Interesting LED dimmer circuit

oracacle

Senior Member
This isn't directly connected to picaxe, but thought that some here would be interested.

I have just purchased a white LED strip with dimmer. I don't need the housing on the dimmer and while removing it thought I would take a closer look at the circuit.
It wasn't long before I started to get quite intrigued by how the 555 was being used in combination with an op-amp in comparator setup.
the higher the voltage going into the non inverting input the higher the duty cycle.
The traces show the saw tooth signal from pin2 on the 555, while the other trace shows the signal coming out of the op-amp.

Please note the MOSFET shown in the schematic isn't the one the circuit, just a stand in from what I had in my eagle library. The actual component is 090N03L, which appears to be a power MOSFET.
 

Attachments

tmfkam

Senior Member
That is quite a novel way of creating PWM, with a 555 just to generate a ramp.

As the 555 can be configured to generate PWM which could drive the mosfet without needing the 358 you wonder what the thinking was when designing the circuit. Could it be more efficient with the 358?
 

oracacle

Senior Member
I suspect that it maybe be down to the mosfet itself.
driving a mosfet at high frequencies can require extra hardware to drive correctly. If you have a frequency that sits in the best switching area for the mosfet and then adjust the on period to make the LEDs brighter.

I do wonder if something like this could be used to control a heating element with the use of a thermostat.
 

cpedw

Senior Member
I have had no difficulty using picaxe PWM applied to the gate of either 2n7000 or STP36NF06L via 330ohm with 12k to earth to give adjustable LED brightness. Currently I'm working at about 50kHz but that was selected mainly to achieve a wide range of duty with fine adjustment; I wouldn't know how to assess the suitability of the frequency for the MOSFET.
 

inglewoodpete

Senior Member
Currently I'm working at about 50kHz but that was selected mainly to achieve a wide range of duty with fine adjustment; I wouldn't know how to assess the suitability of the frequency for the MOSFET.
At the risk of straying off the original topic, the STP36NF06L has an input capacitance of 660pF. Without a MOSFET driver, you will find that if will not respond very well to PWM frequencies over about 2kHz. For a linear response driving LEDs, I don't drive large MOSFETs at over 1kHz; lower if the PICAXE can do it. You can get 10-bit PWM out of a PICAXE at that frequency.
 

Pongo

Senior Member
As the 555 can be configured to generate PWM which could drive the mosfet without needing the 358 you wonder what the thinking was when designing the circuit. Could it be more efficient with the 358?
The totem pole output of the 555 is pretty good current sink, but a much weaker current source. I wouldn't consider it suitable to drive a power FET directly.
 

tmfkam

Senior Member
The totem pole output of the 555 is pretty good current sink, but a much weaker current source. I wouldn't consider it suitable to drive a power FET directly.
The LM358 is not a device that would automatically come to mind as a power mosfet driver. I presume the output stage of the 358 is a standard class AB push pull one using bipolar transistors, fairly similar to the 555? A 555 is capable of directly driving a mosfet at reasonable frequencies of a few hundred Hertz as an LED dimmer. I've built a number of them for use at home and work. I can't vouch for the ultimate efficiency of my designs, but they have been reliable.
 

fernando_g

Senior Member
Indeed, this whole circuit could be replaced with a single Picaxe 8M2, a pot and some resistors, driving a logic level Mosfet.
Or if you want to drive a larger Mosfet, a suitable low-side gate driver, which are dime a dozen.
 
Top