1979 Honda Electric Dream (CB250N conversion)

[einkleburt]

Hello All,

I've been thinking about doing an electric vehicle conversion for a few years now and have now actually started!

The overall goals are to:

• Get road legal
  Speed - Not hold traffic up too much.
  Range - Initially 5-10 miles (50 would be nice but batteries are expensive!)

So a week ago when I saw a Honda Super Dream for sale for not a lot of money, i decided to buy it.
It was listed as "running but in need of repair soon" condition. (advisories for: exhaust noise, exhaust blowing, front and rear sprockets worn, chain worn)

It didn't make it home as the head gasket blew after 10km causing it to lose oil quite rapidly! This pretty much guaranteed that it would become an electric bike.

The first plan was to find out how much space is available and work from there.
I've now started stripping it down to see what I will be able to fit in the place of the engine

Having pulled the engine and associated bits out, the bike is MUCH lighter (the engine alone was ~60kg)



Battery
I think I should easily beable to fit 10x 4s 16Ah Multistar Li-po under the tank.
Arranged as 20s2p for 74V 32Ah giving just over 2kWh capacity
They are rated as 20C peak 10C continuous, hopefully I will be able to draw 10kW without too many problems (which equates to 5C discharge) I could always add some extra packs if it is not sufficient.

Motor
I am struggling a lot, there seem to be very few motors in the UK.
Current options that look possible:

• Used AGNI 95R, the problem being that i'm still not confident about the condition of the used motors that i've seen (either a mildly bent shaft or a grooved commutator and brown from running too hot)
  A Motenergy ME1003 imported from the USA
  Dragging the bike round with a MXUS3000 bolted into the forks!?! (Probably not powerful enough to be safe on the road but should get it rolling or could supplement another motor)

Controller
Dependent on the motor chosen

Frame
I will need to add some kind of frame (probably intergrated with the motormount) to the bike as the engine was a load bearing part of the frame

This may well be a long term project (depending on how long it takes to find a motor!)

 

[jonescg]

Looks like a great project and there are plenty of 10-15 kW motor options which would give the same kind of performance as the original motor. DC will be the cheapest, easiest way to go, but AC is always nicer. Motenergy ME13 series are a good option for AC and Kelly controllers are fairly cheap for the job they do. I think there's a DC motor option with the same dimensions as the ME13 too.

Also, it's in pretty good condition, and the CB250 was a rather handsome bike If you can, try and make it look as well integrated as possible. I've seen a few bikes with blue CALB cells poking out from underneath the frame and it just looks trashy. If you can get that bit right your bike should deliver the goods and get plenty of looks

Chris

 

[gogo]

Nice styling on that bike and I always liked those wheels. Because that heavy engine was a structural frame element, I wouldn't use that chassis. Its unlikely that anything shy of putting the engine case back in place will replicate its structural function.

 

[SlowCo]

Or exchange the rear wheel for a 17"hubmotor wheel from QS motor:
http://www.cnqsmotor.com/en/article_read_248.html

That can take up to 16kW peaks and goes up to 70 mph and will not take any space in the frame leaving enough room for batteries and the controller under the tank. Vito Ho of QS motor is a member on ES here:
https://endless-sphere.com/forums/viewtopic.php?t=65972
He will be able to help you with more info.

Good luck and keep us updated.

 

[Noq]

+1 for QSMotors. Currently getting very high speeds (100-120kmph) with an 8kW motor at 84V. You can go faster with higher voltage and diameter. Their build quality is high, and service is excellent.

 

[einkleburt]

I am tempted by the QS hub motor but it would mean that I lose the comstar wheel (which lots of people like the look of) and also means that I would need to fabricate some kind of disk brake mount for the swing arm, which i'm not keen on doing

I'm struggling to find out how much it costs in import duties and tax to buy from china to the UK.
So far i've got:
$658 motor
$323 shipping
= $1177 with 20% VAT
+ Unknown import duties

Anybody got any comments on putting a MXUS3000 (20/16mm axle) in the front forks? (The MXUS is on order anyway)
If I have two motors on the bike (front and rear hub motors) am I likely to need two 'engine numbers' for the registration document?

Structure wise
I think i can manage to make something that is structurally sound to replace the engine, I've got some experience with designing structural elements, the difficult bit is estimating the amount of load which the new frame will need to take.

Appearance

try and make it look as well integrated as possible

I'd like it to look like it's "custom" and not "home made" or "hacked together" so there will be a reasonable amount of thought put into the bike's cosmetics.

Also, it's in pretty good condition

Looks can be deceiving, It turns out that the previous owner was good at covering things up and making it look better than it is. (not all that surprising)
There are some areas near the rear light that should have been welded up and were insteadbuilt up with silicone sealant...
The orange paint was applied directly over the old paint WITHOUT sanding the old paint first, (The new paint is peeling off very easily with alcohol!)
The fork gaiters appear to have been fitted to cover up the lack of fork dust covers.
The down pipes were half full of exhaust putty... (as in INSIDE the pipe where the exhaust gasses should be flowing!)

Basically I've got plenty of restoration and un-bodging work to get on with prior to fitting a motor and batteries!

 

[einkleburt]

[strike]Currently at the top of the list of possible motor options is a Motenergy ME1012[/strike]
Edit: I just purchaced a Motenergy ME1012 (I found one in the UK, which was about £300/$400 less than importing one!! )

The ME1012 is mentioned very little on ES so I've been looking for the differences between it and the ME0913 (Which is much more popular and has more documentation)

So far I have found:
----------------------ME1012 -- ME0913
Rated Power: -----10kW ----- 12kW
Peak Power: ------24kW ----- 30kW
Rated Voltage: ---72V ------- 96V
Number of turns: -20T ------- 28T
Phase to Phase R: 0.065Ohm 0.013Ohm

Both motors have the same current rating. does the equal current rating mean that the ME1012 just has less copper? i.e. same winding cross section, just with less turns

If anyone has any comments on the ME1012 I'd be happy to hear them

Edit:
I think that this motor should hopefully give the bike a vaguely similar power to the 250cc dino juice burning lump that was in it, which apparently made 27hp when reved up to 10000 rpm. Based on the gearing choices i'm likely to make, I am expecting a lower top speed and more low speed torque

 

[einkleburt]

The motor has arrived and I am now working out where it will go and how the mounting plates will fit in.

A very rough mock-up is shown below: (Purely to see how it MIGHT look)

The current thinking is to use some fairly thick 10-15mm aluminium plates extending from the front engine mount to the two rear engine mounts. These will be stabilised with spacers and bolts running between them Red dashed lines. At the bottom of the mount the two plates will be joined with a third plate (red rectangle) to provide some rigidity to the structure. In the likely case that this is not rigid enough I will join across between the two main plates with a bent and welded aluminium sheet for the majority of the upper and lower sections.

The volume in front of the motor will either be used to house the controller or to house the cooling equipment (fans/air-ducting/filter or radiator).
The controller could alternatively fit vertically, above the front end of the swing arm, between the tyre and main frame.

Controller

Controller selection is down to either a
Trapezoidal Controller:
Kelly KHB 72701 (72V 700A Peak)

• 
  less expensive
  noisy
  powerful (would be limited by the motor)
  larger
  heavy

Sinewave Controller:
Kelly KLS7250D (72V/160A Peak 400A) or KLS7275D (72V/200A peak 500A )

• 
  more expensive
  less powerfull (possibly limited by the controller and the motor)
  Quiet
  Small ~1/2 the size
  3kg lighter

Edit: Tidied up layout of post

DC DC Converter?
12V 12A out, 43-110V in, More than enough for headlights/accessories (not sure how the efficiency compares)
http://www.mouser.co.uk/Search/ProductDetail.aspx?R=V72C12C150BLvirtualkey54930000virtualkey492-V72C12C150BL

 

[flippy]

looks like a nice setup.

i recommend getting the KLS controller. it's smaller and drives much nicer due to the "proper" sinus, the bump in efficiency is also nice.

what idea's do you have with the cells? if reliablillty and lifespan is key then pouches are out. lifepo has poor density and you need more to get the voltage you need. 18650 might be a bit more difficult to build but can yield the best performance and range and wear out the least if you stay within spec.

 

[einkleburt]

Drivetrain design time
This is in part written so people can see what i'm thinking (so if there are mistakes they don't have to make them in the future) and partially for my own future reference.

To calculate the power needed to move the bike I have been putting some parameters into the EBikes.ca simulator.
As the simulator doesn't accept adding motor data, it has only been used to find the power required to move the bike at different speeds.
Parameters used:
Weight - 250kg (the worst case including me and some shopping)
CdA - 0.7 (estimated from some online research / comparisons with other bikes)
Cr - 0.02 (This is almost 100% guess)

To maintain a steady speed of 47Mph I will need ~5.2kW
30Mph = 1.8kW,40Mph = 3.4kW 60Mph = 10kW, 80Mph = 24kW

So having finally found a graph of the ME1012's performance I can then compare the motor power to the required power (something you should do before buying a motor)

Picture found on the kelly controller website, lines enhanced as they were difficult to read, units converted because Lb/In

I'm not sure how to accurately translate the motor graphs for higher voltages so I am basing my calculations on the 48V data.
Based on the motor power output the bike could sustain a constant speed of 47Mph at 48V (but would never reach it without extra power for acceleration)
Fortunately the motor will take more current 420A (peak for 1 min) and I will be running ~72V so overall I should have lots of spare power to provide some acceleration.
Based on the the local roads 50Mph should be plenty!

Gearing
Based on the above calculations running at 47MPH with a motor speed of 3000RPM. This would translate to a gear ratio of ~4.77:1*
*(3000RPM = 50R/s, Back wheel is 2m circumference so in direct drive 100m/s = 224Mph, 224/47= 4.77, ratio 4.77:1)

This can be roughly achieved with fairly standard components: 15T taperbush sprocket , [strike](7/8" taperbush to fit motor shaft)[/strike] , Rear 70T sprocket giving a 4.66:1 reduction.
Edit: Motenergy motors use 3/16" keyways on their shafts, 7/8" taperbushes come standard with 1/4" keyways it seems

I would prefer a shorter gearing with a 90T rear sprocket (6:1 ratio) for an improved acceleration as the increase in voltage from 48-72V should allow me to approach the motor's max RPM (5000).
Unfortunately the largest aluminium sprocket blanks I’ve found are 70T for 428 chain. Initially I will live with a 70T rear unless anyone has any suggestions?

 

[jonescg]

Looks like you're headed in the right direction. Check out Ian Hoopers NSR conversion - http://forums.aeva.asn.au/ian-hoopers-honda-nsr150-conversion_topic3949.html

He used a fairly low gearing due to the current limitations of the motor, but by all accounts it goes well.

420 chain is perfect for the job.

 

[spata]

Oh this should be a fun conversion! Nice to see another one done in the UK. For your gearing have you thought about using a jackshaft? If you are eventually looking at 5000 rpm then that is quite a reduction for a single stage. It is a major pain to design in but it does give you much greater flexibility in terms of gearing albeit at a loss of efficiency. The other benefit is with chain tension. Have you measured the original geometry of the rear drive sprocket relative to the swinging arm pivot? You should try to keep to this geometry as closely as possible, using a jackshaft you should be able to perfectly match the original.

I would worry that the diameter of your motor would put your rear drive sprocket much further away than the original. This can in turn give issues with variable chain tension. Finally have you analysed the chain run from your motor sprocket to the 70T sprocket at all possible swinging arm position? You may find that you hit a frame member or the swinging arm pivot somewhere in the arc of travel. Take a look at my Honda MT50 moped conversion here I had very similar issues which I had to overcome. I also have a section in there on getting the bike through DVLA and MOT testing which hopefully will be of use to you.

I used this DC-DC converter (72V-12v 10A £4.67) I use 2 of them, one for the bike electronics, contactors, cooling etc and one for the normal motorcycle functions. I would not trust the 10A rating, but I run a 35w/35w headlight with no issues - and at that price you can always dedicate one just to the headlight! here is a picture.


- Spata

 

[einkleburt]

Transmission/gearing
I hadn't even considered a jackshaft!
I've got a 70T sprocket on order so I’ll see if/how well it fits with a single stage reduction before I look into the details for adding a jackshaft.

The axis of the front sprocket isn't too far out from the original position and the short travel of the rear suspension -65mm theoretical (in reality it's rather stiff and probably only manages 40mm) means I’m not too worried about the chain length at the moment. I may struggle with the chain catching the swingarm pivot though.

Registration
I've been following your conversion thread, reading about your success with the DVLA was reassuring
My bike is currently fully legal (MOT/Insurance/V5c'ed in my name) so hopefully it will be a fairly straightforward to change the registration to electric. I'm trying to keep the build modular so the motor/controller/frame will hopefully just bolt into the existing engine mounts with no frame modifications then I can legitimately say that the engine has just been replaced with a motor which should also help with the ease of getting approved.


DC-DC
I worry about the reliability of that kind of electronics... I've had things die from voltage spikes from cheap DC-DCs so I’m slightly sceptical of what's in that box especially as it has only 3 wires and the instructions mention not needing to connect ground??????!!!!
That said I could run it for just the lights, with everything else run on a low power but higher quality one like this

Previously I had been thinking about getting either a Vicor (150W £120) or Meanwell (100W £58) or Delta (300W £96)

I like the look of the vicor one as being potted there is less to worry about in terms of vibration and moisture.

 

[spata]

I've got a 70T sprocket on order

Did you use UK Bike parts for the sprocket?

I may struggle with the chain catching the swingarm pivot though

Hopefully with relatively limited suspension travel you will be ok and nothing will foul.

It kooks like the budget for your build will be more than mine and to be honest I would have liked to go for better DC-DC converters.
The ones I linked to are used in almost every Chinese made e-scooter, and I have not heard anyone having a problem with them yet in terms of reliability.
However they are only 3 wire, so there is no ground isolation (the input and output share a ground). This becomes more of a saftey issue as you move to higher voltages. In my case all my lights, horn, indicators, etc are 2 wire and none of them use chassis ground, I did this for a bit of additional safety.

- Spata

 

[einkleburt]

Did you use UK Bike parts for the sprocket?

Yes, ordered on the 26th. I'm not sure how long it takes them to make them - they start from blanks so I'm hoping to see it some time early next week.


My budget isn't really defined... but I figure that a good DC-DC converters will last and I can use it in anything else I build.
I think so far i'm on ~£800 for the bike, motor, sprockets and cabling. The other main purchaces are a controller (£500?) and batteries (£370 - 10x 4s 16Ah HK lipo - I'm not after amazing range!), I think i've got most of the other stuff already so hopefully can manage ~£2000 total. (not including any extra tools I may need)

Also I've still got the old engine and some other IC parts that I can hopefully sell.

 

[spata]

This can be roughly achieved with fairly standard components: 15T taperbush sprocket , (7/8" taperbush to fit motor shaft) , Rear 70T sprocket giving a 4.66:1 reduction.
Edit: Motenergy motors use 3/16" keyways on their shafts, 7/8" taperbushes come standard with 1/4" keyways it seems

Have a google for stepped (step or offset) Keystock. This is specifically designed for when the shaft and pulley have different keyways. They do 3/16" to 1/4". But no idea if you can get a single key, or of the depth will be suitable for you application - but might be worth a look. I found a supplier here


EDIT: Just found this I think the 1/2 pitch Type B Sprocket - 15 teeth, 7/8 inch bore should be what you want, it has a 3/16" keyway and I seem to remember that a #40 sprocket is compatible with 428 chain and sprockets, again worth a check

- Spata

 

[einkleburt]

Thanks Sparta, the stepped key is interesting, I'd not thought to look for that, sadly uncommon outside of the USA.
I've had a look at the taper bush specs: the bush without key can do 51 N.m
The motor with torque constant 0.12N.m/A would need 420A to get the sprocket to slip so for initial testing atleast I'm going to leave the key out.
_______________________________________________________________________________________________

The custom sprocket has arrived, Possibly bigger than I expected (physically) but still about 3x lighter than the original (undrilled steel)


_______________________________________________________________________________________________

Also now on order: Kelly KLS-D 72V 500A Sinwave controller, £5 DC-DC Converter
_______________________________________________________________________________________________
Throttle

As a side project while I've been waiting for parts, I've been looking into the possibility of building a throttle [strike]as the current quality options seem limited to Magura or Domino which are expensive[/strike] as i've clearly got too much time available.
Whether I continue trying to develop a throttle will depend on if other things turn up quickly or if i have lots more waiting to do!

 

[spata]

I am guessing the green string from the sprocket is to check for any contact with the frame, looks like it was ok? And I agree -they are amazingly light compared with the stock steel ones.

I like the idea for the throttle - did you see this thread? It is for a remote cable operated hall throttle, but you could but a linear pot in something similar. I did that on my previous bike, just so I could use the original throttle ... worked a charm. It would be even safer if you have a push pull arrangement on the original CB250N throttle.

- Spata

 

[einkleburt]

I am guessing the green string from the sprocket is to check for any contact with the frame, looks like it was ok?

It's actually pretty close...

based on typical levels of chain sag, It may* very occasionally touch the top of the swingarm pivot with the current positioning of the motor, to increase the clearance (move the motor up) I'd have to move it further froward which I'd rather not do.
The engine's sprocket to swingarm distance was 105mm, the motor's axis will be 120mm from the swing arm, hopefully this will not cause too much in the way of problems.

*Hard to tell without having the chain yet

If all else fails I can shave 10mm off of the frame and move the motor up and backwards and have it in exactly the same place as the stock front sprocket.

I've considered a POT/hall box, but it seems a little against one of the things I was aiming to do with the bike - a reduction of mechanical/oily bits, especially in areas that they're not needed.
If the designing a new throttle falls through then I'll just put a pully on the rotary pot in a box for use with the original honda pull-pull throttle.

 

[spata]

It's been 6 months since you posted anything, any update?

-Spata

 

[einkleburt]

I did very little on it over winter, the project is still going - just slowly.

I've been working on finishing an MXUS based bicycle project first.

Hopefully in a month I should be back to the motorbike full time.

 

[einkleburt]

Well... that was a really long month!
*Bumping the thread so that I get back to working on it*
It turns out that moving a few hundred miles away and leaving the bike in storage isn't a good way to be productive


Let's see if I can get the motor mounted by Christmas!

Also, I need to re-research battery options. The prices have changed a LOT since I planned this build and the equivalent cells from HK are now £114 /pack up from £37!!!

That said, there are probably some better options than drone-spec LiPo around now.