Go Down

### Topic: Combining transistors for more current? (Read 5462 times)previous topic - next topic

#### dkl65

##### May 03, 2013, 10:05 pm
Hello forum!
I was trying to power 2 DC hobby motors with a 500mA NPN transistor each, along with an Arduino UNO. The power source was 4x AA Alkaline batteries, all relatively new. However, the motors were struggling to move, and the Arduino was constantly restarting.  I used my multimeter to measure the resistance of the motors. Both were 2.5 ohms. So, does that mean that the batteries do not have enough current output to power 2 motors and an Arduino?
Another question: is it possible to combine 2 transistors to double the current output (e.g. 200mA transistor + 200mA transistor = 400mA transistor?)
BTW, I know Ohms law.
Thanks in advance! If my explanation is unclear or missing details, please politely point it out.

#### AWOL

#1
##### May 03, 2013, 10:19 pm
Quote
is it possible to combine 2 transistors to double the current output

Bipolar transistors, no - they won't share nicely.
MOSFETs yes.
All to do with temperature coefficients.

As to your first question, feel free to furnish a schematic.
"Pete, it's a fool looks for logic in the chambers of the human heart." Ulysses Everett McGill.
Do not send technical questions via personal messaging - they will be ignored.
I speak for myself, not Arduino.

#### Erdin

#2
##### May 03, 2013, 11:03 pmLast Edit: May 03, 2013, 11:06 pm by Erdin Reason: 1
Your battery is probably too weak.
Can you use seperate power supplies for the Arduino and the motor ? That is easier while testing.

Allthough AWOL is right, I must admit that I used NPN transistors parallel. I had a transistor as emitter follower which could not do a start pulse, it needed a little more. By adding a transistor parallel the problem was solved.
If you use a base resistor for each transistor, the collector current could double for matched transistors in theory. But only if the transistor is not used as a switch with a saturated transistor. If you use the transistor as a switch, one will get most of the current and blow, and after that the other transistor will blow.

#### dkl65

#3
##### May 03, 2013, 11:28 pm
Quote from: AWOL
As to your first question, feel free to furnish a schematic.

This is the current schematic. If I connect the motors parallel directly to the batteries bypassing the Arduino, the motors spin very fast. There is also a spark when I connect the power. I have no datasheet for the motors. The motors are connected to wheels powering a four wheel drive skid-steering robot. Even without load, the motors struggle to spin.

#### Erdin

#4
##### May 03, 2013, 11:41 pm
The schematic is okay.
You could add a capacitor in parallel to the battery near the motors.
I think the motors need a lot more current and larger transistors (or mosfets) to control them. Even a small motor could require more than 500mA.

#### dkl65

#5
##### May 03, 2013, 11:47 pm
Quote
You could add a capacitor in parallel to the battery near the motors.

What kind of capacitor and where should I put it?

#### strykeroz

#6
##### May 03, 2013, 11:53 pm
Hi,

Since you don't have spec on the motors, do you have a multimeter you can measure their current draw when you're running them directly from the battery? To get an idea of the range, measure first the current when they're free spinning.  Then, see if you can prevent the motor from turning to measure the stall current.  This will give you an approximation of the range of current draws each motor will demand of your circuit.  From this you'll be able to determine if your 0.6A (absolute max) on those transistors will do the job allowing for some headroom.

Then if you're confident 0.5A will run the motor, and your circuit connections pass a double-check, the next variable is the Arduino - ie reconfirm the value on the pin that's driving the base of these 2N4400s.  Perhaps output that activity to the Serial monitor for debugging.

Hopefully the issue will jump out at you in that process.  All the best!
Geoff
"There is no problem so bad you can't make it worse" - retired astronaut Chris Hadfield

#### dkl65

#7
##### May 04, 2013, 12:05 am
Quote from: strykeroz
can measure their current draw when you're running them directly from the battery?

Yes. The current wildly fluctuates between 1.2A and 0A when the wheels are off the ground. Stall current is steady at 1.2A.

#### dc42

#8
##### May 04, 2013, 08:45 amLast Edit: May 04, 2013, 08:48 am by dc42 Reason: 1
6V is only just enough to power an Arduino from the Vin pin, and with a load of 1.2A, your battery voltage will drop to well under 6V. Either use a separate battery to power the Arduino, or use 5 or even 6 AA cells instead.

Although connecting BJTs in parallel for higher current isn't recommended, for switching applications you can sometimes get away with it if the transistors are of the same type, preferably from the same batch, and you use a separate base resistor for each transistor. But don't expect the current sharing to be even. I would never do this in a commercial design. For switching 1.2A it would be better to use a mosfet.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

#### 123Splat

#9
##### May 06, 2013, 09:40 pm
I don't understand all this about bipolar junction transistors not working well in parallel.  I've been doing it for forty years, in regulated power supplies (O.K. maybe not a great idea for a high fidelity amplifier, but My favorite 12 VDC source is an old Pyramid PS-52K, delivering 45A. it uses eight 2N2771's regulated by a 723 op amp regulator).  Is it the fact that you are trying to drive them PWM, or what?

#### Erdin

#10
##### May 06, 2013, 09:48 pm
For a power supply the transistors are not in saturation.
I also have a power supply with three parallel 2N3055.
The transistors didn't match very well, but I compensated that with a slightly different base resistor.

If a transistor is used in saturation (as a on/off switch) a very minimal difference will have a large effect.

#### 123Splat

#11
##### May 06, 2013, 09:52 pm
Thanks Erdin, beginning to make sense.  Just wake me when it's over.....

#### sparkylabs

#12
##### May 06, 2013, 11:20 pm
BJT's will pass more current as they heat up so the hotter they get the more current they pass and the hotter they get and it goes on into thermal runaway and meltdown. If one transistor can handle the motor then no problem as even if one takes over it will handle it using the second one as backup for starting.

you can parallel BJT's by using an emitter resistor, the more current that flows the higher the voltage drop on the resistor and the lower the Vbe and the lower the base current and so it pulls the transistor back into safe limits before thermal runaway. This is of course inefficient as you loose power in the emitter resistors but it was the best thing until mosfets came along that have the opposite behavior so share naturally.

#### MarkT

#13
##### May 06, 2013, 11:22 pm

For a power supply the transistors are not in saturation.
I also have a power supply with three parallel 2N3055.
The transistors didn't match very well, but I compensated that with a slightly different base resistor.

If a transistor is used in saturation (as a on/off switch) a very minimal difference will have a large effect.

In linear circuits you add a small amount of emitter resistance (if necessary) to promote fair-sharing of
current, and mount them on the same heatsink close together (so Vbe will be matched).

In theory you could use the same trick with switches, at the expense of more heat loss in the emitter
resistors - but it would need tuning to the actual load current.

The key point about BJT's is that Vbe drops for a given current as temperature increases, so the hotter
device turns on more, gets hotter, the colder device is starved of current and cools down - thus the
important of mounting on the same heatsink.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Go Up