All,

New to the forum. Have a circuit (see attached pdf) to add speed control to a Solar panel-powered DC fan. The circuit works and allows good speed control of the fan. But the 2N3055 transistor which has a healthy heatsink and thermal grease is getting too hot, since this will be in use a lot I don't want to fry the circuit.

Not a real transistor guru so I suspect I have missed something or did something dumb, any help or insight would be greatly appreciated.

Bill

 

If you use a linear speed control like that then the transistor has to dissipate whatever power the solar panel generates that isn't going to be used by the motor.
To avoid that you could use a PWM circuit where the transistor will be either fully on, or off, repeated at perhaps a few kHz which minimises the power in the transistor. The inductance and mechanical momentum of the motor smooth out the pulses to produce smooth motion.

 

If you use a linear speed control like that then the transistor has to dissipate whatever power the solar panel generates that isn't going to be used by the motor.
To avoid that you could use a PWM circuit where the transistor will be either fully on, or off, repeated at perhaps a few kHz which minimises the power in the transistor. The inductance and mechanical momentum of the motor smooth out the pulses to produce smooth motion.

Albert Hall,
Thank you for your response. I appreciate the help.
I had been thinking a PWM circuit but I had seen some complaints that as the solar panel went from nighttime minimal voltage to a slow ramp up, it caused the PWM circuit to not start oscillating and the fan was not on.

 

As long as the PWM circuit will with the lowest voltage that will run the fan all will be well.

 

As long as the PWM circuit will with the lowest voltage that will run the fan all will be well.

Albert,
Thanks

 

Here is a PWM Circuit guaranteed to auto-start,
and which partially compensates for sagging Solar-Panel Voltage when
Speed-Control is set to less than 100%.
( It does not "boost" Voltage, so it can't create more Voltage than the Solar-Panel is producing )
And it uses your existing Speed-Control-Pot.
It will continue to work down to around ~7-Volts.
.
.
.

Schematic Corrected Oct-5-2021

 

The datasheet for the 2N3055 and every other transistor says that for it to saturate (turn on well) then the base current must be 1/10th the collector current. The collector current is 3.47A (and more when starting) then the base current must be 347mA or more.

The 12V from the solar panel is reduced 1V by the series diode and the base-emitter voltage of the transistor is another 1V then the 1k base resistor has 10mA when the speed control is at max. 35 times not enough current!
Then the transistor only partially turns on and gets very hot.

You cannot put 347mA in a pot so use a darlington transistor that needs a base current of only 14mA. But then the motor gets only 10V. The base resistor is (12V - 1V - 2V)/14mA= 642 ohms, use a standard 620 ohms/0.25W. Use a 2k pot.

 

Here is a PWM Circuit guaranteed to auto-start,
and which partially compensates for sagging Solar-Panel Voltage when
Speed-Control is set to less than 100%.
( It does not "boost" Voltage, so it can't create more Voltage than the Solar-Panel is producing )
And it uses your existing Speed-Control-Pot.
It will continue to work down to around ~7-Volts.
.
.
.
View attachment 238591

Hey LowQCab,

Thank you for the PWM circuit, it looks great!! Digikey part numbers and everything as well. I will make a PCB for it and let you know how it all works out.
Ok you got me with the screen name LowQCab ??

I appreciate the effort you put into your answer.

Bill

 

It's my Gmail address that I've had for ~20+years.
It comes from a time when I was determined to figure out everything about Speaker Design.
Speaker Cabinets, ideally, should have a "Q" factor which is as Low as possible, hence, "LowQCab".
I'm semi-retired now, so I've got time on my hands.

BTW, I forgot to mark the Op-Amp Inputs, it's the standard orientation with the
Non-Inverting on the bottom, and Inverting on top for both Amps,
the Pot works "backwards", moving towards ground increases speed.
If you want to control speed with an external voltage, the 0 to 100% range is
similar to a standard 555 Circuit, 33% to 66% of the 5-Volt Supply, over 66% of 5V is Off.
.
.
.

 

It's my Gmail address that I've had for ~20+years.
It comes from a time when I was determined to figure out everything about Speaker Design.
Speaker Cabinets, ideally, should have a "Q" factor which is as Low as possible, hence, "LowQCab".
I'm semi-retired now, so I've got time on my hands.

BTW, I forgot to mark the Op-Amp Inputs, it's the standard orientation with the
Non-Inverting on the bottom, and Inverting on top for both Amps,
the Pot works "backwards", moving towards ground increases speed.
If you want to control speed with an external voltage, the 0 to 100% range is
similar to a standard 555 Circuit, 33% to 66% of the 5-Volt Supply, over 66% of 5V is Off.
.
.
.

LowQCab,

Thanks for the opamp info, I had the polarities added on the oscillator, but was just debating on the top OA as to gnd looked like more speed. Hey back in my 20's I messed with making speaker cabinets (going for loud back then) so I should have recognized the screen name - good one.
I got recently fully retired by covid events but I have been enjoying fixing the music gear for the church worship team and making some home projects. Having way more fun retired.
Bill

 

If you're still into Speakers, I have an excellent Website that you must check-out......
https://billfitzmaurice.info/THT.html

 

Oooo, fun speaker stuff, thanks.

 

The Schematic I provided back in May had a serious mistake,
a Resistor was out of place,
it has now been corrected.
Sorry if it caused too much aggravation.
.
.
.