KDX 20.0 cells trailbike conversion

thoroughbred

100 W
Joined
Apr 11, 2018
Messages
225
Location
virginia
My electric moto experience is limited to the little oset bike i bought for my daughter. I would love a Redshift or Light Bee to be able to ride my own and adjacent farms quietly but this project might suffice until more play monies are available.

I would love to find a combo that would ride like a normal offroad dirtbike for an hour, weigh about 200 pounds and cost about $2000.

Here is my rough outline

12 KW motor/controller kit: http://www.thunderstruck-ev.com/budget-screamer-72v-brushed.html
$1400 and 40 pounds

DIY 18650 batteries 20s10p somehow configured
$600 and 40 pounds?


Here is the free roller someone generously donated to the cause. I was looking for a perimeter framed Kawasaki thinking battery placement would be more flexible. I had a gas kdx200 so I know the chassis will perform well enough on the terrain I wish to ride which is mostly 0-30 mph singletrack. I'm surprised this skeleton weighs 130# but I think if the project proves worthy, I can shave some weight off the wheels and forks.

kdx roller.jpg


So far $2000 and 220# estimated. 200 18650 cells at 2.5Ah each in a 20s10p pack gives me 25Ah and 1.8kWh at 72 volts?

Am I way off on anything here? Ride time is a big unknown for me and maybe the proposed battery won't come close to one hour. My last trailride loop was 19 minutes with a rolling average of 9 mph, 24 mph max. Not sure how many kW it takes to push 390 # offroad at 10 mph...

I don't know what I don't know and look forward to learning with some guidance or suggestions :D
 
In the long run I think you would be better with brushless but a budget is a budget. I think your battery is going to need to be bigger though. My bike has a 20s15p of 30Q so I’d think that would be the smallest you would want to go. I’d tell you that you would need 20p to hit 400 amp peaks be it 30Q, 25R or VTC5 or 6’s. I don’t really know the details of that power systems requirements except for the 400 amp max on the controller.

I rode a KDX 200 for a while. It was a very nice bike. Being bigger myself 6’ 4” 235 lbs at the time I went with a DRZ400.

Looks like a great project.

Tom
 
Looks like a great frame to start with. I'm looking to start something very similar soon and just noticed the perfect battery popped up over on the classified section. If the stated specs are true than a 20s 25ah pack would be 5x6x25"long and be capable of 500amp bursts! All for $440+shipping in the US.

As for motor choice. It would probably be worth it to go up to a brushless setup if you are going to ride offroad. Its Probably going to run you an additional $500 but Id rather hose mud out of a brushless motor. If those 25ah Panasonic cells from the link end up being available you will save that $500 over the 30q pack to pay for the brushless motor.

https://endless-sphere.com/forums/viewtopic.php?f=9&t=93941
 
Yes! Promising indeed on those batteries. Time to make some building blocks to test layout.

https://www.aliexpress.com/item/1pc-battery-3-7v-25ah-lithium-ion-battery-can-build-lithium-ion-battery-12v-24v-36v/32816535689.html?ws_ab_test=searchweb0_0,searchweb201602_2_10152_10065_10709_10151_10344_10068_10130_5722815_10324_10342_10547_10325_10343_10546_10340_10341_10548_5722915_10545_10697_5722615_10696_10084_10083_10618_10307_10710_5722715_5711215_10059_308_100031_10103_441_10624_10623_10622_5711315_5722515_10621_10620,searchweb201603_1,ppcSwitch_5&algo_expid=c307041c-ec2c-4cbd-82d3-f9204a8c043d-3&algo_pvid=c307041c-ec2c-4cbd-82d3-f9204a8c043d&transAbTest=ae803_1&priceBeautifyAB=0

how might these brushless motors perform?
$645 me1012
http://www.evdrives.com/product_p/mot-me1012.htm
$850 me1114
http://www.evdrives.com/ProductDetails.asp?ProductCode=MOT%2DME1114
 
I'm considering the Light Bee in addition to this thing at some point. I love the SRLB but I'm slightly worried about the Sur Ron being less stable over roots and rocks. The KDX can be licensed here.

I mocked up some wood blocks in the configuration of those prismatic cells suggested above. I think i end up with one block at 5x6x12 between the frame rails and a 5x6x8 block hanging below it (20 cells). I need to learn what is meant by "requires compression" etc.

kdx batt.jpg
 
Further investigation on those prismatic batteries has been fruitful, i guess they are a PHEV2 standardized size format

https://pushevs.com/2016/09/13/volkswagens-strategy-for-batteries/

https://endless-sphere.com/forums/viewtopic.php?f=14&t=79639&start=25

https://www.ebay-kleinanzeigen.de/s-anzeige/panasonic-25ah-3-7v-li-ion-akku-92wh-akku-batterie-e-golf-hybrid/656054997-168-2453

https://avt.inl.gov/sites/default/files/pdf/phev/batteryCMax0852.pdf

https://cleantechnica.com/2018/01/01/volkswagen-e-golf-production-capacity-double-next-year/

http://i.imgur.com/htq8Fi1.png

https://pushevs.com/2018/01/22/best-ev-prismatic-battery-cells-chinese/

BLog entry talks about charge rate:


Michael Papke
31 Aug 2017 Reply



I’m still driving my e-up! with pleasure. But sometimes on my 500 km trips a bit more battery capacity would make things easier. The Samsung 37 Ah PHEV2 battery cells don’t perform good regarding fast charging. VW decided to not use any cooling for the e-Golf and e-up! and so there is a demand for battery cells with much lower internal resistance. The 25 Ah PHEV2 battery cells from Sany/Panasonic doing this really well. A fast Charge up to 80% is done in only 20 minutes, but when battery temperature reaches 38°C, the charging power drops. At 43°C the charging power is limited to 16 kW.

However, I would be glad if there will be a battery upgrade available for existing cars. If VW will ask me around 2019/2020 if I would like to upgrade my e-up! battery pack to a new one with 27 or 37 kWh, then, depending on the price, I will really think about this option. But my hope to get such an offer is really low.
 
Is this motor/controller combination feasible? Motenergy ME1012 motor with a Sabvoton SSC72250 controller. I'm confused by this in the motor description "Continuous current of 125 amps AC (180 Amps DC into the motor control)." Isn't this a BLDC motor?

http://www.evdrives.com/product_p/mot-me1012.htm

https://www.aliexpress.com/store/product/Sabvoton-Sine-Wave-ControllerSSC72250-for-electric-in-wheel-hub-motor-with-bluetooth-adapter/1389549_32848488838.html
 
It is indeed a brushless DC motor. The Current sent from the controller to the motor is bi-directional so it can be considered AC. Just a mixing of terminology.

The different stated amps in from the battery vs amps out to the motor is due to that switching to AC that occurs in the controller. Their numbers are fine and will prevent you from choosing an inadequate battery to get the desired performance from the motor.

Controllers are always juggling volts vs amps. So if you had a system that was battery limited to 72v and 200amps and it was running at 1/2 of your max rpm than it could be outputting 36v and 400amps to the motor. In this case the 400amps would be the Phase current so you would look at the motor specs to determine if it could handle it. This is a really broad oversimplification but should give you the basic idea.

The motor and controller you listed will work fine together based on the spec sheets. I have not used either personally so that is not an endorsement. Just saying you got the right idea.

The higher power motenergy motors use sin cosine encoders instead of hall effect sensors and require a controller like the Sevcon. You can get a sevcon for about $300 more and be capable of growing into the 40+hp range. Its hard to know when to stop and easy to keep upgrading each part here and there until the budget is out the window. I really don't know at what power level the hall sensors become a problem compared to the encoders and better controllers like the sevcon.

You may want to contact user frodus on here. He runs http://www.emf-power.com/ and just recently did a group purchase on sevcon controllers and the big motenergy motors. But again, sorry to keep upgrading your project with $$$uggestions. :D
 
Thanks for the reply. Incrementalism is hard to resist and I love spending other people's money, too.

I found another build using the me1012 but its hard to tell if it got completed:
https://endless-sphere.com/forums/viewtopic.php?f=10&t=81612&p=1202035&hilit=me1012#p1202035

I think to step off the metaphorical edge, I'll order batteries today to get things rolling. Not sure if its wise to start a build around a certain choice of battery but the fitment and layout of these in the frame is a huge unknown for me. I think I can figure out mounting the motor etc fairly easily.

I'm going to try the 24Ah panasonic prismatics for the sake of simplicity (20s1p) and modularity. I have read that 50ah cells in the same configuration might be in development ....plausible upgradability 8)
 
panasonic4.jpg
Here are my batteries in their original configuration. They are held together in a line like stacked dominoes sandwiched between a couple of aluminum endplates. I think I will have to separate them into one 12 cell brick and one 8 cell brick.

As you can see, they are separated from each other by a thin plastic spacer. Not sure if that is primarily for insulation or for airflow. (they are not solid spacers, you can see a bit of daylight in there).

panasonic5.jpg

The same cells are used in a different vehicle with what appears to be no spacers between the cells:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=92246&p=1349312#p1349312

My thought is to build a rectangular box (steel or aluminum) to house the cells within the KDX frame using the plastic spacers so I can use the supplied busbars between the cells. Do you think such a box needs to be ventilated for cooling? I can only look at how the OEM packs were set up and guess what the engineers were designing for regarding cell compression, airflow etc? Any ideas on these issues?
 
It's hard to guess how much ride time might be available once this bike is done. I need more examples of power consumption during trailriding conditions. One theoretical comparison is the Alta Redshift bike at 260# of machine. Its battery is 5.8 kWH energy capacity.

Alta claims that you can ride 4 hours singletrack per charge meaning that you are using 1.45 kWh per hour riding
(1.45kWh/h = 1.9 hp constant output during your ride, I think. Way less than I would guess! maybe my math is fuzzy).

Here is my battery:

24Ah x 20 cells = 480Ah

480Ah x 3.8V = 1.8kWh

1.8 / 1.45 = 1.25 hours singletrack riding time. So far it looks ok on paper....
 
My project hasn't seen any progress because I bought a pair of Zero X bikes which are very similar to the path I was heading down with the KDX. This has made me rethink how I want to proceed as I don't nee a third bike with the same weaknesses found in the Zeros.

The Zeros are great for just tooling around on but performance anywhere near a standstill is deficient. i need throttle response and scalability from 0 to 5 mph that a brushed motor, Alltrax controller and Magura throttle don't provide.

Is a brushless motor combined with a Sevcon going to be exponentially better in this area? Any other combinations worth considering?

i am sort of leaning towards incorporating a clutch into the drivetrain somehow if that is what will be require to give me more trials-bike performance.
 
Have you been able to do any tuning of the controller. The zeros are de-tuned from the factory down low to preserve reliability. Even an underpowered electric drive system should be able to tear a bike in half or flip you over. You may want to look into unlocking and tuning the controllers for a steeper throttle curve.

As for clutch, its not like a gas motor where you get into the power by allowing the motor to spin. There could be a slight advantage to having a bit of rotating mass you can grab hold of with the clutch but it will likely not be the feel you are looking for unless you had a weighted flywheel which has its own down sides. It can and will be debated to death here but by the math says its not going to have an advantage over a properly sized drive train.

I had a brushed electric conversion car once with the factory 5 speed and clutch. Shifting made absolutely no difference in the power. It was a really strange feeling.
 
I will try and do some tuning on my zeros but only one has a programmable Alltrax. I'll see If i can tune the Main Bike Board on the other.

Scenario 1 : rider stalled on a very steep incline. The bikes are unable to move because the load will overload the system. Either the Alltrax or the MBB prevents sufficient torque at 0 motor speed. A clutch would allow a motor to spin up above stall speed with a load progressively applied to get a bike moving on steep incline from a standstill with 100 percent traction.

If you look at Lukes offroad videos from when these zeros came out, you see them unable to recover from failed hillclimbs as they had no torque or excessive wheelspin depending on how much traction was forced upon the rear tire.

I think this was a problem even with the vaunted Alta bikes at Erzberg Rodeo. The bike wouldn't climb out of a dead stall or if they did, there wasnt enough throttle control precision to prevent loss of traction (wheelspin) as the motor overcame stall inertia. They tried to tune it with a heavier rotating mass in the countershaft but a clutch would have been the proper solution.

A half turn throttle just doesn't have the input resolution to control all scenarios from 0 mph to topspeed mph on these dirtbikes. I would be happy to be wrong with my conclusions and would love to hear about a good motor controller combination addressing this.
 
This is basically what i'm after as far as the clutch/throttle control I want. If there is a motor/controller/throttle combination out there that will do this, please advise :D

<iframe width="500" height="300" src="https://www.youtube.com/embed/Osk9Jqs6hyg" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
 
Got it. I think you are correct that you will need a clutch of some sort to achieve that level of control. When most people start talking about clutches on here they are usually thinking of them in terms of rev up and dump like a gas bike.

Electric motor torque is amazing but zero rpm is still zero and I can see where it would be hard to overcome on a vehicle that spends so much time right at the threshold like a trials bike.

It's possible a torque limiting clutch could be setup to help with this but then that max torque limit is always there.

Perhaps a high end centrifugal kart racing clutch would be worth looking into. They are highly adjustable but I think it would need to be customized to operate properly at the crazy low RPM you are talking about. Sounds like you literally just need to overcome the absolute stall point at just a few rpm. These cart clutches use swinging centrifugal levers against a spring to actuate but if you made the spring light enough for the low rpm you need I would think it would be subject to vibrations and act really unpredictable. At $400-500 for one of these clutches its probably just easier to gut a motorcycle transmission and keep only standard clutch.

https://www.buller.net/Category/clutches
 
Don't know much about gokarts but that clutch look like the primary on a snowmobile or even sxs's? Wouldn't that need to be couple'd to a secondary clutch as well with a belt between em? As I said don't know nothing about gokarts but on snowmobiles we have a primary with arms and weights that shifts as motor rpm increases and work spring to grip the belt.
 
There are jackshaft mountable versions and motor mountable versions. I believe the idea is that the ice needs to be able to idle but engage as soon as the revs increase. I think these fancy ones are very adjustable and have multi disks to hold high powered 2 stroke motors compared to the small non adjustable versions on cheap 3-5hp lawnmower style engines. If this could be tuned to engage at super low rpm like 10-50rpm then it would perhaps be a hands free solution to the zero rpm stall issue on TB's bike. I would think having a few rpm such that the controller hall signal was nice and happy would be a good thing for precise control. The slight progressive engagement of the clutch would be nice too to keep the phase amp spike reasonable compared to load at totally zero rpm.

I'm brainstorming a set and forget solution because it seems like a shame to have a nice electric conversion and then have to introduce a clutch lever. Simplicity is a huge benefit to electric and the lever controlled variable clutch is not simple IMHO.

In defense of a moto clutch, they do come free with every motorcycle.
 
Its possible that the centrifugal clutch would be effective. An ideal solution for me would be a speedway bike clutch but they are very expensive and maybe overkill for a 30 hp machine.

https://www.youtube.com/watch?v=Ys0hSCrPnTc

https://litespeedbikes.mybigcommerce.com/72v-30ah-20s10p-surron-rebuild-drop-in/

https://www.amainhobbies.com/rcproplus-prod6-supra-x-battery-connector-4-sets-810awg-rppreb6808prod6p8/p298526

https://evtrader.com/l/plant-zonebuilding-bhehetian-sanhe-industry-zone-hengkeng-liaobu-town-dongguan-city-guangdong-province-dongguan-knaptech-energy-technology-co-ltd/

https://www.aliexpress.com/item/4000299065244.html?spm=a2g0o.productlist.0.0.4b5434d6sU0x38&algo_pvid=88052d2b-761f-4911-bad6-aeaf564af7ef&algo_expid=88052d2b-761f-4911-bad6-aeaf564af7ef-0&btsid=63929607-9875-4bf7-bc6d-692430145a09&ws_ab_test=searchweb0_0,searchweb201602_3,searchweb201603_53

https://ebikezilla.com/battery-engine/5009-battery-pack-litespeed-ultimate-72v.html
 
Many owners of dirt bikes are running those non-stall clutches from ie rekluse. If RPM drops too low, the clutch is automatically disengaged. So if you ride tight trails and drop your bike it will not stall. Never owned those kits but I've tested one a few years back and it was actually great for technical riding. Maybe something like that could work for your setup?
 
2 years later update: no progress! (That's not entirely true as now I have a bunch of spare parts to work with.) I ended up using those panasonic prismatics in one of my zero motorcycles and they work great.

In the meantime I have bought and maxxed out a light bee and found an Alta. The KDX had been rolled outside to rot. Now its back in the shop and I'm interested in it again. I'm still wondering if i can build it to weigh way less than the Alta while being more planted than the Light Bee. I haven't given up on the clutch idea yet either.

The extra parts I have now are a Mobipus 200A controller, a Nucular 24f controller and a rolling CRF150r bike including the gearbox/clutch. If I have room, Id like to use the gearbox somehow. If that is impossible, a jackshaft would be nice for countersprocket positioning.

For battery I'm going to use some 50Ah nmc prismatics of the same dimensions as those green panasonics. It looks like the perimeter framed KDX will allow two blocks of 12 cells. The question now is whether to run 24S1P vs 12S2P. I'd have to go with Sintech controller or something other than the spares I have if I go 24s. I think all 3 of those controllers max out at the same wattage, 20,000.

I guess what controller I use will depend on what motor I come up with as well. QS 138 3000 watt midmotor looks like it will fit in addition to the crf gearbox. That same motor with its integral jackshaft is very interesting but I don't know if its available yet. A zero 75-5 would fit without the gearbox, not sure about a jackshaft.20201221_072717.jpg
 
https://endless-sphere.com/forums/viewtopic.php?f=10&t=108178&start=50

This guy is building a light trail/trials bike and bought the qs 3k motor with the integral gear reduction. I think he said it was around $300. So i think it is available.
 
Thanks! In learning more about the qs motors I'm leaning towards the 4000 Watt version +/- the votol 150 controller. The votol 200 doesn't generate any positive info in my research.

The integral gear reduction looks interesting but is obviously untested still. Radio silence so far. Not sure if 2.5:1 reduction is ideal or enough.

I've never done aliexpress so I'm still mentally processing $308 for the motor and then $350 for shipping DHL :shock:
 
Yeah. The shipping thing is always crazy. Given our current trade relationship with China and the Pandemic I think its the worst possible scenario right now. Hopefully that price goes back to what It was.

If that gear reduction is halfway decent in terms of reliability then I think its absolutely perfect for what most of us are trying to do with these 20kw ish light moto builds. For years every direct drive motor option from 20-50kw have all required some insane large custom rear sprocket and too small of a front sprocket. This presents durability and noise issues as well as ground clearance and adverse effects on suspension. Getting the reduction at the motor output within the range of off the shelf moto drive train parts is awesome as far as I'm concerned.

The durability remains to be seen. This motor is called 3000watts. Generally with electric motors you can put a temp sensor in them and then totally ignore the ratings as long as you are careful. Obviously mechanical gears don't allow you the same freedom to push the limits. If you blow up a unique small production qty chinese gearbox with $300 in shipping attached to it you can't just buy another one at the salvage yard.

If you are wanting a clutch then obviously the simplest motor with the easiest output shaft to work with is the way to go. At that point you are making your own primary reduction gearbox thingy yourself anyway.

I'm just glad to see that someone on here at least ordered one of those gearboxes so we know if its vaporware.
 
Back
Top