do most of the compression. each tie is rated for 40lbs of tension. there are 8 ties per buss bar (120 in total). that is 4800 lbs of theoretical force distributed over an area of 24x32cm!

i am hoping the silver electrical paste will prevent corrosion (only real life testing will tell). there is also a product called No-ox-id specifically made for battery connections. after all the connections are made, the pack is shrink-wrapped to prevent water ingress. initially, i'll be using transparent shrink to check for charring from bad connections or broken cable ties.

Wow, I've never seen magnetic buss bars. What if one cell is a little shorter than the others?

Do the magnets really give you enough pressure to prevent corrosion over time?

 

[ghettoracer]

i can has color zip ties! did you color coordinated those blue zip ties benji ha ha. hi luke! are you gonna hit us up on layover in TPE air port or what??

 

[j3tch1u]

hey luke, u should drop by taipei cycle mar 4-8.

ghettoracer, they are blue because i was originally gonna use Tefzel ethylene-tetrafluoroethylene high temperature zip ties (but they are kinda overkill and costly). lower cost heat stabilized ties are good enough.

i can has color zip ties! did you color coordinated those blue zip ties benji ha ha. hi luke! are you gonna hit us up on layover in TPE air port or what??

 

[Tekler]

Hello j3tch1u.
I'm a Dogati follower on facebook and E.S. I'm building my own electric bike inspired in your work and I was near to buy a flat cells battery 48v 15ah (similary you used in your dogati V1) But now I see that you're going to use completly different battery cells. Please, can you explain me this change. Have you some tech problems with flat cells. Can you tell me how many cicles gives you the "old" battery?
Its the same C rate discharge of both cells type? It's better this new cells? Explain/help me/us please.
Tekler

 

[jateureka]

I really like the idea of magnets on the buss bars for cell contacts!
This would make it much easier to service/repair the battery if you get a bad cell.
Also makes it easier to change the shape of the battery to fit different size/shape frames.
I guess the trade off is added weight over the usual spot weld tabs

 

[j3tch1u]

Hello j3tch1u.
I'm a Dogati follower on facebook and E.S. I'm building my own electric bike inspired in your work and I was near to buy a flat cells battery 48v 15ah (similary you used in your dogati V1) But now I see that you're going to use completly different battery cells. Please, can you explain me this change. Have you some tech problems with flat cells. Can you tell me how many cicles gives you the "old" battery?
Its the same C rate discharge of both cells type? It's better this new cells? Explain/help me/us please.
Tekler

the old dogati pack was 0.8kwh. new 18650 pack (approx same size, a little wider) is 2.1kwh using the panasonic ncr18650-pf. you can drop in any 18650 cell you want. 120A @ 60v is a good match for my current setup (the a123 can put out more amps but it wouldn't be used). quality of the a123 cells were hit and miss--a few cells puffed and didn't get many cycles.

my supplier tells me the samsung inr18650-25r will be available next month. samsung may be a good bet since it's likely they will partner with tesla on their new gigafactory.

i think 18650 will be future-proof for some time. i would not go back to the flat packs.

 

[j3tch1u]

I really like the idea of magnets on the buss bars for cell contacts!
This would make it much easier to service/repair the battery if you get a bad cell.
Also makes it easier to change the shape of the battery to fit different size/shape frames.
I guess the trade off is added weight over the usual spot weld tabs

yes, if cell chemistry starts to follow moore's law then it would be great to have a mechanical re-usable kit where you could swap-in the latest cells.

word of caution though, this method is untested and i'm not sure how good the connections will be under vibration and heavy bumps. also not sure how much resistance the silver paste adds. there's quite a bit of tolerance required (up to .8mm i'd say) cuz the cell holders don't all seat completely flat /w each other. also, any torsion of the pack might create intermittent disconnections. more real-life testing to be done.

the extra weight is pretty minimal (the buss bars are only 3mm at their thickest point).

 

[Arlo1]

. also not sure how much resistance the silver paste adds.

I did some tests with silver paste. I found it conducts electricity better when added to polished aluminum then just bare polished aluminum.

 

[j3tch1u]

hey that's great to hear. does the thickness of the silver paste also have any affect (too much, too little)? i'm also curious if you experimented with different brands (i'm using MG 8463).

in my application, the silver paste is between the nickel-coated magnets and the cell terminals. on the bussbar side, i use a dab of silver epoxy and press-fit the magnets into 1.5mm deep holes (the magnets are 6mm diameter). i'm more concerned about the cell connections.

in order to get nice flat bottomed holes for the magnets to sit flush and maximize contact, i used a 4mm end mill with a helical tool-path (straight plunge would leave a conical bottom).

almost finished "lacing up" the pack (ran out of cable ties). the connections seem pretty solid so far.

. also not sure how much resistance the silver paste adds.

I did some tests with silver paste. I found it conducts electricity better when added to polished aluminum then just bare polished aluminum.

 

[kfong]

Ben, one concern are the contacts using dissimilar metals. You get galvanic corrosion if this happens over time. There are charts on the web that show which metals won't work well with others.

 

[Tekler]

the old dogati pack was 0.8kwh. new 18650 pack (approx same size, a little wider) is 2.1kwh using the panasonic ncr18650-pf. you can drop in any 18650 cell you want. 120A @ 60v is a good match for my current setup (the a123 can put out more amps but it wouldn't be used). quality of the a123 cells were hit and miss--a few cells puffed and didn't get many cycles.
my supplier tells me the samsung inr18650-25r will be available next month. samsung may be a good bet since it's likely they will partner with tesla on their new gigafactory.
i think 18650 will be future-proof for some time. i would not go back to the flat packs.

Thanks for answer j3tch1u. I'm very interested in your battery construction process. I'm readding about these cells and thinking about following your way to this cells. It's true that each cell brings his own little BMS? This simplify the construction. Are these the cells that are using?

http://es.aliexpress.com/item/10PCS-lot-Protected-Original-3400mah-18650-NCR18650B-Rechargeable-battery-with-PCB-3-7V-For-panasonic-Free/704254229.html

Thanks again.

 

[j3tch1u]

those are not the same cells i am using--they are lower discharge. the higher discharge cells are not usually protected.

Thanks for answer j3tch1u. I'm very interested in your battery construction process. I'm readding about these cells and thinking about following your way to this cells. It's true that each cell brings his own little BMS? This simplify the construction. Are these the cells that are using?

http://es.aliexpress.com/item/10PCS-lot-Protected-Original-3400mah-18650-NCR18650B-Rechargeable-battery-with-PCB-3-7V-For-panasonic-Free/704254229.html

Thanks again.

 

[j3tch1u]

that is a good point. i'll have to research that some more.

Ben, one concern are the contacts using dissimilar metals. You get galvanic corrosion if this happens over time. There are charts on the web that show which metals won't work well with others.

 

[j3tch1u]

hi guys, a moment of your time if you will. i'm trying to decide on handlebar styles.

Riser or Straight clip-ons, which do you prefer?

 

[kudos]

Clip-ons suit the motorbike looking forks better, but the risers make it look more like a bike and would probably be more comfortable on a long ride....

 

[j3tch1u]

thanks kudos...that pretty much sums up what i think as well.

Clip-ons suit the motorbike looking forks better, but the risers make it look more like a bike and would probably be more comfortable on a long ride....

 

[Danny Mayes]

At first i though the straight clip-ons hands down, but on a longer look, the riser ones grew on me. They are cool because they are still clip-ons but they are like a hybrid clip-on, somewhere between a motor bike and a bicycle. A bit like the bike itself

D

 

[johnrobholmes]

Straight clip on for the looks, but I would need risers in real life. On the risers, the amount of rise and where the clips are mounted look funny without a crossbar connected them.

Maybe a shorter rise with the clips above the top triple clamp? Would keep the grips at the same height without so much sweep to the bars.

 

[gogo]

Add an 'enduro' brace clamp between the risers?

 

[Green Machine]

I vote for risers.