# Question for the clever AC experts

Discussion in 'The Projects Forum' started by Denesius, Feb 11, 2014.

1. ### Denesius Thread Starter Member

Feb 5, 2014
89
14
Folks: I have a sine wave output at 400hz. How do I calculate the parameters for an LC circuit to shift the output 120 & 240deg? The output is driving a transistor stage, so current is minimal (and I don't believe it effects the shift, anyway). If you haven't guessed, I'm converting a single phase source into a 3 phase output. The sine wave origin cannot be triggered reliably to obtain a consistent shift. Any pointers would be appreciated, or an alternative means of obtaining the end result. Thanks.

2. ### Papabravo Expert

Feb 24, 2006
11,163
2,187
What makes you think this technique will work to achieve the desired result? Come to think of it, what are you actually trying to do? You say the output is a transistor stage with minimal current. Why is this not shown in your drawing?

3. ### alfacliff Well-Known Member

Dec 13, 2013
2,449
429
the problem with this is that you cant get the 120 degree whift with either lc or rc, the max is 90 degree.

4. ### crutschow Expert

Mar 14, 2008
16,575
4,477
You can use two op amps connected as all-pass phase-shifters to shift the phase 120 degrees and 240 (-60°) degrees. The two types are adjustable in phase between 0 and 180° or 0 and -180° respectively by using a trimpot for R.

5. ### Papabravo Expert

Feb 24, 2006
11,163
2,187
I don't know of many 3-phase sources that have output sages made from opamps. I'd like to see you run a 3-phase motor with one -- LOL

6. ### crutschow Expert

Mar 14, 2008
16,575
4,477
He did state the output was a transistor stage with minimal current..

7. ### Denesius Thread Starter Member

Feb 5, 2014
89
14
ALPHACLIFF- I wondered. I recall reading somewhere that an R-L (with the inductive reactance comparable to the resistance), with a parallel C circuit would shift current cycle 120-deg in relation to the voltage cycle- and the current cycle can be used, in turn, to induce a voltage that is in turn 120-deg out of phase. If I'm barking up the wrong tree, let me know!

8. ### Denesius Thread Starter Member

Feb 5, 2014
89
14
PAPABRAVO- If the cycles are there, and in proper relations, I can always boost the power up to the point where I can drive anything. At this point, however, this is an attempt to obtain the results as simply as possible. I'll try the op-amp approach this weekend & post the results.

9. ### Papabravo Expert

Feb 24, 2006
11,163
2,187
There was some ambiguity in that statement. I thought it might be the output of the generator, but I wan't sure.

Still -- you can't just generate the waveforms and then "boost the power". It is just not practical or efficient to do it that way, but nevermind you're having fun with this.

10. ### GopherT AAC Fanatic!

Nov 23, 2012
7,292
6,032
Are we guessing what the goal is here because it certainly is not clear so far?

My guess is that the OP has a 400 Hz military generator and wants to make it 3 phase.

Phase converts are available for that if needed (static or rotary models).

11. ### Denesius Thread Starter Member

Feb 5, 2014
89
14
Actually, I'm starting with a DC power supply & am using an NTE864 to generate the 400hz AC. I'd like to see if I can convert to 3-phase to operate 3 coils of a tuning circuit. The power needed is in the 1-2VA range (at 12v), so commercial inverters/phase converters are impractical. While the current draw is minimal, the distortion of the sine wave & the phase relations are critical. I looked into the "all pass filter" mentioned earlier, and if I can work out the math, I think it'll get things done. Personally, I had never heard of such a thing (it initially sounded like a joke), and it reminds one of why these forums are so useful.

12. ### crutschow Expert

Mar 14, 2008
16,575
4,477
Below is a simulation of the all-pass op amp phase shift circuit. You can replace R1 and R2 with pots to compensate for component tolerances and tweak the phase-shift to the exact value.

Of course the R's and C's can be changed as long as the RC time-constant is the same.

Last edited: Feb 12, 2014
Denesius likes this.
13. ### Papabravo Expert

Feb 24, 2006
11,163
2,187
So when the three tuning coils are excited from the 400 Hz generator what will exactly will you be tuning?

14. ### Denesius Thread Starter Member

Feb 5, 2014
89
14
I did the math & came up with something similar, using 4700pF caps & 5-turn 1K5 pots for R1 & R2. Doing the math with 1042.8 ohms, the shifts were +/-120.033deg, close enough! Now to put it all together & note stability over time & temp change....

15. ### crutschow Expert

Mar 14, 2008
16,575
4,477
If you want temperature stability then the resistors and capacitors need to have a low temperature coefficient. In particular the capacitors should be high stability types like NPO ceramic.