Kawasaki KXF 250 QS 138 90H SEVCON conversion

General Discussion about large electric scooters and motorcycles and other things with no pedals.
User avatar
ZALO   1 mW

1 mW
Posts: 16
Joined: Dec 25 2019 7:27am
Contact:

Kawasaki KXF 250 QS 138 90H SEVCON conversion

Post by ZALO » Feb 01 2020 6:12pm

I'm working on a kawasaki kxf 250 cross motorcycle, I decided on the qs 138 70H motor.
I plan to use the sevcon gen 4 size 4 controller. I have Sony VTC5A lithium-ion cells at my disposal.
What package configuration is best to do?
How many cells would be optimal?
Attachments
ex.jpg
ex.jpg (250.07 KiB) Viewed 5165 times
ex2.jpg
ex2.jpg (203.81 KiB) Viewed 5165 times
ex3.jpg
ex3.jpg (137.75 KiB) Viewed 5165 times
Last edited by ZALO on Oct 28 2020 9:37am, edited 1 time in total.

User avatar
amberwolf   100 GW

100 GW
Posts: 31756
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by amberwolf » Feb 02 2020 1:54am

ZALO wrote:
Feb 01 2020 6:12pm
I'm working on a kawasaki kxf 250 cross motorcycle, I decided on the qs 138 70H motor.
I plan to use the sevcon gen 4 size 4 controller. I have Sony VTC5A lithium-ion cells at my disposal.
What package configuration is best to do?
How many cells would be optimal?
what current do you need, peak?

what current do you need, continous?

how much range do you need? (which includes questions like, what wh/mile do you expect to have, which includes: what is your terrain, wind, weigth, speed, etc).

the above leads to: how many wh do you need?

you'll need to answer those questions before you can get a *minimum* number of cells requirement.

then you'll want to decide if you want to run the cells hard, or be easy on them. running htem hard will mean running them at their max abilities, so using only the absolute minimum number of cells. running them easy means using more than the minimum; the greater the increase, the easier it is on them.

it's possible that just to get the range you want you'll already have much more than minimum...but you'll have to work that out first to find out.

User avatar
ZALO   1 mW

1 mW
Posts: 16
Joined: Dec 25 2019 7:27am
Contact:

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by ZALO » Feb 02 2020 8:04am

Gen4_Size4_80v_Info-500x400.png
Gen4_Size4_80v_Info-500x400.png (141.48 KiB) Viewed 5111 times
Specification Size 4:
Nominal Supply Voltage - 80V
Maximum Supply Voltage - 116V
Peak Current - 350A
Boost Current - 420A
Continuous Current - 140A

Specification QS 138:
Rated Power: 3000W
Rated Voltage: 72V as default
No-load rpm: 4400rpm
Max Torque: approx 56N.m
Peak Battery Current: 120A (72V)
Suggest Peak Phase Current: 450A

Specification Sony VTC5A:
Max continuous discharge current: 35A (with temperature cutoff circuitry at 80 degrees C)
Max continous discharge current without temperature cut: 30A
Weight: 47.1g +/-1.5g
Nominal capacity: 2600mAh

I would like to over voltage the motor a little and supply it with 88.8V (24s) to get a higher speed.
Will working with this voltage be suitable for such a motor ?
I intend to take no more than 140A from the battery, so I plan to 16p in pack.
Summary, I did a preliminary try on the 24s16p package. The weight of the entire battery should not exceed 20kg.
I care about the low weight of the whole project, around 100kg.
I hope that 3.7 kWh energy is enough for about an hour of off-road fun.

User avatar
macribs   1.21 GW

1.21 GW
Posts: 3676
Joined: Jul 22 2014 5:59pm

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by macribs » Feb 08 2020 8:25am

You converting a dirt bike. For off road use I assume? I would forget about over volting. For the simple reason of more range and more possible amps.

If you stick to 20s20p or 20s22p your cell count will not be that much higher then 24s16p but you will have more ride time, you can drain much more peak current from the pack and for getting the most fun out of your build. I think maybe stay at 20s or even 22s, and put even more cells in parallel if you find the real estate.

Acceleration should be your priority. Whether you top out at 90 km/h, 100 km/h or 130 km/h it will likely not mean much to you off road, but faster acceleration, longer ride time/practice sessions and longer battery pack life will for sure put a big smile on your face.

User avatar
ZALO   1 mW

1 mW
Posts: 16
Joined: Dec 25 2019 7:27am
Contact:

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by ZALO » Feb 11 2020 3:29pm

After all, the range will not depend on the number of cells in the series, but on the total energy stored in the battery measured in kWh. I want to power the motor with a higher voltage to get a higher speed, because I use any gear anyway. I care about the greatest possible moment without overheating the motor, which would be inevitable at high phase current.

larsb   1 MW

1 MW
Posts: 1984
Joined: Dec 10 2014 5:12am
Location: Gothenburg, Sweden

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by larsb » Mar 24 2020 3:40pm

Looking to be on a nice track :D
You’ll need 500A phase, 200-250A battery current to fully utilise the motor. It will easily take 22 or 24S.
Vtc5a are capable of 30A, barely, so i’d choose at least 20S20P as it will ease the stress on the batteries and give an ok range - if you can fit them all?

User avatar
ZALO   1 mW

1 mW
Posts: 16
Joined: Dec 25 2019 7:27am
Contact:

ZALOMOTO EX350 - Kawasaki KXF 250 QS 138 70H conversion

Post by ZALO » Apr 13 2020 9:19am

I decided to use a battery with 24s16p configuration, this arrangement of cells will fit perfectly in the frame space and additionally welding of the packets will be regular. The energy stored in the battery should be enough for a moment of fun on the track, while the entire motorcycle will be light.

I will use BMS 300A LCD, I hope that work with maximum current will be possible
91262175_2994948210564998_6638611974538133504_n.jpg
91262175_2994948210564998_6638611974538133504_n.jpg (15.93 KiB) Viewed 4286 times
91009506_2994948157231670_2643202708810498048_n.jpg
91009506_2994948157231670_2643202708810498048_n.jpg (23.65 KiB) Viewed 4286 times
90801345_2994948113898341_2949219029866250240_n.jpg
90801345_2994948113898341_2949219029866250240_n.jpg (56.06 KiB) Viewed 4286 times
I have also designed the entire drive system to match the original engine mounts.
89288450_2960734593986360_5620731672010948608_n.jpg
89288450_2960734593986360_5620731672010948608_n.jpg (18.52 KiB) Viewed 4286 times
89512344_2960734633986356_662060472714919936_n.jpg
89512344_2960734633986356_662060472714919936_n.jpg (14.78 KiB) Viewed 4286 times
For now, I'm staying with the sevcon 350A, because it fits perfectly into the airbox space, if there is too little phase current, it will mount the sevcon 550A
89348292_2960734550653031_4765941379475963904_o.jpg
89348292_2960734550653031_4765941379475963904_o.jpg (74.54 KiB) Viewed 4286 times

User avatar
tylerwatts   100 W

100 W
Posts: 155
Joined: Sep 14 2012 3:10am

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by tylerwatts » Apr 14 2020 8:49am

Very nice design.

Cheers
Tyler


MorbidlyObeseKoala   1 W

1 W
Posts: 57
Joined: Apr 06 2018 3:53pm

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by MorbidlyObeseKoala » Apr 14 2020 10:56am

Clean looking. Maybe this was answered elsewhere, what made you decide for a belt driven reduction? Why not run the sprocket off the motor?

User avatar
ZALO   1 mW

1 mW
Posts: 16
Joined: Dec 25 2019 7:27am
Contact:

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by ZALO » Apr 14 2020 3:34pm

Thanks, I wanted the drive to be as quiet as possible, the belt transmission would be properly built-in.
This is where the highest speed occurs and the chain would generate unbearable noise.
I hope that I will not have to change this belt often.

User avatar
macribs   1.21 GW

1.21 GW
Posts: 3676
Joined: Jul 22 2014 5:59pm

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by macribs » Apr 16 2020 1:54am

I like that battery and motor box. This will be a cool bike.

For side by side with power upgrades we have used temp sensor inside the cvt cabinet to prolong belt life. When temp spikes ease off the throttle until temp falls. Works wonder.

If you decide to enclose the belt consider mounting a temp sensor so that you can monitor belt temperature if you experience increased belt wear.

Daversj   1 µW

1 µW
Posts: 3
Joined: Apr 16 2020 12:53pm

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by Daversj » Apr 16 2020 3:05pm

First post!

I have been trying to figure out this same conversion...

Some issues to consider:

Your QS motor makes 4400 rpm @72v with no load. Under load it will only make 75% of those rpm. So 3300 rpm under load.

3300 rpm with a final chain drive of 3:1or 42,14 teeth, will give a theoretical top speed of 75mph WITHOUT your belt reduction. With your belt reduction you are looking at about 25mph.

If you use a flux weakening controller it will help a little but not sure if its enough. They claim 6500 rpm....so 4875rpm under load. This translates to 111 mph theoretical and gives some room to reduce gearing there. But i doubt you will see 4875rpm, maybe 4k under load. These motors lose torque near the top. So maybe a reduction belt 25% at most. And it will be a pig off the line.

A CVT as an intermediate might work better. It would let you gear for speed and help off the line.

The QS motor you chose has 50nm of torque. That is about the same as a 450 4 stroke gas engine at peak. But the gas motor has 5 gears to help get it going and keep top speed potential. You only have 1 gear so you have to pick one....4th gear in the stock tranny is 1:1.

If you could output that motor through the stock transmission it would be a beast. More power from start than a 450f.

Or find a motor that has higher rpms. The Alta redshift motor goes to 14k and is gear reduced. Where can we source that baby?

So I don’t mean to throw a wrench in the works but i am trying to figure out the same problem. I haven't been able to find a suitable motor that has enough torque and RPM to work direct drive. Unless i pick something huge that weighs 60 lbs. That wont work for a dirt bike or supermoto. And a reduction of a smaller motor will make it slow. So really we need either a higher rpm motor for reduction or a transmission. A large diameter out runner would work but they don't really exist other than hub motors. This is why the Vector Vortex mounted a hub motor mid drive.

If you try to get rpm by increasing voltage to 100 i don't know if it will get you there. Will the motor handle it? Maybe? Your battery will start to get huge with any decent AH capacity at those voltages as well.

So when you figure it out let me know. I am still stuck.😬
Will be following your progress.

larsb   1 MW

1 MW
Posts: 1984
Joined: Dec 10 2014 5:12am
Location: Gothenburg, Sweden

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by larsb » Apr 16 2020 3:26pm

Loaded rpm is more like 85% in my experience.
max torque for this motor was mentioned close to 70Nm in Vasilisk testing of nuc controller
There was a youtube video linked here at ES with this motor at >100volt, think top speed was something like 140kph
..anecdotal but i think it’s possible to push this motor further than the votol em-150 controller allows.
I’d choose a 30S 600A phase controller if max output is wanted :D

If you need more torque and speed than this, me1302 or me1507 motors aren’t that bulky but certainly in a different price range.

Daversj   1 µW

1 µW
Posts: 3
Joined: Apr 16 2020 12:53pm

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by Daversj » Apr 17 2020 12:11am

70 Nm is a lot of torque but sounds improbable. Thats like 50ft lbs. I had an Italian 1100cc bike with 80lbs at the shaft and it was a hoot.

Found a vid of a guy using this motor in a ktm. Looks like he is 22 cells with 8:1 gear reduction. Not sure of his chain ratio. Seems underpowered but maybe his controller is the choke point?

https://youtu.be/3l8ZKQOKlNE

larsb   1 MW

1 MW
Posts: 1984
Joined: Dec 10 2014 5:12am
Location: Gothenburg, Sweden

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by larsb » Apr 17 2020 12:43am

Yes, controller chokes, it’s in the votol controller thread.
Also a six gear 30kW ICE engine powerplant will never be matched with a one gear 15kW electric, right?

Sport bike ICE engines lack torque in comparison to electrics but they have it in high rpms, that’s why the redline needs to be at 13K rpm to get decent power out of them :wink:
the QS 138 70h would be a 100hp motor if it could have the max torque at 13K rpm.

User avatar
ZALO   1 mW

1 mW
Posts: 16
Joined: Dec 25 2019 7:27am
Contact:

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by ZALO » Sep 26 2020 1:59pm

Ok, the project is finished and it fully met my expectations.
Finally put the QS 138 90H motor.
A few photos of the final effect:
Zalo Cross-12.jpg
Zalo Cross-12.jpg (305.92 KiB) Viewed 3276 times
Zalo Cross-24.jpg
Zalo Cross-24.jpg (218.05 KiB) Viewed 3276 times
Zalo Cross-33.jpg
Zalo Cross-33.jpg (168.07 KiB) Viewed 3276 times
Zalo Cross-42.jpg
Zalo Cross-42.jpg (181.24 KiB) Viewed 3276 times

User avatar
gogo   10 MW

10 MW
Posts: 2482
Joined: Mar 17 2008 2:20pm
Location: Iowa USA

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by gogo » Sep 26 2020 2:56pm

Beautiful. Looks like you retained the stock motor sprocket location and replicated the stiffness of the original gearbox to maintain original chassis engineering. Seems like the CG is probably the same, too.

I wonder if wax would dampen the chain sound without being messy?
"A conclusion is the place where you got tired of thinking." -Steven Wright

danli_new   1 µW

1 µW
Posts: 2
Joined: Sep 27 2020 7:28pm

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by danli_new » Sep 27 2020 7:55pm

Great conversion and final product looks so good. It reminds me of sur-ron storm been that suppose to come out soon later this year. I was just thinking about buying a mid drive motor kit (that including 138 4000W PMSM IPM mid drive motor with sprocket, controller, display, and throttle and light switch) from QS to try to convert a dirt bike to electric.

I'm very new to this field. I only had limited experience upgrading electric bike from lower motor/controller/battery to higher ones. I know there's gonna be challenges when it come to mounting the controller, the mid drive motor and battery, since I don't have any experiences in welding. I wonder if my plan will be a little more than I can handle.

j bjork   10 kW

10 kW
Posts: 903
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by j bjork » Sep 28 2020 3:52am

Very nice, but what happened to the build thread?
There was planning, and finished.

How is the belt drive holding up?
Are you still using the size 4 sevcon?
Did it make a difference to change to a 90h motor when you are controller limited?

SlowCo   1 MW

1 MW
Posts: 1654
Joined: Jan 05 2015 5:43pm
Location: The Netherlands

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by SlowCo » Sep 28 2020 2:07pm

What a beautiful build :thumb:
Any numbers you can share? I.e. 1st and 2nd gear reduction, max. battery and phase amps, acceleration and top speed, range/riding time? And if possible a video 8)

Routybouty   100 W

100 W
Posts: 161
Joined: Jul 23 2013 6:44pm
Location: Fort Wayne, IN, USA

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by Routybouty » Sep 29 2020 7:35pm

Looks awesome! Videos please!!
Waiting for that perfect ebike.

User avatar
Chambers   100 W

100 W
Posts: 237
Joined: Nov 11 2011 8:01pm
Location: Brisbane

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by Chambers » Oct 23 2020 2:31am

Really cool!

User avatar
Chambers   100 W

100 W
Posts: 237
Joined: Nov 11 2011 8:01pm
Location: Brisbane

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by Chambers » Oct 23 2020 2:36am

This looks like it? [youtube]8-S9Z1yHvZg[/youtube]

larsb   1 MW

1 MW
Posts: 1984
Joined: Dec 10 2014 5:12am
Location: Gothenburg, Sweden

Re: Kawasaki KXF 250 QS 138 70H conversion

Post by larsb » Oct 23 2020 9:14am

Looking good! What gearing did you use in the end?

User avatar
ZALO   1 mW

1 mW
Posts: 16
Joined: Dec 25 2019 7:27am
Contact:

Re: Kawasaki KXF 250 QS 138 90H SEVCON conversion

Post by ZALO » Oct 28 2020 9:39am

Total 1:8,25

First competition!
[youtube]https://youtu.be/Z04VutFY3dk[/youtube]

Post Reply