Trampa EMTB EToxx Dual VESC SK3 6374 10S6P DIY Battery

whitepony

10 kW
Joined
Feb 19, 2015
Messages
663
hey guys, if everything goes well I can testride my trampa tomorrow! soo, time to put up some pictures before all the shiny new stuff is dirty :)

Speccs:
Board: http://www.trampaboards.com/35-holy...het-bindings--685t-mountainboard-p-12369.html
Mount: eToxx V2 chaindrive 1:6 http://www.e-toxx.com
Motors: 2x SK3 6374 192KV http://www.hobbyking.com/hobbyking/store/uh_viewitem.asp?idproduct=42024
Enclosure: B&W Outdoor Typ 500 http://www.b-w-international.com/product/type-500/
Battery: DIY 10S6P (750Wh, 60A continuous current) with Sanyo NCR18650 cells (10A continuous, 3450mAh, fairly stable capacity under load)
Controller: GT2B Badwolf V2 mod
Estimated range: 40km
Topspeed: 43km/h
Weight: 14.5kg :? (7kg trampa, 3kg battery, 1.7kg 2x SK6374, mount + box + capacitors + vescs + 6x long phase cables = 2.8kg)

lets go with pictures and minimum blabla! :wink:

here it is:
trampa_emtb_done1.jpg


and thats what its made for - we probably have about 10-20km dirt trails in our forests nearby :D
trampa_emtb_done2.jpg


the gorgeous etoxx motor mount, "unblinged" a little with some 3M carbon sheets:
trampa_emtb_etoxx1.jpg


flattened the shaft with my dremel and pimped the chrome looks of the sk3s so they would be more stealthy and less shiny - used the same 3M carbon design:
trampa_emtb_motorcarbon.jpg


some impressions of mount and motors with the beautiful board *gasp*
trampa_emtb_motors1.jpg

trampa_emtb_motors2.jpg

trampa_emtb_motors3.jpg


I didnt use 2 boxes like the other builds, cause I wanted to keep the battery wires short due to some sensitive vesc behavior on longer battery cables (its generally not good to have long battery wires - much better to have long motor wires!). in the end I had a fairly complicated looking shrink tube setup that seems to work well at least. shrink tube is pretty thick special stuff with glue inside. after it cools down, it really hardens a lot:
trampa_emtb_wires.jpg


speaking about boxes: many builds use the ultrabox uk308, but I found the perfect match in the B&W Outdoor Typ500! the box is mounted with that sick 3M dual lock and has room for my 10S6P, 2x vesc, 1x capacitor bank and maybe some keys & money. also, it looks a little better than the uk308 in my opinion :D
trampa_emtb_outdoor500_1.jpg

trampa_emtb_outdoor500_2.jpg


the 6 phase wires arrive at the box in a really thick glued shrink tube. pretty glad that I made such an exact hole for the tube! it will be easy to get this box really waterproof!
trampa_emtb_motorinput2.jpg

trampa_emtb_motorinput1.jpg


as you can see from the last picture: I decided to remove the capacitor bank from the vesc since it gave me trouble on my last board and replaced it with a shared giant capacitor bank (its way oversizes, but I had the caps and I had the room in the box - in the end it will smooth out spikes very well):
trampa_emtb%20capa1.jpg

trampa_emtb_capa2.jpg

trampa_emtb_capa3.jpg


also I added heat sinks to the vescs - they are cheap aluminum gfx card coolers with double adhesive thermal tape:
trampa_emtb_capa4.jpg


the layout of the box is like that:
trampa_emtb_layout.jpg


but before getting to the battery, I fixed some tooling ... also cause I was building another evolve 10S6P battery at that time. used a bench from my granddad and equipped a piece of hard wood with an aluminum angle from my old enclusures of my other builds. that way I can easily align up to 10 cells in parallel!
trampa_emtb_newbench1.jpg

trampa_emtb_newbench2.jpg


also reshaped my electrodes to be a little thinner:
trampa_emtb_eletrodes.jpg


and off we go with another DIY battery:
trampa_emtb_newbench3.jpg


all serial connections are made from 6 nickel strips, each able to carry 15A while the final plus and minus pole is reinforced with my good old copper wick to actually get the current to the capacitor bank!
trampa_emtb_battery1.jpg

trampa_emtb_battery2.jpg


everything wrapped in fiber glass reinforced tape for structural stability:
trampa_emtb_battery3.jpg

trampa_emtb_battery4.jpg


shrink wrapped (didnt even add balancer cables for now, will check in half a year probably, but so far I never ever had drift in my diy batteries, even after heavy abuse):
trampa_emtb_battery5.jpg


and fitted into the B&W box: perfect fit really:
trampa_emtb_battery6.jpg


you can see the XT90S loop key in that picture, also the vescs are still naked, gotta shrink tube them after setting them up. minor things to do tomorrow before my first testrun. more updates tomorrow :)
 
as usual with you... this is a masterpiece. I like the caps stuff, going out of standard. Each of your builds bring innovation and a new approach. Awesome !!!!!!

10S6P... have you plan to travel to Paris by the road with your board ?!!!!
 
okp said:
10S6P... have you plan to travel to Paris by the road with your board ?!!!!

I like longer tours and with the trampa Im free to go wherever I want, thats why I wanted a large battery. the way I ride and knowing pneumatic wheels, I will probably only get a range of 30-40km. downside of the battery: its only good for 60A continuous current. hopefully the board doesnt feel tame with only 2kW continuous and about 4kW burst power. :?
 
With 10A continuous per battery you get
10A x 6p x 10s x 4v = 2,4kW, no?

Thats actually a sh*t ton of power!
Do not complain :)
 
Another great board!
Sanyo NCR18650GA? I'm doing also 10s6p with those cells, and hope they deliver enough amps for my dual hubs:/
 
2 lessons learnt from my tonights maiden test rides:

* 3M dual lock adhesive tape is NOT enough to safely secure the battery box
* I should stop fiddling with FOC for now, grilled my DRV8302 chip when I switched to foc, ppm & current control :cry:

my setup with 1 remaining vesc had quite some trouble getting up a steep hill - it looked like okp's video where he got slower and slower. my single rspecc board wouldve performed better there. not sure what the extra motor with dead vesc did though, possibly it was braking quite a bit. when i checked the temperature inside the box, the single vesc was quite warm.
 
RIP DRV ! By the way I went back today to the hill... the Turing 40/50C have HUGE power delivery. I can climb the same hill starting from zero speed... at 1m distance from the start of the hill. Impressive power. Multistory 10C is just bullshit.
 
okp said:
RIP DRV ! By the way I went back today to the hill... the Turing 40/50C have HUGE power delivery. I can climb the same hill starting from zero speed... at 1m distance from the start of the hill. Impressive power. Multistory 10C is just bullshit.

I find the multistorys car parks ( turnigy multistar) fine at 4c, their description says they are for a low current long duration discharge eg small prop multitrotor use not any high amp use so for their purpose i suppose they do deliver, but the 15c peaks do seem unrealistic the voltage sag would be highly noticeable more like 5c 10c peaks.

Great build though lots of quality throughout looks very professional i had a monster truck skate board as a kid but not agile enough now or your build be right up my street.
 
after killing both of my vescs (1x FOC, 1x unknown) and switching back to tanky torqueboard 12S escs from my topspeed, I still encountered heat problems in the dirt when others apparently have no issues at all ... even with vescs stuffed in smaller enclosures without air vents and under higher loads!

Im still in the process of finding the reason, for now I took a new path for the motor wires cause I was suspicious about the 6-wire-loop around the truck, which actually looked like a copper coil tightly wound around a closed loop metal core. at first I thought "OMG, IDIOT!!!", cause I felt like I was inducing currents into the truck, which in return induced currents into the other wires, acting like a higher resistance, leading to higher load --> overheating of escs, possibly inducing weird voltage spikes and killing off vescs, and and and ... but then I realized: current flows in and out of these 6 wires, so there wont be no real induced net current in the truck. but yea, in that process I cut off about 20-30cm of the motor wires, so thats good anyway I guess. during that inductance mania I also separated the 2 motor wires leading to the 2 motors cause you never know. also I fixated the motor wires to the etoxx motor mount to make sure that they wont move and kill my motor like on my tesseract.

trampa_emtb_cables.jpg


my short test run where I ran up a hill kind of went halfway well, but also the torqueboard escs got quite warm again like in the dirt ... BUT they didnt shut down yet, so maybe a small partial success.

after that test I drilled in more air vents into the enclosure (same amount of holes in rear). in the end I can live with vacuuming the box every now and then if the soon to be repaired vescs and battery stays cooler. :)

trampa_emtb_heat.jpg


overall I had a lot of fun though - we have really awesome dirt trails in a forest thats only a 2min skateboard drive away. crashed once because I couldnt recover from unwanted drifing ... went drift to the left *uhohhh* drift to the right *shiit* drift to the left *hnnnghh* and my tyre got sudden grip on a rough rock, which finally made me fall over. :lol:

definitely should ride with more respect, crashing on forest ground also hurts and worst case noone will find you :p made a few speed jumps on the trails, but nothing like the miles high jumps of mister nowind. love how the flex of the trampa irons out everything thats in your way. from a board perspective Im really really happy, the traction is insane compared to my tame evolve, bindings are just awesome but also quite scary (especially with the heel loops - you just wont get out), but well: electronics bleh. Ill keep working and testing. :wink:

trampa_emtb_dirt.jpg
 
my 2 vescs finally returned from the service guy who exchanged 2x DRV8302 and replaced a few burnt connections on the board by wires. turns out only one of the vescs worked, the other had trouble during motor setup and I returned it already again... :(

I zipped around on one vesc again like last time, but left the battery current at only 30A max like I would run it as a dual rear configuration. didnt have heat issues at all this time hmmm, could be my few air vents in the battery case, could be the reduced battery current, could be the dual 2000uF that I soldered to the vesc with just 2cm wires or really could be my changed motor wires setup. im waiting for vesc no2 and will do extensive long distance testing next weekend!
 
wohoo, 2nd vesc came back repaired, I hooked them both up with my nunchuck for now, connected via can and while setting up the motors, I noticed in the vesc live monitor: braking induced 47Vish, accelerating brought the battery down to 30V immediately, lots of over & undervoltage notifications, started to wonder whats wrong with my battery. the motors werent running at full speed either. :?

I really didnt get it until I started to smell melting plastic: the loop key!!!! I always wondered about how good a "loop" is in the connection from battery to vesc, since is should increase the inductance of the wires somewhat? anyway, the loop became scorching hot and melted the XT90S plastic.

when I removed it and directly connected the vescs to the battery, everything was suddenly running perfectly! gave it a quick night tour: awesome! had a lot of fun and tortured the vescs with all the hills we had ... got some steep 25% slopes and the vescs were just mildly warm - nothing compared to my very first try.

ooking forward to this weekend :)
 
Good to hear all is working as it should now, hope it stays that way.
Looks like the reason is coming for the weekend here so I hope you have it better and can get a good burn in.
Getting quite a collection of beauties there I must say.
Have fun. :D
 
Hey.
Nice to hear that the setup is working now!
Sounds strange that the Loop is getting that hot :shock: i didnt have any issues with it.

Anyway, wish you a nice testing weekend with lots of fun with your Trampa.
 
nowind said:
Hey.
Nice to hear that the setup is working now!
Sounds strange that the Loop is getting that hot :shock: i didnt have any issues with it.

nunchuck isnt really the right controller for this, but I just didnt have anything else available, since I killed the receiver of my gt2b with some soldering. also the cruise control KP and KI settings of 0.002 that work great for my tesseract are NOT working for me trampa - the integral part is still too large I think, slightly oscillating.

about the loop key: I know quite well why my loop exploded and yours worked: my loop was too large! most of you directly bridge the 2 soldering joints of the XT90s plug and then attached something to pull the plug out again. I thought I just use a cable and wind it in a slightly larger loop, so that I can use it to pull the plug out at the same time. the problem is, that the inductance of a single wire loop is increasing with loopsize (~ radius*log(radius)), so the inductance of my loop key was much larger than that of most others.

at least that my working hypothesis. I have a 2nd loop key with a smaller radius and that one is working. its only a problem anyway when you keep changing the currents, which I did when I spin up the motors -> break -> repeat. that induced quite large currents apparantly, large enough to really heat up 10AWG wire that killed the XT90 plug! :shock:
 
Concerning long battery wires ?,... did your previous 10s4p battery pack with one Pos or Neg lead at the far end from your Vesc cause any problems?
 
Just a note on long battery wires from my experience. Right or wrong I am just saying what I found on my setup.
My dual 500 motor build running on 8S has very long battery leads, the leads actually loop under the VESCs I have (they are installed on stilts so room for heat sinks) and back to the battery.
I had a lead already made lying around so used it, was going to shorten it but never did as all worked absolutely fine.
With my single 6374 build and VESC I had temperature problems and as a test, to see if it made any difference, I shortened the leads to the minimum. Made no difference at all. (Tested at 8S and 12S)

Generally, sure - shorter can only be better.
 
randyc1 said:
Concerning long battery wires ?,... did your previous 10s4p battery pack with one Pos or Neg lead at the far end from your Vesc cause any problems?

I didnt have any issues with that so far, but thought about turning all cellpacks by 90° quite a few times now. in that Kind of Setup, youd have + and - pole at the vesc and it would allow battery flex too in the serial connections with nickel strips!

WeeChumlee said:
Just a note on long battery wires from my experience. Right or wrong I am just saying what I found on my setup.
My dual 500 motor build running on 8S has very long battery leads, the leads actually loop under the VESCs I have (they are installed on stilts so room for heat sinks) and back to the battery.
I had a lead already made lying around so used it, was going to shorten it but never did as all worked absolutely fine.
With my single 6374 build and VESC I had temperature problems and as a test, to see if it made any difference, I shortened the leads to the minimum. Made no difference at all. (Tested at 8S and 12S)

Generally, sure - shorter can only be better.

an eye opener for me certainly was my loop key incident. wire inductance pays a major role in the stability of the power source for your esc and I think I did everything right with short battery and Long esc wires!
 
Do you think this Loop Key will cause a problem ?
 

Attachments

  • 2016-04-22 00.10.27-640x360.jpg
    2016-04-22 00.10.27-640x360.jpg
    24.9 KB · Views: 14,323
thats EXACTLY how mine looked ... before it melted. yea, I think it might be a problem when you play around with the throttle. I could actually watch the effect live in the vesc live sampling: quick throttle jerks -> battery voltage sagged down from 41 to 30, quick braking from full throttle -> battery voltage jumped to 47! I have a much smaller loop key now and I dont see ANY sag at all.

or well, maybe there was something wrong with my loop key *shrug* the effect fits inductivity very well though!
 
Ok Thank you whitepony !!, i will shorten it just to be sure ,and find another pull method for it .


...Just want to thank you for all your posts and pictures, your builds have inspired & helped me to build my own, which i will be posting soon.

You are helping many people understand this hobby !!
 
whitepony said:
nowind said:
Hey.
Nice to hear that the setup is working now!
Sounds strange that the Loop is getting that hot :shock: i didnt have any issues with it.

nunchuck isnt really the right controller for this, but I just didnt have anything else available, since I killed the receiver of my gt2b with some soldering. also the cruise control KP and KI settings of 0.002 that work great for my tesseract are NOT working for me trampa - the integral part is still too large I think, slightly oscillating.

about the loop key: I know quite well why my loop exploded and yours worked: my loop was too large! most of you directly bridge the 2 soldering joints of the XT90s plug and then attached something to pull the plug out again. I thought I just use a cable and wind it in a slightly larger loop, so that I can use it to pull the plug out at the same time. the problem is, that the inductance of a single wire loop is increasing with loopsize (~ radius*log(radius)), so the inductance of my loop key was much larger than that of most others.

at least that my working hypothesis. I have a 2nd loop key with a smaller radius and that one is working. its only a problem anyway when you keep changing the currents, which I did when I spin up the motors -> break -> repeat. that induced quite large currents apparantly, large enough to really heat up 10AWG wire that killed the XT90 plug! :shock:

I was just about to start using one of these loop keys in a project I'm working on, so my curiosity is piqued here... Why do you think it was the parasitic inductance that did this? The whole scenario seems more consistent with simply a bad solder joint in the loop key... If the connection were bad, then the resistance could cause the heating you saw. It seems like it may also explain the voltage sag you saw because voltage was being dropped across the bad joint (i.e. resistor). I guess I just don't understand the exact mechanism at work under this hypothesis... Even if there was a lot of energy stored in the magnetic field of the inductor, why would all that energy get dissipated only in the inductor itself and not divided among the various loads, parasitic and otherwise?
 
bobfandango said:
The whole scenario seems more consistent with simply a bad solder joint in the loop key... If the connection were bad, then the resistance could cause the heating you saw. It seems like it may also explain the voltage sag you saw because voltage was being dropped across the bad joint (i.e. resistor). I guess I just don't understand the exact mechanism at work under this hypothesis... Even if there was a lot of energy stored in the magnetic field of the inductor, why would all that energy get dissipated only in the inductor itself and not divided among the various loads, parasitic and otherwise?

hmmm, I agree on the heat, it makes no sense that it heats up, impedance wouldnt result in heat dissipation and the currents that flow are rather smaller compared to those without parasitic inductance. what a bad solder point does not explain is the drv8302 over currents errors though - and these were definitely gone with my 2nd loop key with the much smaller loop!

on the other hand, if I actually compare inductances of my scenario (20cm straight battery cable vs. a 20mm loop of the same cable -> https://www.eeweb.com/toolbox/loop-inductance/), the cable has a 10-fold higher inductance, so that probably wouldnt really explain that weird inductance effect either, since the loop doesnt really play much of a role. :|

* straight wire, 20cm long, 3mm cable: 19.2e-8H
* wire loop, 20mm diameter, 3mm cable: 2.5e-8H

great, now I dont know anymore what to think :p
 
Inspiring Stuff!
I ride an evolve carbon 2 in 1 and found your original post about extending range. I've bought a second hand carbon deck with EMC, controller and receiver and a 10S battery pack which apparently has one failed cell.
Should be testing that this week on night shift :)
I'm going to try and get a dual setup on the spare carbon deck as a test then I'm keen to start an AT board such as this.
Thinking a http://shop.mbs.com/boards/mountain...-mountainboard-tom-kirkman-pro-model-869.html
A friend of mine has been investigating using old 18650's for a solar charged power pack for camping and now he's keen on one of these as well.
Did you approach EToxx to make the mounts custom or do they make these retail ? cannot find info on their website besides the images of what I'm assuming is your build.
Why did you decide on the enertion motors ?
 
Back
Top