Building my first ebike, would appreciate some advice

[I4N]

If your charger is fixed at 3.6v, it'll be OK. I thought that you were just going to charge at 1 amp with no voltage cut-off, and stop when you saw 3.6v. The charge and discharge curve is "S" shaped, being nearly flat between 3.0 and 3,3v, but climbs/drops increasingly more rapidly either side of that.

Excellent D8veh, thanks for the info just what I needed to know. Over the weekend I connected a single cell to the dc converter at 3.60V and let it sit there until the amp flow slowed to nothing.

I let it sit off charge overnight and it held a steady, but lower voltage. Then when I was happy I hadn't destroyed it I connected the remaining cells up in fours in parallel (4p?) and put the steady 3.60V on them until baked.

Here's the part no. of the piece I was using.. LM2596HV (the HV stands for High Voltage) that is overkill for changing a single cell, the lower voltage (and cheaper) LM2596 would have been enough. However I want to input around 55V from the pack and convert to around 6V to run my bike lights.

I am thinking of buying 8 of these and building a little float charger, two cells at a time. I'll see how well the BMS keeps them all in check. At a couple of quid each it's a pretty cheap way of keeping them all in line.

Ignore the wire colours in the pic.. it was very late and I soldered the board upside down forgetting to swap the wires over.. I've only soldered stuff a handful of times so I was very pleased it all worked..

 

[I4N]

Hi just one problem you need to fit a washer on the inside with the round cut out with the 2 flats as if you just tighten up as you show the motor shaft can pull through the aluminum frame as the difference between the slot in the end off the frame and the diameter off the shaft is very close and the aluminum frame is very soft

Hi Frank, thats great advice thanks, I wouldn't have thought of that until after the motor destroyed my frame! I've added two washers on one side and a third by itself on the cassette side so hopefully won't have a problem.

 

[I4N]

I didn't have much time over the weekend to work on fitting everything to the frame, but managed to test fit the components and it all looks like it should fit okay.

The battery bag is pretty tight but I'll be able to get all the cells in there and the BMS too, the controller will be in a separate bag behind the seatpost. I'm going to run a small tube under the top tube to contain the wires from the handlebars going to the controller. I could run them through the bag but this bike will have to do double duty until it's paid for a dedicated leisure bike.

The only thing missing from this pic is the seatpost bag with the controller in it, hopefully it'll be a very neat installation...

 

[oigoi]

It might be a good idea to put something between the blocks of battery cells to prevent them touching each other.
As they are in your photo they could touch into each other and cause some almighty short circuits and explosions occurring between your legs!

Even just some pieces of plastic cut from milk bottles or simlar

 

[I4N]

It might be a good idea to put something between the blocks of battery cells to prevent them touching each other.

Thanks, in the pic I was just testing for fit, the orange blocks put the zipper under too much strain so they'll have to come off. I was thinking of wrapping the batteries somehow but hadn't really considered the exposed ends bashing into each other!! So thanks!

I can't find any decent foam tape over here, I was rummaging for some spacing material to wrap the batteries in when I looked at the box the batteries were delivered in, it's filled with a decent thickness of foam so I'm going to cut that into strips and make my 'cell blocks' with that, they can be properly insulated then with just a couple of wires popping out.

 

[Thamosy]

Hello. I’m following this with great interest. I’m intending to swop my direct drive hub for a geared one – a MAC – and I too have been in touch with Paul at Em3ev. I hadn’t thought about building my own battery though – and now think that should be my priority, mainly so as to have a good battery that will fit in the right place. I’d be grateful for a bit of advice about the choice of BMS. The ones listed on ‘EclipseBikes’ are rated as 30A/60A or 60A/150A (for whatever chosen voltage) and I’m not sure how to match up this up with the headway cell I’m proposing to use. Can anyone advise?

 

[I4N]

I’d be grateful for a bit of advice about the choice of BMS. The ones listed on ‘EclipseBikes’ are rated as 30A/60A or 60A/150A (for whatever chosen voltage) and I’m not sure how to match up this up with the headway cell I’m proposing to use. Can anyone advise?

Hi Thamosy,
I bought the whole battery kaboodle as a kit from Eclipse, they can supply everything you need to build a custom shaped pack. The BMS was the one they had tested and recommended for use with the Headway cells so I plumped for that.

As I understand it the cells can pump out 150 amps for a short period, much less for a sustained duration, so there's no problem taking that many amps from the batteries, however the controller I have is only rated for 35 amps max (I think) so I decided to go for the 30/60 amp BMS as it can handle 60 amps peak supply, 30 amps sustained.

I don't think the cell is your 'low point' in the system, I think it's most important to match the BMS and your chosen controller, that way the controller can get all the amps it needs from the BMS and you don't end up with a situation where the BMS shuts down the battery pack when the controller draws too many amps.

I'm no expert however so don't take that as gospel, you might want to drop the guys at Eclipse a line they generally respond to emails after a day or so.

Here's the key parts of the Headway cell specs from the Eclipse Bikes site:
Max Discharge Current (Continuous): 5C
Max Pulse Discharge Current (30 seconds): 10C (150A)
Recommended Discharge Current: 1C (15A)

 

[I4N]

Thamosy has started me thinking about how well I'd matched up the components, I ended up buying a 72V max controller in case I wanted to add more batteries in the future, however looking at the space in the frame I think it's doubtful I could get another four cells on, the main problem is cable runs and the little tabs holding them. Obviously if I buy more cells and increase the voltage I'd have to buy another battery charger too, an expense I'd rather avoid.

I've just re-checked the specs on the controller I bought and it can draw a maximum of 30 amps, which is a prefect match for the BMS. So hopefully the BMS won't be tripping out due to a mismatched controller trying to draw too many amps from the BMS.

9 fet IRFB4110 Infineon Controller (Motor With Hall Sensors)

I decided to go for a programmable controller as being a bit of a tinkerer I know I'm going to want to adjust the power delivery of the motor, ideally I would like to make it as similar as possible to a standard motorcycle engine, so you get max torque while you're moving not from zero so the bike flies away! This isn't so important for me, mainly for when I lend it to someone else.

Also the controller is a high quality piece with genuine quality components, so it should be reliable and give me years of service and seeing as this is a commuter bike that's the highest priority for me.

 

[oigoi]

I had some trouble with the bms I got from eclipse where the initial surge of current when I opened the throttle and the motor started drawing power would trip the current overload on the bms. This surge would only last for a tiny fraction of a second but it was enough to shut down the bms.
I got around it by adding a capacitor fitted on the power cables between the bms and the controller, the capacitor thus compensating for this surge and giving the controller what it needed without tripping off the bms overload.

 

[I4N]

I had some trouble with the bms I got from eclipse where the initial surge of current when I opened the throttle and the motor started drawing power would trip the current overload on the bms. This surge would only last for a tiny fraction of a second but it was enough to shut down the bms.
I got around it by adding a capacitor fitted on the power cables between the bms and the controller, the capacitor thus compensating for this surge and giving the controller what it needed without tripping off the bms overload.

Hi Oigoi,
Good idea! How big was the capacitor you used? What motor and controller were you using?

I'm hoping that I'll be able to avoid any issues like this by re-programming the controller, either that or the Cycle Analyst but adding a capacitor inline seems like a good reliable solution!

 

[oigoi]

I am using a 24v cyclone 500w crank drive motor with the built in controller, on a 24v 16ah headway battery. At first I used a 10000 microfarad capacitor but that was overkill now its got a 470 microfarad and seems okay. It depends how big the capacitors are in your controller, if they are big enough to cope with the initial current surge then it wont cause a problem with the bms. One of these "try it and see what happens" situations!

 

[Thamosy]

I’d like to add another question and it’s this: how did you select the particular motor – the 10T? I’m interested because I’m on the point of buying a MAC for myself and I don’t know which ‘loading’ to opt for. Other than controlling the top speed, I wonder what effect the 200rpm and 255rpm loading will have? I’ve calculated a top speed (with a 26” wheel) of around 19mph with the 255 loading and around 15mph with the 200rpm loading, but other than that can anyone help me over what else might be affected?

If opting for the lower, 200rpm, I’m guessing that might provide more torque, but would it mean a tendency to ride with maximum power from time to time and so use up that precious battery power? And apart from concerns over power usage, I also wonder what might happen if (say) I’m belting downhill at a speed greater than that calculated for the motor’s top speed.

Can anyone give me any pointers?

 

[I4N]

I’d like to add another question and it’s this: how did you select the particular motor – the 10T? I’m interested because I’m on the point of buying a MAC for myself and I don’t know which ‘loading’ to opt for. Other than controlling the top speed, I wonder what effect the 200rpm and 255rpm loading will have? I’ve calculated a top speed (with a 26” wheel) of around 19mph with the 255 loading and around 15mph with the 200rpm loading, but other than that can anyone help me over what else might be affected?

If opting for the lower, 200rpm, I’m guessing that might provide more torque, but would it mean a tendency to ride with maximum power from time to time and so use up that precious battery power? And apart from concerns over power usage, I also wonder what might happen if (say) I’m belting downhill at a speed greater than that calculated for the motor’s top speed.

Can anyone give me any pointers?

Hi Thamosy,

I went for the 10Turn motor as I wanted a torquier motor to climb the 25-30% hills I have right outside my house. Paul said the 8Turn motor which is slightly faster would work okay. As I understand it, the 8T motor would be generating a more heat while climbing and the ambient temperature around here can be over 40°C for much of the summer, which doesn't give it much chance to cool off..

Hence I decided to go for the slower torquier 10T motor and then over volt it slightly to 48V to increase the top speed, not that I'm too fussed about going very fast.. but my setup should get me to 30mph or so on the flat which is more than fast enough for me, I mainly want the higher speeds for hill climbing.

If you haven't seen it already you can use the simulator at Grin tech here, it's worth just playing around with it for a bit ebikes.ca Hub Motor and Ebike Simulator

I used the BMC Trq for the 10T motor and BMC Spd for the 8Turn, not quite accurate but close enough. Stick your controller details (I used Custom 30Amp) in and battery (I used 52V 20Ah Ping). It looks like my top speed is around 33mph, peak efficiency at 25mph (say 75% throttle) with that setup.

the thing that sold me most was a guy in San Francisco riding up a 20 or 30% hill using a Mac 10T without having to pedal..
MAC motor hill climbing test - YouTube

 

[Deleted member 4366]

There's a slight trade-off of torque for speed, but the most important thing is the efficiency. If you have alot of steep hills, then a low-speed version will be more efficient. You also have to remember that power drops off in the last 25% of the speed range, so if you want to be able to maintain 15mph against a strong head-wind, a 20mph one will be at maximum power and efficiency at that speed. To calculate the speed, use 201 rpm as 15mph in a 26" wheel, but then you have to multiply by 40/36 to compensate for a 36v battery actually averaging about 40v.

I've found that a motor of about 270 rpm is ideal. It still has good climbing ability, it's happy at 12 to 15mph, but if you want to turn the speed up a bit, when you're in a\hurry, it'll run up to about 22mph.

The Bafang CST motor is IMHO better than a MAC. It's a good quality build, and takes a cassette rather than the crappy free-wheels. Probably a bit cheaper too. The Bafang BPM is similar to a MAC, very reliable and a lot cheaper.
Bafang CST 36V 500W REAR Driving E-Bike Conversion Kit. - BMSBATTERY

 

[MikeyBikey]

When you click on that link it says '48V1000W-FRONT' , are BMSBattery reliable?

 

[Deleted member 4366]

When you click on that link it says '48V1000W-FRONT' , are BMSBattery reliable?

Don't take any notice of the URLs. None of them relate to the product shown. BMSBattery are as reliable as any Chinese supplier. No warranty, but problems are rare.

 

[I4N]

Well I finally got some time today to finish the few things needed to get the bike running.. And quickly took my first ride!!

First impressions are good, it's really hilly here and I'm getting used to the throttle but I had to slow down round corners going uphill! I was cycling with the motor, it makes you feel bionic!

I had a few issues with the bike, namely tyres that don't fit, I'm still experimenting. And the ebrakes aren't compatible with my setup as mine is a cheap integrated brake and gear lever. I'll have to figure something out.

I also couldn't fit the wired up battery in the frame bag with the bms and all the wires, it was just too tight.

I also thought I'd blown the bms at one point but it seems to be working okay now..

In short my top speed was 46kph on the only flat bit and an average of 20 odd kph over 4km of some very hill terrain.

I'm pleased with the performance it is pushing too fast for me at wot and cycling uphill feels downright bizarre!

I'll post more later.

 

[I4N]

Don't take any notice of the URLs. None of them relate to the product shown. BMSBattery are as reliable as any Chinese supplier. No warranty, but problems are rare.

This is what concerned me about BMS Battery and Green Bike Kit, apparently they order from the same wholesaler or factory, from what I've read BMS have had a reputation for poor customer service so don't expect much. I contacted Green Bike Kit but had ordered from Paul before i got a reply back from them. However my biggest concern was not getting exactly,what I wanted, I've heard it's a bit pot luck getting a torque or speed motor, having said that their website has been updated recently and it has an option to specify the motor you want.

What I appreciate about the Mac kit from Em3ev is that it's all setup and tested before it leaves the factory, I just had to plug a couple of cables in and flicking the on switch it all came to life, it was really very simple. Plus I got a half decent branded and well balanced wheel too.

 

[Deleted member 4366]

And the ebrakes aren't compatible with my setup as mine is a cheap integrated brake and gear lever. I'll have to figure something out.

Hidden wire brake sensors from BMSBattery
HWBS - Hidden Wire Brake Sensor 1pcs - BMSBATTERY
or I've got a pair you can have for £10 plus postage. You have to do a small amount of wiring too connect them to your brake connectors. Basically one of the three wires has to go to a 5v,like the throttle red, or solder a wire to a spare pad in the controller, then the other two wires go to tour brake-switch connector.

 

[I4N]

Well here's the finished article.. I had a few teething problems with the battery and the Headway cells are LARGE, thats the disadvantage of LifePo4, however I finally managed to pack them in the frame bag and this is the final result.

I need to trim the battery to controller wire down to size and I'll make the charger connector more accessible, there's a couple of well shielded holes at the front and rear of the bag so the controller will pop out the back and the charger will pop out the front..