DRZ 400 SM - First time build seeking advice

[frostnova37]

Hello,

I have been wanting to start a conversion project for a while now and recently managed to pick up a suitable donor at a decent price.




The previous owner had the top end blow out mid 2019, he put the registration on hold with plans to rebuild the engine. A few years passed and decided to move it on.



All cleaned up and ready for the fun part.



New Zealand has some regulations around EV conversions, one of which is that the final weight cannot exceed the 'manufacturers gross vehicle mass' without recertification becoming a headache. Out of the factory a (wet) DRZ 400 SM weighs 146 kg / 321 lbs, stripped back my 2016 model weighs 82 kg / 185 lbs. There is still some weight loss to go once I take a grinder to it but at this stage its looking like a 64kg / 141 lb allowance for the new electric drivetrain.

The intent/purpose of this build is to have an electric motorcycle suitable for low distance, daily on road use. These are my existing and ideal parameters.

Conditions:
-Low to no wind.
-Sealed roads.
-Daily use would be ≤10km / 6.2 miles at a stretch, I live very close to work so range is not a driving factor.
-Overall flat terrain, no notable inclines.
-Around town riding so a moderate amount of stop/starts.

Vehicle:
-Bike and rider combined will be ≤220 kg / 485 lbs (Assuming the weight is kept close to the manufacturers GVM).
-Capable of 115-120km/h with majority of the riding being done between 50-70km/h.
-17 inch wheel (motor will dictate gear ratio).
-12v system for lights/signals.
-Budget is flexible but living in NZ means that accessing parts takes longer and is more expensive.


At this point I'm looking at 3 potential candidates for motors:

QS 138 70h v3 (w/ reducer) - 3000W
-Easiest bolt in solution.
-Marginally more expensive than the 90h.
-Very popular, lots of community support/advice.
-Lowest footprint.

QS 138 90h - 4000W
-Requires 2 stage reduction (personal preference).
-Cheapest off of the shelf.
-Seems like the safe middle of the road pick.

QS 180 90h - 8000W
-Also requires 2 stage reduction.
-Nearly twice as heavy as the others.
-Twice as expensive.
-Not limited to 72v.

Majority of the builds here that use one of these three motors are dirtbike conversions and have what I would call 'limited' traction by comparison. Am I over egging it by considering a 8000W mid drive? The way I see it I would be pushing the 3000W to its upper limit to meet my needs but my opinion is far from informed. I am also interested in seeing some other before and after conversion weights to see where I stand.

I will be following up with a build thread once I have the motor, controller and battery locked in.

 

[larsb]

I’ve commuted for some years on the qs 138 70h and it seems your goals aren’t that different from my riding.
I’d go with the QS 138 90h with single stage gearing. It’ll give you more room and weight left for batteries and still enough performance.

I don’t think two-stage reduction is needed, why do you?
Why do you think the motor is limited to 72V?

 

[frostnova37]

I don’t think two-stage reduction is needed, why do you?
Why do you think the motor is limited to 72V?

The main reason for wanting a two stage is being able to change/play around with ratio's at will. Its more work setting up a jackshaft but I think I would rather have the ability to use 'standard' sprockets. Though this isn't a hill I'm willing to die on. I am also considering designing a reducer not too dissimilar to the one on the 138 70h to mount on the 90h.

I found examples of people running them at ~80-90V I'm interested to know how much higher you can go without repercussions.

 

[amberwolf]

The main reason for wanting a two stage is being able to change/play around with ratio's at will.

Couldn't you just change the rear sprocket, or the motor sprocket? Or is there a chain clearance problem somewhere by doing so?


The motors can probably handle over 200v easily enough. The current flow thru them, and watts at a particular speed (such as high watts at low speed) are where you would need limit things.

Most motors are rated at a particular voltage for a particular speed; this would be much better stated as a kV value, or RPM/Volt, but most places don't do it that way (they should).

 

[frostnova37]

Couldn't you just change the rear sprocket, or the motor sprocket? Or is there a chain clearance problem somewhere by doing so?

I need to get CAD models of the motors and chuck them in a frame to see if there are clearance issues. What are my odds getting them from QS? I have a feeling drawing them from scratch might be faster.

 

[amberwolf]

Pretty sure they have drawings of some of them; you could print those out and use cardboard-aided design with your actual frame.

 

[larsb]

I’ve seen cad of these here and i think j bjork even got the 180 motor 3d printed. Ask QS.

Check my rieju thread for reductions - I could fit the 138 motor sprocket in the same place as original motor sprocket on my moped frame. You’d have a smaller rear sprocket than me since you want higher top speed, i had 110kph top speed with the 22/105 geared belt drive at 20s.

It should work to go to at least 120V and single stage, something like 72/12 tooth :thumb:
I’ve seen QS state a 7500rpm max for the 138 motors, could be the max for rotor integrity reasons but better ask them about that too.

 

[frostnova37]

Pretty sure they have drawings of some of them; you could print those out and use cardboard-aided design with your actual frame.

Does this count as cardboard aided? :wink:

Check my rieju thread for reductions - I could fit the 138 motor sprocket in the same place as original motor sprocket on my moped frame. You’d have a smaller rear sprocket than me since you want higher top speed, i had 110kph top speed with the 22/105 geared belt drive at 20s.

I will have a play around over the weekend and come back with my findings.

 

[larsb]

Does this count as cardboard aided? :wink:

Yes, that would be one of the higher levels of CAD i’ve seen here

 

[frostnova37]

Here we go. For easier viewing open the images in new taps and swap between them.

First up is the 3000W V3. Easy placement as expected, lots of available frame to weld to. Stock sprocket position is maintained.


The 4000W is roughly the same outer diameter but the centered output means the front sprocket is pulled forward from the OEM position. For the sake of this exercise the sprocket setup is 10T/60T.

This is it sunken slightly further.


Lastly the 8000W. Noticeably bigger but still fits within the width of the frame. Same deal as the 4000W but the shaft is even further forward, again with the 10T/60T.

Same thing dropped down.


After trying a bunch of motor locations I came up with the idea of mounting it to the swingarm in front of the main pivot, giving a fixed distance between sprockets (pseudo coaxial?). Meaning the motor would oscillate inside of the frame but because of its proximity to the pivot it doesn't actually require much space to move. I don't know what influence it would have on the suspension if any but there's only one way to find out Not sure how well I've explained this as a concept but I'm working on a draft now.

Failing that I will mount it to the frame like a normal person.

 

[larsb]

Looking good!

My 2 cents on these pics:
Swing mounted motor is nice for a belt drive since it solves the required constant belt tension. I did a trial with my moped and compressed the suspension to the max with straps and found the swing envelope of the motor takes up a lot of space - I don’t think it would work really since to fit the compressed suspension location the uncompressed position of the motor would be so high that bottom side of the belt/chain might rub against the pivot. For a chain drive i would say there are only small benefits but a lot of added complexity.

10t front sprockets are not so easy to find for these motors, they create a lot of noise and wear chain so to find the boundaries of the build a 12t or larger could be better to design around.

 

[j bjork]

Very nice drawings

I think a single stage reduction with chain and sprockets will be too noisy for a street bike.
Well, a two stage with the same construction seems to be just as noisy though..

A construction with gears as the first stage might be better.

 

[larsb]

Belt drive is a dream for both noise and the user
completely silent and no chain oil/wax getting flung everywhere. Not so good for really poor roads, the belt can be sensitive to small rocks getting sucked into the belt/sprocket

 

[TheSammich18]

What CAD software are you using for these drawings and how did you get such a high fidelity model of the donor bike? Nice work.

What are you planning on using for your battery and have you determined its specs yet?

I can't imagine you gain much by attaching the motor to the swingarm, so I would personally rigid mount it to the frame. You'd be adding to your unsprung mass which will only hurt the handling characteristics while adding complexity. It would certainly be unique though!

 

[SlowCo]

Obviously go for the QS180 90H, big Fardriver controller and a strong battery pack for most fun
Good luck and enjoy building it!

 

[frostnova37]

What CAD software are you using for these drawings and how did you get such a high fidelity model of the donor bike? Nice work.

The photos are from Rhino5 and the model is a mesh made from a 3D scan (not mine) of a DRZ 400 E, I just added the motard specific items and tweaked a couple of things.

What are you planning on using for your battery and have you determined its specs yet?

This is up in the air until I pick the motor/driver. Likely going to be 72-96V ~50AH. I will be building the pack myself to make the most of the available space.

 

[larsb]

Belt drive is a dream for both noise and the user
completely silent and no chain oil/wax getting flung everywhere. Not so good for really poor roads, the belt can be sensitive to small rocks getting sucked into the belt/sprocket

On request from the OP, i think it’s best to share with everyone:
The spec of the QS belt pulley on the 138 motor is HTD-8m-20
The original c-c length you will get from the CAD and then you need to find a belt size that’s close and then move your motor to fit the c-c distance of that belt (with margin for proper tensioning)

I’ve used continental belt calculator for the belt selection and determining the c-c distance:
https://www.continental-industry.com/en/solutions/power-transmission/industrial-applications/drive-belts/service/products/service-tools-catalogues/conti-professional

 

[frostnova37]

Tread revival! A lot of real life things came up which benched the project for a while but I'm back, and with new toys



I have been supplied CAD files for the new QS138 90H with gear reducer (under NDA so I cant share). More updates soon but I will likely make a proper build tread for documentation moving forward.

 

[halg]

I am doing a conversion on a KLX400 which is a kawasaki version of the DRZ400. I got the motor and controller (QS138 70h & EM260) this week and am going to start the tear down and mock up soon.

Mainly, I'm interested if you were able to design any motor and battery mounts and what size battery you used.

Did you complete this build? Would you be willing to share the motor and battery mount design you chose?

Thanks.