555 timer PWM circuit

Thread Starter

Boost Creep

Joined Sep 1, 2022
9
Hi All,

I'm looking for a simple circuit to dim a LED dash. It's for a project I'm helping a friend with. He is building a factory 5 cobra.

The new style dash they offer is a full LED backlight and clearly won't dim on a standard rheostat. The limitation I'm working to is that he wants to keep the dash looking original, so he will use the Lucas-style bakelite knobs.

The options are as I see it:

1, Use a potentiometer (modify it to take the Lucas knob); this clearly is not the most efficient circuit. But given its light use, it should be OK.

2, Use a 555 timer to build a PWM circuit.

The question I have is, can I control a 555 timer with a rheostat rather than a potentiometer? Using the original Lucas rheostat will mean I don't have to modify the dash and the overly brittle bakelite knob.
 

ericgibbs

Joined Jan 29, 2010
18,766
hi BC,
I guess you realize you will need a little extra circuitry with the Rheo' to enable it to control the PWM with a 555.
Have you already designed a basic 555 PWM circuit, that you could post?

What current drive will the panel LEDs require at full intensity.?
E

@crutschow @Alec_t @Papabravo
 
Last edited:

Thread Starter

Boost Creep

Joined Sep 1, 2022
9
What current drive will the panel LEDs require at full intensity.?
E
I will need to measure that, but I will get that done and come back to you.

I planned to base it on this circuit, replacing the POT with a rheostat.

I'm unsure what is required circuit-wise to do that.

1662057388869.png
 

crutschow

Joined Mar 14, 2008
34,281
Using a rheostat will change the frequency as well as the duty-cycle.

Using two 555's (or one 556) might be better as one can generate a fixed frequency to trigger the second 555 whose pulse-width is varied by the rheostat.
If that sounds interesting, I can gin up a circuit.
 

crutschow

Joined Mar 14, 2008
34,281
I came up with a circuit using two 555's but it was somewhat complicated, with the PWM duty-cycle versus the 15Ω rheostat being very non-linear with rotation, and I haven't come up with a solution for that.

I do have a circuit that uses a quad LM324 op amp which is much more linear.
A preliminary, proof-of-concept circuit is shown below:

U4 is the 15Ω rheostat.
The left op amp is an astable oscillator which generates a quasi-triangle, and the right op amp acts as a comparator to generate the pulse-width as determined by Mod voltage versus the Tri voltage.
The middle op amp provides gain for the low voltage change generated by the rheostat rotation.
The PWM duty-cycle (bottom plot pane) is shown for the 0% (green trace), 25% (yellow trace), 50% (red trace), 75% (blue trace) , and 100% (purple trace) position of the rheostat.
It's a little hard to follow since all the waveforms are on top of each other, but basically the 0% is fully off, and the 100% if fully on, with the other duty-cycles closely proportional to the rheostat position.

If that will work for you I can add an output buffer transistor to drive the LEDs.

1662086226481.png
 
Last edited:

eetech00

Joined Jun 8, 2013
3,858
Hi All,

I'm looking for a simple circuit to dim a LED dash. It's for a project I'm helping a friend with. He is building a factory 5 cobra.

The new style dash they offer is a full LED backlight and clearly won't dim on a standard rheostat. The limitation I'm working to is that he wants to keep the dash looking original, so he will use the Lucas-style bakelite knobs.

The options are as I see it:

1, Use a potentiometer (modify it to take the Lucas knob); this clearly is not the most efficient circuit. But given its light use, it should be OK.

2, Use a 555 timer to build a PWM circuit.

The question I have is, can I control a 555 timer with a rheostat rather than a potentiometer? Using the original Lucas rheostat will mean I don't have to modify the dash and the overly brittle bakelite knob.
Look for an LED PWM driver IC (automotive type) that has analog dimming control feature.
You could then control the LED current with a 0v-10v signal (for example) via the reostat connected to the analog dimming control pin.
They are very easy to use.
 

Thread Starter

Boost Creep

Joined Sep 1, 2022
9
I came up with a circuit using two 555's but it was somewhat complicated, with the PWM duty-cycle versus the 15Ω rheostat being very non-linear with rotation, and I haven't come up with a solution for that.

I do have a circuit that uses a quad LM324 op amp which is much more linear.
A preliminary, proof-of-concept circuit is shown below:

U4 is the 15Ω rheostat.
The left op amp is an astable oscillator which generates a quasi-triangle, and the right op amp acts as a comparator to generate the pulse-width as determined by Mod voltage versus the Tri voltage.
The middle op amp provides gain for the low voltage change generated by the rheostat rotation.
The PWM duty-cycle (bottom plot pane) is shown for the 0% (green trace), 25% (yellow trace), 50% (red trace), 75% (blue trace) , and 100% (purple trace) position of the rheostat.
It's a little hard to follow since all the waveforms are on top of each other, but basically the 0% is fully off, and the 100% if fully on, with the other duty-cycles closely proportional to the rheostat position.

If that will work for you I can add an output buffer transistor to drive the LEDs.

View attachment 275384
thanks crutschow

I'm going to dive into my parts bin and see what I'm missing to build :)
 

crutschow

Joined Mar 14, 2008
34,281
Here's the simulation of the LM324 circuit which better shows the PWM duty cycle (green trace) as the rheostat resistance goes from 0Ω to 15Ω (blue trace):
I also modified the circuit slightly to eliminate the need for the 9 ohm resistor R8, so the rheostat can be connected directly to ground.

(I didn't mention it before, but the circuit is essentially insensitive to changes in the battery voltage.)

1662176967757.png
 
Last edited:

Thread Starter

Boost Creep

Joined Sep 1, 2022
9
Here's the simulation of the LM324 circuit which better shows the PWM duty cycle (green trace) as the rheostat resistance goes from 0Ω to 15Ω (blue trace):
I also modified the circuit slightly to eliminate the need for the 9 ohm resistor R8, so the rheostat can be connected directly to ground.

(I didn't mention it before, but the circuit is essentially insensitive to changes in the battery voltage.)

View attachment 275516
Thanks so much, crutschow, I have ordered thebits I don't have. I will prototype it and report back.
 

crutschow

Joined Mar 14, 2008
34,281
I don't show the buffer drive circuit (MOSFET) for the LEDs.
How much current do they take?
Do they go to ground or are they connected to the battery?
 

Thread Starter

Boost Creep

Joined Sep 1, 2022
9
I don't show the buffer drive circuit (MOSFET) for the LEDs.
How much current do they take?
Do they go to ground or are they connected to the battery?
The LED backlights pull 115ma at 12v full brightness, I will probably use FQP47P06. I did a quick test with a 10K POT and they seemed to work OK.

https://4donline.ihs.com/images/Vip...8-1.pdf?hkey=6D3A4C79FDBF58556ACFDE234799DDF0

I have a bunch of them laying around.

Connected to the positive from the headlight in the original circuit, so battery.
 

crutschow

Joined Mar 14, 2008
34,281
I will probably use FQP47P06.
That's a P-MOSFET so I assume you want a high-side drive with the LEDs going to ground.
For that to work properly you will need a pull-up resistor (i.e. 2kΩ) between the MOSFET's gate and source terminals.

That will give maximum brightness at a rheostat resistance of zero.
Is that ok?
If not, just reverse the input connections to U1-c.
 

Thread Starter

Boost Creep

Joined Sep 1, 2022
9
That's a P-MOSFET so I assume you want a high-side drive with the LEDs going to ground.
For that to work properly you will need a pull-up resistor (i.e. 2kΩ) between the MOSFET's gate and source terminals.

That will give maximum brightness at a rheostat resistance of zero.
Is that ok?
If not, just reverse the input connections to U1-c.
Yeah, the loom is mostly closed (rapped up), so the LED's going to the ground will mean less alteration. Thanks for the tip on adding the pull-up resistor.
 
Top