EEVblog #392 – 555 LED PWM HackPosted on November 29th, 2012 9 comments
555 T-Shirt HERE
Hacking the Mantis Elite microscope to control the LED brightness with a 555 PWM circuit.
Forum Topic HERE
This dimmer should work as well connected between the power supply and the DC jack in the arm. Then you could mount it in a nice enclosure, so it would have a look of a quick and rough hack and, more important, you wouldn’t need to touch a single screw on the microscope head.
Yea, great idea !
A further noise maker in your lab. Beware of this squarewave transmitter. I have LED lights working in my own lab benchtop illumination and I prefer a self-made analog current control solution even if the power dissipation is much higher. If you want to make low-level measurements in the sub- millivolt range this is a must.
Thank you Dave
CFL bulbs also are a nightmare in that context.
My thoughts exactly.
But I’d use the already present LM317
Just a OneHungLow 10R 4W pot in series with the feedback resistor will give a usable dimming range. (and I’d add in a couple of ceramic caps for good measure, so RF won’t throw the LM317 out of whack!)
I’d guess that, deep down, Dave would also prefer this solution, but then he couldn’t throw in a PWM dimmer tutorial for more views.
There are a number of ways this can be fixed/ameleorated:
* Add an inductor in series with the LEDs to give amoother current flow. Optionally also add a cap to form an LC filter, effectively making this a buck regulator without feedback.
* Add a snubber (series resistor/capacitor) from the output to ground to reduce dV/dt.
* Shield the dimmer (and the wiring, if it isn’t already).
* Increase the oscillator frequency so that it’s well above the bandwidth of the measurement devices it’s most likely to interfere with. However this may also increase the overall amount of EMI.
* Change the circuit so the frequency/duty cycle varies over time, spreading the EMI spectrum out.
All sorts of high-frequency, sensitive equipment (such as the spectrum analyser Dave tore down recently) contain switching circuits and switchmode regulators and they manage to avoid serious interference. So clearly it’s possible.
These switching circuits are evil. Believe it or your life will end in the hell of digital noise.
Interesting to reflect (as those of us with that sort of mind do) that this collection of analog bits (and the nasty switching spike generated by a regular 555) could all be reduced to just three parts with a cheaper 8-pin ATtiny or PIC chip. (Variable R, the chip, plus a FET, I reckon)
On the other hand, it’s a darn sight more difficult to push an Agilent 1GHz scope probe into the guts of a micro to show it working, and arguably of limited appeal, while probing around the 555 circuit will reveal much of interest.
Just a thought….
#392 is one of my favorite videos of all since I started watching your blog, Dave. I luv it! Awesome. Great job.
Leave a reply