Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

teach old dog new LED tricks

Status
Not open for further replies.

zipdogso

New Member
Until last year when I dabbled briefly I hadn't picked up a soldering iron for over 2 decades and forgotten most of what I learned....but my recent project requires amongst other things for me to go back and do electronics again.

My problem is I need to have an LED display for the state of a fan thus:-

1./ the LED is off either the fan or something worse has gone wrong.
2./ the fan is attached to power but not rotating. (RED LED.)
3./ the is powered and running. (GREEN LED.)

To break myself back in gently I have been trying a few things on the bread board.

The schematic attached ( not a good one i admit.) is where I have got to so far and works.

On the left of the relay (as you look at it.) is a simple constant current LED driver circuit. On the right is the fan it is 12v 1.5amp 12cm. The relay is DPDT. I am running both the LED circuit and the fan from the same supply and I wanted to be able to adjust the fan input hence the constant current LED circuit.

As shown in the schematic everything works fine just for testing I also ran the relay coil off the same supply as well with a switch to turn it on and off.

BUT.....

1./ It doesn't satisfy my first fan state i.e. if the fan breaks down the LED will continue to show red or green. At the moment I can only think I will have to run the fan on the relay coil circuit with the coil going to the base of a transister and the collector driving the fan (due to the 1.5amp current draw of the fan.)and add an extra relay for the red/green states but I havent come up with anything beyond this yet.

2./ I really wanted to use 2 lead bicolor LEDs I have 3 fans to monitor, temp outputs and allsorts so I would prefer one LED display per fan.

Any ideas ?

EDIT - I did have a litle fiddle around with the bicolors to try and get them working but I couldn't it going.
 

Attachments

  • fan led setup.pdf
    1.1 MB · Views: 266
Last edited:
Suggest to place in front of the fan, a vane that makes contact on one side when there is wind (green led), and contacts the other side when there is no wind (red led) acting on it.
The leds supply should be derived from the fan supply. That way your 3 conditions meet.

+Fan--------------------------Red--------o_o Vane---------/\/\/\/\/---------Fan-
+Fan-------------------------Green------o

The vane shown making contact for the red with wind absence


Miguel
 
Suggest to place in front of the fan, a vane that makes contact on one side when there is wind (green led), and contacts the other side when there is no wind (red led) acting on it.
The leds supply should be derived from the fan supply. That way your 3 conditions meet.

+Fan--------------------------Red--------o_o Vane---------/\/\/\/\/---------Fan-
+Fan-------------------------Green------o

The vane shown making contact for the red with wind absence


Miguel
Showing red and green is not the problem (though I would like to get it down to just one LED.)...showing when the fan has broken down is.
Oh sorry...I get it .... but having vanes on the fans will be a pain (impossible in one case.) where they are situated...I'd rather not. These aren't big fans I'm talking about only 12cm.
 
Last edited:
Well if you can't actually test fan RPM, there are still 3 simple things you can test;

1. No voltage at fan terminals (PSU or FUSE failure)
2. Fan not drawing enough current (fan stopped, open circuit failure)
3. Fan drawing too much current (fan stalled, siezed, cooked, shorted etc)

You can probably refine that to just one transistor and a couple of resistors per test.
 
Well if you can't actually test fan RPM, there are still 3 simple things you can test;

1. No voltage at fan terminals (PSU or FUSE failure)
2. Fan not drawing enough current (fan stopped, open circuit failure)
3. Fan drawing too much current (fan stalled, siezed, cooked, shorted etc)

You can probably refine that to just one transistor and a couple of resistors per test.

using above parameters, it will sure make it easier to make the circuit..
 
Well if you can't actually test fan RPM, there are still 3 simple things you can test;

1. No voltage at fan terminals (PSU or FUSE failure)
2. Fan not drawing enough current (fan stopped, open circuit failure)
3. Fan drawing too much current (fan stalled, siezed, cooked, shorted etc)

You can probably refine that to just one transistor and a couple of resistors per test.

I hadn't actually gone as far as considering three states of failure !

So how would this work ?
There wont be much availability for continued testing - the tests would have to be hardwired and permanent

I was originally thinking a simple - fan busted no voltage - would of been fine after reading this I am not so sure.
 
Last edited:
Mr RB nailed it as far as the test criteria. Here is a new one: If the fan is place where there is lots of ambient light, mount a photodetector so that the light falling on it is modulated by the moving fan blades. Use an AC-coupled amplifier, band-passed to the rotation rate, followed by a diode detector to indicate that the fan is turning at near the right speed. If there is not enough ambient light, you might have to mount a light source on the other side of the blades. You need the bandpass filter to eliminate the 100/120Hz flicker if 50/60Hz powered artificial lighting is used.

afterthought: use an electret mic mounted close to blades to detect the turbulence created by the moving blades. You should get pulses at the blade frequency...
 
Last edited:
Showing red and green is not the problem (though I would like to get it down to just one LED.)...showing when the fan has broken down is.
Oh sorry...I get it .... but having vanes on the fans will be a pain (impossible in one case.) where they are situated...I'd rather not. These aren't big fans I'm talking about only 12cm.

You can use a single bicolor LED.

The vane in front of the fan can be a simple 1cm x 1cm brass foil exposed to the airflow hinged on an horizontal shaft. If you imagined something complex, back-up.
Like a rigid flag on a horizontal pole.
If you want to get fancier, optocouplers can be used too. A moving fan will pass intermittently a beam trough its spinning blades. (Or reflect from a shiny adhered whatever to one blade) That pulsing can be detected.
Pulsing = green led on; no pulsing = red on.

Miguel
 
Mr RB nailed it as far as the test criteria.

Well I hope Mr RB comes back otherwise I am totally clueless and I'll have to give up....saying the tests can be reduced to to a transister and a couple of resisters doesn't help me at all.

Vanes, optocouplers and mics - basically anything external to the fan are no go....as I intended to originally state I need something to sense voltage or current. Though after Mr RB's post I realize it might not be that simple.

It looks like I'll have to give up using a single bicolor in my circuit as well....


Well thanks for the suggestions so far anyway...
 
Last edited:
For using a single transistor to turn a LED on when a current gets over a set level see here;
Motor over-current sensor

and you don't need R3 so its 3 resistors a trimpot and a transistor.

You can start with 2 of those circuits, one will light the green LED if the fan is drawing normal current, the other circuit lights the red LED if it draws too much current.

Another transistor can turn off the green LED if the red LED is lit, so the fault condition gets priority.

That covers all your needs, if the power fails (or fan is open circuit) no LEDs will be lit. If running ok the green LED will be lit. If overcurrent fault (stall etc) the red LED will be lit.
 
For using a single transistor to turn a LED on when a current gets over a set level see here;
Motor over-current sensor

and you don't need R3 so its 3 resistors a trimpot and a transistor.

You can start with 2 of those circuits, one will light the green LED if the fan is drawing normal current, the other circuit lights the red LED if it draws too much current.

Another transistor can turn off the green LED if the red LED is lit, so the fault condition gets priority.

That covers all your needs, if the power fails (or fan is open circuit) no LEDs will be lit. If running ok the green LED will be lit. If overcurrent fault (stall etc) the red LED will be lit.

Thank you very much a.) for pointing out my lack of understanding on motor faults and b.) linking to a suitable test circuit.

Luck would have it I have a bunch of 2N2222 and two trimpots so I can started just need a few more pots.

Many thanks.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top