Originally posted by larryross
View Post
The facts mentioned below are not meant as criticism at all but should forward the notion that - depending on the actual circuit - it might give no success to fight against nonlinear behaviour of the duty cycle. Total simplicity suffers on certain drawbacks but I think we can live with this and it will not prevent any initial experiment.
I experienced such nonlinear behaviour if the circuit charges the corresponding timing cap via resistors. Then the voltage is not of true linear sawtooth shape but is bent depending on the value of the charging resistor (shark fin shape). So the following comparator chanages the duty cycle because it senses voltage and not time.
This behaviour can be minimized if the charge discharge rate is low (i.e. 0V....5V) while the charging resistor is being connected to a considerably higher regulated voltage i.e. 24V. In this case the resistor acts like a current source of moderate quality. (We make use of the mostly linear bottom part of the shark fin only.) The current simple circuits in this forum try to make use of high voltage at the timing capacitors being almost alike the supply voltage. This is the source of the effect mentioned above.
Exact this nonlinear behaviour was defended thoroughly by applying a simple current source (see my doc) for charging the timing capswhile running completely at 5V supply.
Please bear with me! Simple approaches are entirely SIMPLE BUT NOT USELESS.
rgds John
probably it would handle it The schematic is here

Grrr


Leave a comment: