How to wire a potentiometer for use dimming multiple LEDs?

Thread Starter

birckcmi

Joined Jan 1, 2018
210
Final results to the original question: How to wire a potentiometer for use in adjusting brightness of an LED tracing/light panel? It boiled down to 4 suggested circuits, schematics attached. Circuit A, using a 10k pot-the panel illuminated at about 70-80% brightness, but the pot was unable to vary the brightness, when powered by a 5VDC wall-wart. When powered by a variable DC supply, that variability of voltage did the trick, but there was no variability when using the pot.
Circuit B, using a TO-222 NPN voltage regulator gave 2 results: omitting the bypass loop on the right, with a switch, gave no illumination, with or without the 10-ohm resistor; when I added the bypass, the panel came on, but no variation in brightness was attainable. That suggests that the circuit was simply ignoring the IC and using the bypass as a bypass. I should add that the person who posted this schematic later regretted it because of an error he had made. I included it anyway, to see if it could accomplish anything.
Circuit D, frankly, did nothing at all. I will try mocking it up again. It looks as if something should happen. Stay tuned.
There were two successful approaches: using a pulse-wave modulator, and Circuit C, using two pots to achieve a complete range of brightness. I think Bernard suggested it and supplied a schematic. It worked. I will build that circuit. Even though the PWM system worked, as I expected it to, I'm glad it worked, but I want to learn more about circuitry, so I'm sticking with the assembly in fig C. Thanks to all of you for taking the question seriously so that I can get this light panel running again.4-up circuits annotated.jpg
 

ElectricSpidey

Joined Dec 2, 2017
2,758
The only error I made was ignoring the fact that the strips were 5 volts, and assumed they were 12.

My circuit is tried and proven over and over again as I have used it many times.

I would suggest if you want to try it again...

1. Remove the bypass switch.
2. Use the 6 volt supply.
3. Place a resistor between the ground pin of the pot and the supply ground, to change the range from 0 to 6 volts to about 2.5 to 6 volts.

If it doesn't work...well there could be plenty of reasons for that, but it's not an inherent circuit problem.

I'm not trying to convince to use the follower circuit, just as a learning experience.
 

bassbindevil

Joined Jan 23, 2014
824
TLDR, but why not get a PWM dimmer module from China, or make one using a 555 timer chip and a transistor (which is exactly what the $1.35 PWM modules use)?

But after a closer look at your schematics: B should work better if the LED string is in the collector side of the transistor (which should probably be a TO220 package power transistor, possibly with a heat sink). That way the transistor can turn on 100%.

Yet another option: investigate the world of flashlight drivers that use ATtiny micros to control constant-current driver chips via PWM. Things like the NANJG AK47 or 105C can be reprogrammed (need an IC clip and USBasp AVR programmer) , but come with a decent choice of built-in modes (selected by soldering a jumper to a star-shaped pad). If you get the programming gear, there's plenty of firmware written for these flashlight drivers, some with multi-level menus (check out budget light forum, candlepower forums).

Most drivers found in cheap LED lights have annoying modes like high/low/strobe/SOS, but I have come across an LED desk lamp with a touch-controlled driver with a useful range like low/med/high, and as a bonus, can (if I identified the chip correctly) be reconfigured (by jumpering some pins) to do continuously adjustable brightness. This would the Voltax desk lamp sold by Dollarama, the one with a 16-LED COB.
 

ElectricSpidey

Joined Dec 2, 2017
2,758
bassbindevil if you place the LEDs on the collector the circuit would no longer be a follower, and the pot range would be basically ON/OFF. (very small variable range)
 
Like many other home appliances, potentiomenter is used to link the led directly to the potentiometer and fade / dim / control.

By using a variable resistor, either dim or brighten the current can be regulated at will. Potentiometers have three cables for connecting to a circuit and a press.

I was trying to wire the positive dimming leads of all 3 drivers to one wago. The maximum value of the HLG240 for a single potentiometer is 248w. If you are connecting 2 drivers together, go for a 50k pot and a 5k resistor.
 

Thread Starter

birckcmi

Joined Jan 1, 2018
210
The only error I made was ignoring the fact that the strips were 5 volts, and assumed they were 12.

My circuit is tried and proven over and over again as I have used it many times.

I would suggest if you want to try it again...

1. Remove the bypass switch.
2. Use the 6 volt supply.
3. Place a resistor between the ground pin of the pot and the supply ground, to change the range from 0 to 6 volts to about 2.5 to 6 volts.

If it doesn't work...well there could be plenty of reasons for that, but it's not an inherent circuit problem.

I'm not trying to convince to use the follower circuit, just as a learning experience.
Thanks, ElectricSpidey. I believe you. I will try it again. "there could be plenty of reasons for that" indeed, starting with my lack of finesse with a breadboard. Since I now have sufficient parts to put together a few separate soldered versions of the circuit, I'll try it that way. The version of Bernard's circuit that worked well was done that way. BTW, I'm clear on which pin on the pot is ground, but the power supply only has two leads-so the ground would be the negative lead? Isn't that the way it's wired now?
 
Last edited:

Thread Starter

birckcmi

Joined Jan 1, 2018
210
TLDR, but why not get a PWM dimmer module from China, or make one using a 555 timer chip and a transistor (which is exactly what the $1.35 PWM modules use)?

But after a closer look at your schematics: B should work better if the LED string is in the collector side of the transistor (which should probably be a TO220 package power transistor, possibly with a heat sink). That way the transistor can turn on 100%.

Yet another option: investigate the world of flashlight drivers that use ATtiny micros to control constant-current driver chips via PWM. Things like the NANJG AK47 or 105C can be reprogrammed (need an IC clip and USBasp AVR programmer) , but come with a decent choice of built-in modes (selected by soldering a jumper to a star-shaped pad). If you get the programming gear, there's plenty of firmware written for these flashlight drivers, some with multi-level menus (check out budget light forum, candlepower forums).

Most drivers found in cheap LED lights have annoying modes like high/low/strobe/SOS, but I have come across an LED desk lamp with a touch-controlled driver with a useful range like low/med/high, and as a bonus, can (if I identified the chip correctly) be reconfigured (by jumpering some pins) to do continuously adjustable brightness. This would the Voltax desk lamp sold by Dollarama, the one with a 16-LED COB.
Bassbindevil, thanks for your suggestion. I screwed up identifying the transistor. It is, in fact, not a TO-222, but a 220. My apologies. And it is mounted on a heat sink that does a good job. And thanks, but I am not enough of an EE to know if a driver designed for one or two flashlight LEDs at unknown voltage will work as well controlling a ladder of 150-200 LEDs at 5 -6 volts as the circuits suggested here. At least one of these works perfectly, and I suspect that others will as well, as my prototyping skills improve.
 
Last edited:

Thread Starter

birckcmi

Joined Jan 1, 2018
210
Like many other home appliances, potentiomenter is used to link the led directly to the potentiometer and fade / dim / control.

By using a variable resistor, either dim or brighten the current can be regulated at will. Potentiometers have three cables for connecting to a circuit and a press.

I was trying to wire the positive dimming leads of all 3 drivers to one wago. The maximum value of the HLG240 for a single potentiometer is 248w. If you are connecting 2 drivers together, go for a 50k pot and a 5k resistor.
Thanks for the suggestion. When I first read it, I didn't even recognize the name "Wago". I looked into it, and they look like something I could use. At the moment, my circuits are pretty small and simple-I would be surprised if the power got up to 5W. And I'm trying to wind up with a small, pocket-sized controller for a desktop 5VDC light panel. An HLG240 would drive one of my ceiling lights, but it's overkill for what I'm building. But thanks for your comment.
 

Thread Starter

birckcmi

Joined Jan 1, 2018
210
The only error I made was ignoring the fact that the strips were 5 volts, and assumed they were 12.

My circuit is tried and proven over and over again as I have used it many times.

I would suggest if you want to try it again...

1. Remove the bypass switch.
2. Use the 6 volt supply.
3. Place a resistor between the ground pin of the pot and the supply ground, to change the range from 0 to 6 volts to about 2.5 to 6 volts.

If it doesn't work...well there could be plenty of reasons for that, but it's not an inherent circuit problem.

I'm not trying to convince to use the follower circuit, just as a learning experience.
I think I get this. A Follower Circuit is designed to function as a voltage buffer, keeping the voltage at or below its max? "Supply Ground" is...what? The negative pole? Any size resistor? ohms, K-ohms, or meg-ohms? And I appreciate the learning experience. I'm putting the circuit together for real on a perf-board.
 

ElectricSpidey

Joined Dec 2, 2017
2,758
Yes, a voltage buffer with current gain.

Here is a new schematic, I left the mystery resistor up to you to figure out.;) (under 10k)

The additional resistor should be chosen to make the range about 2.5 to 6 volts. (or whatever voltage the LEDs start to light)

A_Emitter_Follower_2.jpg
 

Thread Starter

birckcmi

Joined Jan 1, 2018
210
Yes, a voltage buffer with current gain.

Here is a new schematic, I left the mystery resistor up to you to figure out.;) (under 10k)

The additional resistor should be chosen to make the range about 2.5 to 6 volts. (or whatever voltage the LEDs start to light)

View attachment 221797
I tried it again. This time I treated the circuit as a voltage divider, and found an Ohm's Law formula that purported to give me a value for R2 (the lower resistor). I came up with .714 ohms. By the attached formula. Since it's less than 1 ohm, I figured it had a chance.Formula.jpg
 

ElectricSpidey

Joined Dec 2, 2017
2,758
.714Ω as the lower resistor will do nothing.

Putting 10k at the lower will yield 3 volts at the center of the divider. (assuming the upper set @ 10k.

This is close to where you want to be, so the range on the LEDs ends up being about 2.3 to 5.3 volts.

Then you can raise or lower the lower resistor to get the lowest level you want.
 

Thread Starter

birckcmi

Joined Jan 1, 2018
210
Thanks for the alert about the pinout. That could be the reason I haven't got any results. On the other issue, I used a formula under the "voltage divider" heading to come up with the fact (?) that if I put a 5K resistor in the R2 position, V out will be 2 volts-pretty close to what you suggested (2.5 volts). Sound right? the formula was worked out to deliver a value for V out, but I estimated an R2, and it looks OK:
Voltage equation.jpg
Did you check your mail...I think you may have a problem with the NPN pinout.
PS- Re the pinout: I'll try turning the TO-220 180˚ from its current orientation.
 
Last edited:
Top