transistors connected to PWM

Hi

I have two BD241a transistors connected in parallel to a pwm output on my UNO. Its all on protoboard. The idea it to drive a DC motor with PWM

I have the base connected directly to the pin, collector directly to the +ve on the supplying lipo battery and emitor on the +ve of the motor, the -ve of the motor is going to the common ground of the circuit. There js a protection diode on the motors poles.

PROBLEM at 255 i have 5.4v on the UNO pin and the motor runs fast. Until i put strain on the motor, then thempin voltage drops to 2.5v and slowly drops wkth time.

Is it possiblenthat there is some backfeed from the transistor back to the pin thats killing it?

What's your circuit?

If I read the circuit description correctly (although a pic would be better…) sounds like you have the load on the low, emitter side of an NPN transistor. I don’t know much about this stuff, but afaik the load should be on the high, collector side of an NPN. Maybe that’s part of the problem?

I’m not sure what it means to have 2 transistors in parallel?.. You really need to post a pic of the circuit.

You shouldn't connect two silicon transistors in parallel like that, especially without something (like a 1ohm emitter resistor) to help balance the current. As it is, one transistor will tend to "hog" the current as it heats up. You absolutely need to limit the base current with a resistor. The transistor is an NPN and you should move the motor to the "high" side, above the collector and tie the emitter directly to ground.

I have the base connected directly to the pin, collector directly to the +ve on the supplying lipo battery and emitor on the +ve of the motor, the -ve of the motor is going to the common ground of the circuit. There js a protection diode on the motors poles.

  1. If possible use one NPN transistor rated for the max motor current or higher, two may not share current equally so you don't always get the full advantage of using two parallel transistors.
  2. The arduino output pin must have a series current limiting resistor wired to the base of the transistor. Try 500 ohms to start.
  3. Ground the emitter, collector wires to motor and other motor terminal wires to positive voltage source. Wire a diode across the motor terminals, cathode end wired towards positive voltage source.
  4. Negative voltage source wires to collector and then on to a arduino ground pin.

Lefty

"Is it possiblenthat there is some backfeed from the transistor back to the pin thats killing it?"

The problem is that the NPN Base-Emitter looks like a forward biased diode to the arduino pin. You need some current limiting there to: prevent the arduino pin from sourcing current until it dies; overdriving the base of the NPN until it dies.

Also, with NPN, you want +supply to motor '+', motor '-' to collector, and then emitter to Gnd. NPNs like to sink current, and can do so with 0-5V on their base. PNPs like to source current, but if the supply voltage is >5V, then a pullup resistor to the +supply is needed to turn them off, and a NPN is needed to pull the base low to turn it on.

a diagram is needed. What is the supply voltage for the motor ? this is very important. It sounds to me like you are having problems with different supply voltages.

Diagram:

thanks to everyone,

I feel like a wally not having noticed that the NPN needs to be connected differently to a PNP, I definately have the collector conencted to the source voltage, the emmitter to the pos motor pole and the rest to ground, I had a 500 ohm resistor but it made the output of the transistor lower, so i removed it whilst looking for a reason ,

Am i right i thinking that the NPN is getting mother !"·$ingly hot due to being wired by a dumbass? I have two in parallel to try to stop one overheating, I couldnt work out why pulling just a few amps was frying it,

I will change the layout later and get back , Thanks,

If you put transistor in parallel then one will end up taking all of the load. with BJT's the hotter they get the more conductive they get so one will burn itself out. That is why we put a low value resistor on the emitter so that as the current rises the voltage drop rises helping to dump current onto the other ones.

With mosfets we don't have this problem

Well , i have changed the connections as per crossroad,s advice and now my indicator leds dont dim, the motor seems more responsive and the transistor isnt getting as hot

However, i still get more speed from the motor when connected directly to the battery than when run with the transistor. How much voltage loss should i expect to see?

Also , i have noticed that when the motor is under load, at full speed my indicator leds are still being interferred with to the point where they almost turn off, this doesnt happen with the motor at 100%when not under load.

Assuming my its not the battery lacking power, (the motor pulls well) its not the voltage regulator losing power (the arduino itself doesnt brown out) that seems to leave the same question of back fed voltage on the negative line, Could i just use a diode to sort that out? Or will i need some cleverer solution?

Thanks jn advance

Your voltage drop is in your transistors. Measure the voltage across the motor terminals and compare to voltage applied to the circuit. You would have significantly less loss if you used a proper sized MOSFET in place of the silicon BJT you are using. There will still be some voltage drop, but it will be more of a resistive type drop than a diode-like voltage drop. It should also be significantly less.

What value resistor are you using between the Arduino pin and the base of the transistor?

Ok , not knowing the difference between a bjt (sounds a lot like my favorite sandwich) and a mosfet, I will hunt a mosfet in the parts box,

any decent sized one ok ?

dc42: What value resistor are you using between the Arduino pin and the base of the transistor?

510 ohm

If you do not find a "Logic Level" MOSFET, you will need a BJT in front of it to pull the Gate low, and a pullup resistor to your motor source volage to pull the gate high. (non Logic Level MOSFETs usually need 10V swing on the Gate to go full on or full off)

any ideas why the leds brown out ?

Not enough current available most likely. Or the voltage is drooping (indicating not enough current is available).

i have the following transistory looking thingys,

irf3205 x 2 irf9z34 p60nf06 irfz44n (pulled from an atmel based charger, so i will look at the board later and try to work out how they were using it)

bdx54c looks like its a darlington transistor

and one i cant find a data sheet for, it says NEC 7805 913oe (possibly a votlage regulator ?)

Well, i dont know what im looking for one the datasheet , how to distinguish a good one for my purpose from a bad one, any ideas which is best suited? or should i dig deeper into the parts bin ?