Go Down

Topic: multiplexing with pwm? (Read 9291 times) previous topic - next topic


The FETs look OK but are a bit overkill for a few LEDs. The darlingtons invert, that is a high on the input causes the transistor to turn on. With the LED in the collector then this inverts.
With any transistor you are back to using a series resistor to determine the current through the LED.


Can I use this instead? - http://users.telenet.be/davshomepage/current-source.htm

Will it perform the amplification as well or will I need to use it with a transistor?

If so how would the outputs of the 5940 connect with it?



Can I use this instead?

No, that is a constant current driver. It is not what you need.


But can't I use 5940 > non inverting transistor setup > constant current driver ?


This says it supports pwm http://www.instructables.com/id/Super-simple-high-power-LED-driver/

would it not work with the setup I just mentioned?



No, there is nothing to make the three parallel strings of LEDs share the current.


I'm not planning to use that led arrangment, the image was ripped from that instructables link.

I'm thinking a single one of these or similar  http://uk.farnell.com/lumileds/lxml-pwc2/led-rebel-es-high-power/dp/1834323 in place of those parallel strings


I'm not planning to use that led arrangment

Then don't ask if that circuit will work.  :0

If you replace those LEDs with that one LED it will work providing the current is correct, you don't say what current you want to use. Also it (the LED) will get very hot and need some form of heat sink.


haha  alright Grumpy!

Could've been clearer :)

Think I'm comfortable working out the resistor and regulator values for a particular current.

Presumably I can find arrayed versions of the regulator, transistor, and resistor to make a more compact multi channel version of this circuit.

Yay think I 'm finally there!


Testing the circuit in the image above (but with only one led), changing the voltage going into the lm317 affects the current going through the led. 

I thought that the lm317 will keep a constant reference voltage output regardless of input voltage (within limits) which means the current going through the led remains constant.....?

Is something amiss?


By output reference voltage I mean that between the vOut and Adjust pins


I thought that the lm317 will keep a constant reference voltage output regardless of input voltage (within limits)

Yes it will, what limits were you changing the input voltage over. It is likely that you are not yet into the limiting area of the characteristic. Have you tried some voltage measurements? You need to look at the voltage across that resistor.

You could try a larger value, as you increase the input voltage the voltage across the resistor will increase until it plateaus.

Also look at the signal with an oscilloscope the system might be oscillating with no capacitors to stableise it.


Hi Grumpy.

I think it's ok. I thought the voltage drop of the lm317 was less than what it is. I now get consistent current when the voltage is over 6.5 volts (~3 for the led and ~3 for the lm317)

Thing is, I have removed the transistor and pwm source, and I'm getting a much higher current going through the led than expected.

I'm using a 1.7ohm (tested) resistor between the adj and vout pins, and as I understand it the calculation is

1.25volts (regulated voltage) / 1.7ohms = 0.735Amps

whereas I'm getting 1.03A through the led....

Any idea where I'm going wrong?


I'm using a 1.7ohm (tested) resistor

It might be the accuracy of your test instrument. This is a very low resistance and most medium cost meters don 't do too well down there.
There is a spread of voltages on the LM317 from 1.2 to 1.3V but this would not be enough for the discrepancy you see. Also this voltage is temperature dependent but again I don't see it moving by that much.
That is an odd resistor value how did you get it?

Go Up

Please enter a valid email to subscribe

Confirm your email address

We need to confirm your email address.
To complete the subscription, please click the link in the email we just sent you.

Thank you for subscribing!

via Egeo 16
Torino, 10131