Hi,
I am having the following problem. I need to create a variable frequency variable duty cycle 5V square wave signal. The frequency and pulse width need to be adjusted by pots or something of that user friendliness. My current design incorporates 2 555-timer ICs one run in astable mode to get my frequency that I want and the second run in monostable mode for pulse shaping. My problem is with the pulse shaping design. I have it being triggered by the first 555 and it works, however, my frequency range 1-50Hz requires the pot on the duty cycle adjustment to be astronomically large to achieve 3-97% duty cycle at each frequency. I need some IC that accepts the frequency signal from the first IC and creates the duty cycle by a smaller adjustment independent of frequency. I believe I can do the whole circuit with a MSP430 microcontroller, however, I need two designs and if there is an IC out there that does pulse shaping for this application it would be nice.

Thanks for your time
Chris

 

Looks like you need about a 5000:1 range on the second monostable which would be do-able, but impossibly sensitive at one end of the pot range.
How about using a log pot instead of linear?

 

I could but that makes the duty cycle too sensative at higher frequencies.

 

Try this:

Generate triangle waves at the frequency you require, (I think you may find these on the timing capacitor of the 555).
Feed the triangle wave into a comparator.
Make the camparator reference voltage adjustable with a pot.
The output of the comparator will have a duty cycle variable from 0 to 100% determined in a linear manner from the position of the reference voltage pot.

JimB

__________________
Experience is directly proportional to the value of the equipment ruined.

 

That may work. I thought of that initially but only can deal with a small delay (only a couple of microseconds). Specific Timing is a must therefore comparator response time must be minimal. Do you have any Recommendations for a good Op Amp to achieve this?

 

An LM393 has a nominal delay of 300ns. A little hysteresis would be in order.

__________________
Ron

 

I'm not experienced enough to say whether this circuit really suits what you're after, but I've had good results with it. It's the two-diode circuit second from the bottom of the image:

http://home.cogeco.ca/~rpaisley4/LM555.html#4

(It's a long page...)

Anyway I just built one of these last night for a slow-time fountain project and it works great. I'm using R1 = 470R, R2a = R2b = 10K trimpot, and C1 = 4.7uF. I'm still playing with the values to give the most usable range but these ones are in the ballpark for me.

Anyway, again I'm not sure this is what you want, but it might be worth a look.


Torben

 

Would you consider a different approach using a "function generator" ic? For example, the ICL8038 from Intersil or the Maxim MAX 038.

__________________
RadioRon