Hi everyone, my name is Luciano, I live in Buenos Aires Argentina. After building the bike I decided to upload some of the info to a web page. This electric bike is based on Gordon Wong´s .I totally recommend it: http://gwinfo.dhs.org/ebike01

I have made 127km so far without any problems.

Technical data :

Maximum Speed : 23.4km/h Cruise speed: 18km/h

Autonomy : Aprox. 15km no pedaling.

Batteries : 24V 14A/h ( 4 batteries of 12V 7A/h )

Supposing each battery ( discharged 11.6V máximum charge 12.8V ) :

This is the electric bike, the PWM controller and three of the four batteries are in the inside of a plastic toolbox, while the battery that´s left is actually on the toolbox. It has two motors, one of each side, unified by a roller of 4.6cm, which goes against the wheel. The motors are 12V, but the batteries are used forming 24V 14A/h. Yet there seem to be no problems using 24V on these 12V motors. The batteries are rechargeable gel type. On the back of the toolbox there is a cooler, taking heat away from the MOSFET´s and the DIODE, which are mounted on a heatsink. 06/01/03 : I managed to put the 4 batteries in the box, removing the cooler and placing the half of the dissipator outside.

Batteries :

I used 4 Yuasa NP12-7 ( they cost u$S13 each ) rechargeable lead acid. They rate 12V and 7A/h, and they where found in a broken UPS. The gel type have a discharge curve associated. If for example, you consume 7A from a 7A/h gel battery, you will consume it not for an hour but four 33minutes. And if for the same battery you pull out 14A, they will last only 12minutes. That´s why I chose 4 batteries to get 14A/h instead of 2 getting only 7Ah. The motors with no load pulled 7A/h, and the range will by unsatisfying. Doubling the capacity, increases range always more than double.

Motors :

Two fan blower motors where used ( of a Ford Ranger ), they are 10cm diammeter. They where connected opposed to each other by a 4.6cm alumminum roller.

Note : the motors have the ground shorted with the chasis, when you connect the chasis electrically with the roller, you can not use them in parallel ( be careful ). To connect them in parallel and not in series ( more power and speed ), you need to open the back cover and aislate the connection with the chasis ( you´ll see a copper plate touching the chasis and the metal cover itself). Please verify with an ohmeter.

The roller goes against the wheel, between 0.5 an 1cm. The more you get it in the wheel, more torque, and more power you consume. The less you put it in, less torque ( you´ll notice and slipeage when you climb a hill ), but the range will increase.

This fan blower motors are 12V, 1150RPM, and each one consumes 3.5A/h with no load. Other have 3600RPM, and this changes the roller diammeter. In Gordon Wong´s bike, which uses an EV Warrior motor, the roller diammeter is 3cm ( by the way it is still in Meci at U$S40, and in ebay you can find a 24V 5A charger at u$S20 ).

The roller against the wheel is the easiest system of them all, but you lose power in the friction. They pulley and strap system are best, but they are difficult to build, and they must have a reduction.

PWM controller :

It has two 2 LM555P ( u$S0.30 each ), 1 LM311 ( u$S0.80 ), 2 IRFZ44N ( u$S 1 each), a cooler found in a broken PC ( 12V 0.12A/h ), and a heatsink from the same UPS I found the batteries.

The circuit is the following.

The first 555 is a monostable at 11Khz ( I suggest using between 10 and 20Khz ), it then feeds other 555 in astable mode, the width of the pulse is controlled by a voltage regulated with a pot. The LM311 is used to compare this voltage with aprox. 1V, when it is above, the reset of the 555 goes to "1" and the pulses get to the gate of the MOSFET. This is used the get the motors stopped, and to avoid scratchy sounds while turning them on.

I blowed the MOSFET´s a couple of times, but each time I learned something of it. The first thing I learned was that the ground and the chasis of the motors where connected electrically. Hume and fire came out of the fan. Also the 555 and the 7812 where blown. I suggest not connecting the cooller to the 7812 circuit, because it get´s hot and it finally breaks the rest of the circuit. The way I have it connected now, it with 2 15V 1Watt zenner in parallel, directly to 24V. Now the 7812 is not hot anymore. The blown MOSFET photo is the following ( RIP ):

Note : The MOSFET drain is generally connected to the case, beware then of the DIODE because all the cases can´t be connected electrically. Stick the diode and the 7812 to the heatsink and put a piece of cloth and don´t screw them to the heatsink. Use the case to connect to the MOSFET drain, you´ll save cables and future problems if you need to replace them.

Looking at the MOSFET from left to right ( GATE, DRAIN, SOURCE ), but I insist, use the case to connect to the drain.

The MOSFET I chose is the IRFZ44N, about a dolar each, and I used two in parallel ( Gate and Gate, Drain and Drain, Source and Source ). The IRFZ44N rates 60V and 50A DC ( 120A peak ), with an Rds(on) about 0.024ohm. The pdf with the details is :IRFZ44N. It will dissipate 60W at 50A continuous. But the consumption of the bike if around 7-14A/h aprox.

DIODE : I used the MBR2040 ( 20A40V). As the motors are inductive, when you take away current ( on off PWM ), they still have some power on the inductors, which need to "flyback" with a diode. I have also read to can use a MOSFET with Gate and Drain shorted, and with this you automatically get regenerative braking.

02/01/03 : I replaced the diode with a MOSFET, shorting the gate with de source. I can see an autonomy gain of 30%. To gain regenerative braking you need to turn on the "diode" MOSFET when the drive MOSFET is OFF. But I have to think it for a while, you can´t have both MOSFET ON at the same time cause the battery will be shorted.

6/01/03 : Between gate and battery ( + ), I connected a common diode and 2 13V 1W zenner in series. If the motor generates a 40V spike, the drive MOSFET will be turned ON. This replaces a main capacitor between battery leads.

Here is the heatink with the MOSFET´s, and the big DIODE stuck with the cloth. The circuit was built with an standard card. In my next project I promise myself not to use so many wires.

Mounting the circuit:

A plastic toolbox of less than u$S1.5 was used, and in it I mounted the circuit, heasink, cooler and three batteries. The fourth battery is actually outside the toolbox. An elastic rope was used to "hug" the toolbox to the bike.

Throttle :

The pot was put in a film can, and the a plastic support of a bike´s rear light was used. The other thing you can see is a solar light. I wish the bike was solar, but solar can only be used to charge batteries in day, otherwise the panels would be bigger than the bike.

 For suggestions, questions or exchange of opinions or material on Electric Vehicles please contact :

Luciano trentino123@hotpop.com