18fet controller make click sound

Erikjan

100 W
Joined
Apr 7, 2010
Messages
148
My controller don't work, when i connect te ignition wire it make a little click sound like a relais is switching.
When i twist the throttle nothing happens.

I messure 5V at the red and black throttle wire and 0 ~ 4V at the green and black throttle wire so that looks normal.

When i mesure the hall connector i have 5V at the red and black wire. The other 3 wires toggles between 0.7 and 5V when rotating the wheel.

do somebody now what can be the problem?

The controller is 18Fet lyen controller with irbf 4310 fets.

When i try to program it makes the same click sound when i push the button.

sorry for my bad english
 
i open the controller but dont find lose wires.
I connect the power without the controllercase but i can nog hear what is making the sound.

as you already say there are no moving parts in the controller and i dont know wich part can make a sound.

I will post the pics tomorrow i dont have a cable here.
 
maybe the sound was from the motor. if you don't hear the sound from inside there is no need to open it further.

did they motor ever run before or is this a new build? do you know that the phase and hall wires are connected correctly for the motor/controller combination?
 
It is the controller making that sound becouse is makes it to when no motor is connect.
When i connect only the mainpower wires and then connect the ignition wire it makes that sound.

It is a new controller pcb and fets. I damaged the old, some fets exploded and pcb was burn.
I replace the pcb and fets and caps. Then put all the wires from the old pcb to the new.
The motor run for i think 2 sec. Then noting.

The motor looks good. Between the phase wires i have 0,2 ohm at all combo's.
The Halls are working. It is a scooter hub 1500W but i use it at 60V 60A for a long time always was good.
 
i don't now how i can check mosfets. I will findout.

But when i mesure between the heatsink and the phase wires there is no connection.
When i mesure beween the phase wires all combo's are 17k ohm.

There are no burn parts in the controller.

I go find out how to check mosfets. i only have a simple multimeter
 
some pictures


12502972.jpg


12502961.jpg



More at http://djjr1982.nl.picturepush.com/album/250595/p-ebike.html#column2::/album/250595/detail/12502961
 
judging by your name you are from the netherlands, if you are near zwolle, you are welcome to come by and i could help you figure this out.
 
if you have a digital multimeter it will have the diode tester built in.

the trick to remember is that always the drain is at a higher voltage than the source in these mosfets. these are n channel mosfets and if the source voltage is higher than the drain then the body diode of the semiconductor that the mosfet is made in will conduct current in that direction. when the connection is reversed and the drain is at a higher voltage than the source then the body diode does not conduct.

there are three phases, each has a high side mosfet between B+ and the phase wire and a low side mosfet between the phase wire and the ground or B-.

if the mosfet drain has shorted to the heat sink then you will see a short between the phase wire and the case, or between the red wire B+ into the controller and the case.

to test the mosfets, put the red probe on the large red wire into the controller. then use the black probe and measure each phase wire. they should be open circuit. then reverse the leads and put the black probe on the large red wire, and use the red probe to measure the body diode by probing each of the phase wires. they should all conduct and have the same forward bias. it will read out as millivolts on your meter.

to measure the low side mosfets you do the same thing. put the black probe on the black wire from the controller, and probe each of the phase wires with the red probe. the loside mosfet drain is attached to the phase wire so it will be open circuit. if you put the red probe from the meter on the black wire and probe each of the phase wires with the black probe then you will see the forward bias of the body diode for the loside mosfets. all the forward bias voltages should be identical. hiside and loside.
 
I am from Holland Zwolle is 140km thanks for the offer if i cant fix this it is an option.

I mesure with the diode test als Dnmun say,

No short between case and phasewire also no short between case and red wire.

0,4mv for all forward bias.

And open circuit for black probe at black wire and red probe at red wire.
Is it normal to see a value like 1,8 for a second before i mesure open circuit.

I think mosfets are ok.
the 5V regulator is ok because i mesure 5V at the throttle and hall connector.

Now i dont know what next to check maybe the caps?
 
yep mosfets ok, the 1,8 you saw was from the capacitors charging up. that is common when measuring in a circuit. just to verify your comment, you measured between the phase wires and B+ or phase wires and B-?

430-440 mV is common on the high current mosfets i have seen. you also now know the shunt wiring is ok since you measured through the shunt to get the loside readings.

the click is a stumper. i thought you would find a loose solder joint on one of the big caps maybe.
 
I mesure both phase wires and B+ and phase wires and B-

When mesure the B+ i connect the black probe to B+ and the red to phase wire.
When mesure the B- i connect the red probe to B- and the black to phase wire.

Values are 0,406 ~ 0,416

Tomorrow i have a tool to check the capacity from the cap's
Maybe some is damaged
 
i doubt if the caps are damaged, or they would be shorted out and the controller would not have power. if the source/drain caps were shorted you would not see the open circuit in the diode tester. all jmho.
 
I made a video where you can hear the click sound.

http://www.youtube.com/watch?v=rU3DGJ5ZElg&feature=youtu.be

I don't understand what part is able to make that kind of sound.
 
Ok i dont have a stethoscope but i will change the caps one by one till i find if there is one damaged.

I start on the wire output side becouse i think the click come from that side of the pcb.

But it is not easy to trace the click.
 
use a small stick. maybe chopstick and put it on the top of the cap and put the other end up against the inside of your ear in the flat open part so it pushes against the cranium and listen then.

or you can power the controller up slowly by using a resistor in line with the power so the caps would charge up slowly and that might eliminate the click since the caps would be charging up slowly.
 
Back
Top