Lapierre electric bike

[jpgey]

Hello, I spend so much time reading ES for getting usefull information that it was mandatory for me to show you my new setup.
As you can see this is a french stuff and my english is not as fun as i hope !
Anyway, I had started few month ago like all others with Crystalyte HS3540, Ping battery put on the rear (22lbs ouch!) and nothing special.

Before:


An auxiliary battery was added at the bottle place, without DC converter I could not electrify the lights.

And now I start an upgrade for this stuff, trying to mold epoxy and try to have an usable bike, something cool and really usable.

Here begin the topic, let me share my findings !!!

Final setup:

[jpgey]

Ok, now I need to secure the battery. Ping battery is very reliable (48v 20Ah) but BMS small cables need to be rewired.

During the operation I burn the BMS then it was a good idea to change it with a high rate Bms !
Something around 60A-100A max should be good enough, don't forget that this battery cannot deal with high C

[jpgey]

Ok, now close the battery

[jpgey]

Epoxy molding. First epoxy of my life, hard to say but it was really a nightmare !
And not a tiny box ! A box that need to carry 22lbs with only two anchors, with all electric stuff inside (Bms, DC/DC, switch, pre-charge, fan ...)
The most difficult was the size, I need this box to fill the triangle framework and to let the pedal turn free !

Here the mold inside the hoover

[jpgey]

The Traco Power DC/DC converter seems to be a good choice, I need only 12v for lights, horn and fans (fans for the controller, we will see it later )
It is a 15W converter with a real 86% of efficiency. Deal with perhaps 2w for fans, 10w for led light and keep an extra for some stuff)

Master switch and pre-charge resistor (180ohms 10watts) should charge the capacitors in less than 20seconds (capacitors, ... we will see later )

[jpgey]

Now the most funny !
Try to add all this stuff inside the battery box and keep it possible to open

[jpgey]

OK, after hours and hours trying to add all the parts, here is the battery box !
Oh but what is that strange red light rounding the box
No, this is not a flash reflect !
No, this is not glow paint (from http://www.glowtec.co.uk/ of course !)
No no, this is an EL stripe You know, it is pretty flat (1,5 mm) and works like neon with a small converter (100volts 1000Hz).
Very cool stuff, perhaps dork

[motorino magnet]

Fantastic !

Inspiring !

[jpgey]

As I don't want to loose any power for aerodynamics i try to setup a crown fork (faring ?)





During paint :

[jpgey]

Finishing the battery box with a cover

[jpgey]

Painting, perhaps I shoud use black paint instead white ?

[jpgey]

Closer look under the cover. A small fan will extract hot air from BMS,

[jpgey]

Time for controller upgrade. It is a regular Clx 48v with 12 x IRF 4110, sensorless.
This controller can handle a lot of amp when EMF return from motor coils is under control. Lots of big caps (Ultra Low ESR needed) and plastic caps.
This mod consist in distributing 5 x 1000uF 63v near Mosfet and 2 x 2,2uF 500v plastic caps at the power rail end and begin. Of course a huge copper cable bypass the shunt.
For fun I change also all the other caps on the board with ultra-low Esr caps. Changing heat-sink and add a 10000uF bunch of ultra-low Esr for immediate power request



While I need some battery safety and don't want to destroy the Bms with current return from switching power at low speed/huge amp, I add a Schottky stripe. IRF 40CPQ060 can handle the request power if used with large heat sink. I add a temp sensor for information and I glue the CA shunt. CA shunt is done for 40A cont. and 100A peak.

[jpgey]

Mesuring the stuff before put it in box.
Low voltage is really small because of pre-charge capacitors and because CA register the most small value it can start.
146Amps should be cool enough to destroy an entry-level torque arm, I must weld a nice torque arm.
During the test (standalone) the brake pad start smoking and disk become blue like motor-bike exhaust !
BMS cut at this level but no magic smoke from Mosfet. Heat-sink was near from hot. Add a temp sensor also here for fans control.

[jpgey]

OK, power is under control. Mold a dork epoxy box for controller with fan, schottky, UL bunch caps ...

[jpgey]

Close-up

[jpgey]

Really need some high level brake pads

[jpgey]

I cannot find any hydraulic brake lever with switch from bike supplier. Only from motor-bike. As I do not want to change i add a small switch near the lever.
It will activate a small relay to switch regen (not connected but it stops the controller) and rear brake light.

Sorry for picture quality

[jpgey]

Time now to have some rest with CMS welding


Very small DC converter for led stuff. Adding a copper heat-sink. Front light (cool white 2 x 1watt with angle around 60degx30deg. Rear light home-made from epoxy.
Front: 750mA
Rear: 350mA
Some heat-sink also for the led.

[jpgey]

Front led stuff

[jpgey]

Testing front light. Heat-sinks are doing good job (small warm after half an hour)

[neptronix]

Sorry to rain on your parade, this is looking like a cool build, but the amount of amps you are running on the ping is concerning, especially considering that you're going to enclose the battery where heat cannot escape.

I do not have experience with ping batteries, but they are rated more like 2C, and that is a peak load of 6C on a 20AH battery. I would strongly consider cutting your max amps in half, especially if you live in a hilly area where you will pull very high amps on that high speed wind..

I hope you do not plan on keeping that controller set to over 60A long term.

[jpgey]

Sorry to rain on your parade, this is looking like a cool build, but the amount of amps you are running on the ping is concerning, especially considering that you're going to enclose the battery where heat cannot escape.

Thank's for reading my rainy posts !
I totally agree with you, Ping battery cannot deal with more than 2 or 2.5 C before become damaged. Heat from battery box is under control with a fan. But of course I do not plan to use so much C from this battery, that's why I add a lot of ultra-low ESR caps and Schottky diode !

I hope you do not plan on keeping that controller set to over 60A long term.

Why not ?

[jpgey]

I don't want to spend too much time in engineering, then this Kemo twilight switch will help a lot

I need the front light-up automatically while I drive under tunnel

[neptronix]

Sorry to rain on your parade, this is looking like a cool build, but the amount of amps you are running on the ping is concerning, especially considering that you're going to enclose the battery where heat cannot escape.

Thank's for reading my rainy posts !
I totally agree with you, Ping battery cannot deal with more than 2 or 2.5 C before become damaged. Heat from battery box is under control with a fan. But of course I do not plan to use so much C from this battery, that's why I add a lot of ultra-low ESR caps and Schottky diode !

A capacitor will provide maybe a second of discharge buffering at most, it's not the spike of amps that will kill the ping, it's the prolonged >2C draw that will produce lots of heat & sag, wasting watt hours and damaging the prismatic cells.
A fan will not do diddly squat to cool the cells outside of the cells of the edges. Think about it - the cells in the middle of the pack have about 1/10th of the cooling surface that the ones on the edges have.

Do not think that it will protect you at all.

I hope you do not plan on keeping that controller set to over 60A long term.

Why not ?

The ping batter problem i mentioned, and the fact that You're running a 12FET controller, which realistically is good for 'bout 50 amps continuous in most situations. It can handle power spikes, but if you're going to run much over that continuous ( for example, when hill climbing ), something is going to blow, especially on that amp-hungry high speed wind.

Do yourself a favor and take the bike up some gnarly hills, then measure the temp of the battery and controller afterwards.