# Need help : How to change phase of a 90* phased square wave output

Discussion in 'General Electronics Chat' started by chihiro, May 18, 2008.

1. ### chihiro Thread Starter Member

May 18, 2008
22
0
hello everyone, i am working on a simple project and trying to obtain some info: here s a circuit that produces two square waves with phases 90* apart..

now my aim is to produce 3 waves with 120* phases apart, and i dont have enough knowledge to construct, can anyone help me with this issue?

note: i don't want an already-built circuit, some info and hints will be very useful for me,, thanks

2. ### Papabravo Expert

Feb 24, 2006
10,340
1,850
I don't think there is a general purpose analog solution to producing square waves with a 120* phase difference. It would be a great deal easier and more precise in the digital domain.

3. ### chihiro Thread Starter Member

May 18, 2008
22
0
i am a second year student in my dep. , actually i will have analog 1, and digital 1 in my next year, thanks for the help , but my main concern is about the simple logic of the dc to three phase inverter's process, in other words i need basic knowledge, i read from a pdf that it's not possible to create 3 phased stable sinusoidal outputs, and im trying to find a way to modify the circuit above with parallel connected capacitors to obtain 120* phased square waves, can anyone show me a way to begin ? and shortly how it will work...

for example, last term i worked on a bouncing ball in DSO, used integrator to obtain cosine wave from sine, and summer to get a ball, and another summer to sum ramp function and that ball, and i finally got a bouncing ball, it was quite easy but in this project i need a pattern to follow, but don't know how to begin

4. ### Papabravo Expert

Feb 24, 2006
10,340
1,850
Finding a circuit with a stable 120* phase shift may not be possible. If a circuit exists with that phase shift it will occur only in a narrow frequency range. Any small disturbance like temperature or supply voltage will cause the frequency to shift and the phase relationship will also shift. Your best chance is to do this in the digital domain.

5. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Papabravo's correct - you'll wind up with inaccuracies and frequency drift with an analog solution.

However, that might not be important for your application - it just depends on what you want to do.

You can build a simple 3-phase square wave generator using 1/2 of a CMOS 40106 IC, which is a Schmitt-trigger hex inverter, three resistors and three capacitors. It won't be super accurate, but it'll work.

If you want a spoiler, there's a similar circuit attached using 3 out of four NAND gates in a 4093. The nice thing about using the 4093 or 40106 is the very low parts count.

You can get an arbitrary number of offset square waves by simply adding on more gates, resistors and capacitors.

• ###### 3-phase4093CMOS.PNG
File size:
61.6 KB
Views:
74
Last edited: May 18, 2008
6. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
Another way is to have a 2 bit counter, and generate a square wave using gates. Easy to do, and precision besides. You need a clock frequency 4X the square wave freq. If you're interested in this approach ask, and I'll draw up a schematic. Not counting the ocsillator it is a 2 chip circuit.

7. ### chihiro Thread Starter Member

May 18, 2008
22
0
anyway i designed a square output circuit but i need a current of 0.15A but i just have 1 nA :S can u help me with this, how to amplify that current to 0.15A. A friend of mine told me to use transistor to amplify but don't know how.

8. ### chihiro Thread Starter Member

May 18, 2008
22
0
btw sorry for the size :S

9. ### chihiro Thread Starter Member

May 18, 2008
22
0
i just added a bjt to amplify that current, and got 1.772uA, still too low

May 16, 2005
5,072
6
11. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Look up "emitter follower" or "common collector".

If you put a PNP and NPN emitter follower together by their emitters, you'll have something interesting.

12. ### chihiro Thread Starter Member

May 18, 2008
22
0
i finally got the correct current by using double 2n2222 for each opamp circuit parts, now preparing my report, what i want to ask is how is the phase shifted by opamp and capacitor combination ? i successfully designed the circuit used amplifiers but just don't know the process that how the phase of square waves are shifted by 120* with each opamp.

13. ### chihiro Thread Starter Member

May 18, 2008
22
0
btw i searched on google as "phase shift oscillator", but couldnt find a page that describes the situation

14. ### chihiro Thread Starter Member

May 18, 2008
22
0
wow i read my 1.5 year old post now its so funny and shocking to see how ignorant i was, felt so good when u guys tried to help with all effort, thanks guys its a kind of you

15. ### BillB3857 Senior Member

Feb 28, 2009
2,402
348
If you wanted to do it with digital logic, try this. 120 degree shift at any frequency.

File size:
5.4 KB
Views:
32