Looking for a calc that will solve for R1, R2, C in astable 50% DC with entering desired freq (1KHZ)and DC(50%). All the calcs I find require entering the component values to solve for freq and DC. Pain. Thanks JR
--
--------------------------------------------------------------
Home Page: http://www.seanet.com/~jasonrnorth
You can get a 50% DC with a cap from the trigger and threshold pins to ground and a single resistor to the output. You need the CMOS version of the 555 for an exact 50% DC. The regular version output doesn't swing all the way between the rails, so there will be a small error.
--
You can do the same thing with a bipolar 555, but you have to make
sure that the supply voltage is high enough and the load current is
low enough that the output swing will get you above 2/3Vcc and below
1/3Vcc so the TRIGGER and THRESHOLD inputs will work.
I guess it's just from all of these billions and billions of years of experience. ;-) Or maybe because I had tried it once, so I already knew the answer, and was just tweaking you. ;-)
I was seriously curious how the program would deal with it, and I guess it's another example of why computers will never replace people. ;-) ;-)
One time, I think (it's kinda lost in the mists of time) I actually used two resistors with a diode in series with one of them in place of Rb, and got pretty durned close to 50%, but I like the one-resistor, schmitt- triggeroid version more, I think. :-)
The output must move all the way to +Vcc for a 50% DC. Otherwise, the cap charge time is greater toward 2/3 Vcc and less on the way down, and therefore not equal. From the 555 spec sheet, I see the output moves to
+13.3 with a 15 volt supply and 100mA load. No data for lighter loads. But looking at the schematic, the high level is controlled by a emitter follower, and therefore the high level output will always be less than the supply voltage regardless of load. So, it doesn't look like the DC can ever be exactly 50% using the bipolar 555 with one resistor and cap.
The standard 555 astable circuit will only give you duty cycles above
50%. But you CAN get a 50% duty cycle with a simple mod to the circuit. Put a diode in PARALLEL with the resistor that connects pin 6 and pin
Connect the cathode to pin 6, anode to pin 7. With that diode in place, you can make the low duty cycle as low as you want. Using R1 and R2 equal should get you 50% duty cycle. As a side note, you can make a fixed-frequency astable circuit with continuously variable duty cycle by using a pot; connect the ends of the pot to pins 6 and 8, and the wiper to pin 7, while keeping the diode I mentioned previously.
--
You\'re right, of course.
My thinking was that as long as the voltage on the cap oscillated
between 2/3 Vcc and 1/3 Vcc the charge time and discharge time
would be the same, but since the voltages sourcing and sinking
current for the cap are different, the times _can\'t_ be the same,
Duh...
Thanks,
--
Well, let\'s see...
For way less than a dollar you get an astable or a monostable
multivibrator, a couple of comparators with a nice internal pretty
much isothermal voltage divider that you can mess with if you want
to, a totem pole output that you can get a couple hundred milliamps
out of, an open collector NPN, and a RESET input.
And... you can get them anywhere, everybody makes them, and they\'re
not likely to stop making them anytime soon.
And... you can get it in CMOS.
Crap you say? LOL!
1KHz is 1millisecond making the charge and discharge times 0.72/F or about 720us. Smaller C is better and R in megohms is about the upper limit for the 555, so C somewhere around 720us/megohms gives somewhere around 720pf. 720pf is almost a nf which is 0.001uf a nice standard value. So make C=0.001uf. Then R has to be like 720K. If R1 is the Vcc-connected resistor and R2 is the other one, maybe making R1=39K and R2=680K will do the trick. The charge time is
0.69*(680+39)K*0.001u=496us, and the discharge time is
0.69*680K*0.001u=469us. The frequency is then 1/(496u+469u)=1.036KHz and the duty is 496/(496+469)=0.513. Both of these numbers are nearly one half stock tolerance of R's,C's, and the 555 itself. If you're not worried about a little current, then make R1=10K and R2=720K. This makes charge time 0.69*730K*0.001u=504us and discharge 0.69*720K*0.001u=497us for a frequency of 1.001KHz and duty of 0.503. Close enough yet?
Rich Grise wrote in news: snipped-for-privacy@example.net:
I haven't used 555's much actually. :) I'd have to go back to the data sheet and some experimentation just to contribute specifically to the thread. Like Budgie, I don't like them much, they seem very loose for a standard, not an approach to an ideal the way the average op-amp is. I did use a 556 for a servo tester once though, it was helpful for that, but even then some simple means of direct voltage control would have been nice.
About programs, I agree. While good programming can solve things like this value predictor, there are limits to models. I never use SPICE modelling, I think electronics is as much art as science, and if I want that much detail I explore the real component's's oddities to see what helps and what doesn't. I actually do non-recommended things with op-amps at times when they work in the context I want, and I suspect a model wouldn't do this for me even if I could afford that kind of expensive software.
Chain curves have caused problems for mechanics computer design, and there's recently been work on imaging real ones to speed up the computer modelling. Likewise, modelling a bass guitar is harder than learning to play bass...
I suppose the more intelligent way to use that calculator is to enter duty cycles that are arbitarily close to and above but not exactly equal
50%. Of course, you wouldn't know to do this if all you've ever done is use calculators. Entering 50.3% , 1KHz, and C=0.001u, values obtained with some fairly rough insight into 555 operation, returns RA=8.64K and RB=716K- not std values according to my other formula derived mysteriously from additional applications insight, almost as good as, but a lot less fun than, the manual method.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.