120 Degree Phase Shift Osc.

Lord, what an ass you are, and apparently want to be. JA can out-engineer, out-sing, out-cook, and obviously out-think you without effort. And he's nice, while you are a creep.

But I do wish he'd learn to ski a little faster. A lot faster, actually.

John

=A0 =A0...Jim Thompson

No worries--we knew what you meant.

-- Cheers, James Arthur

=A0 =A0 =A0 =A0 =A0Vcc

=A0 =A0 =A0 =A0 =A0 =A0 |

-[DIODE>]-+

=A0 =A0 | =A0 =A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0 =A0 =A0|

ODE>]-+

=A0 =A0 | =A0 =A0 =A0 =A0 =A0|

=A0 =A0 +--[COIL]--+

=A0 =A0 =A0 =A0 =A0 =A0 =A0|

-+-[DIODE>]-+

=A0| =A0 =A0 | =A0 =A0 =A0 =A0 =A0|

=A0 +--[COIL]--+

The 10F200 is what I had in mind when I said $0.50, but JF's version doesn't need any programming, no development system, and is repairable forever. If you don't need sinewaves, that all's mighty attractive.

-- Cheers, James Arthur

Or a 3-bit Johnson counter. ;-)

Cheers! Rich

rote:

n.

from

our

s...

to

ree

ee

of

of

WM

=A0 =A0 =A0Vcc

=A0 =A0 =A0 =A0 =A0 |

ODE>]-+ =A0

=A0 | =A0 =A0 =A0 =A0 =A0| =A0

OIL]--+ =A0 =A0

=A0 =A0 =A0 =A0 =A0| =A0

]-+ =A0

=A0 | =A0 =A0 =A0 =A0 =A0| =A0

=A0 +--[COIL]--+ =A0

=A0 =A0 =A0 =A0 =A0 =A0 =A0| =A0

-[DIODE>]-+ =A0

=A0 =A0 | =A0 =A0 =A0 =A0 =A0| =A0 =A0 =A0 =A0 =A0 =A0

=A0 =A0 =A0 =A0 =A0 =A0

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0

All of which goes to show that he doesn't notice it when you say something stupid. I'm reminded of the King of Id, who was delighted when a monarch showed up who was even shorter than he was.

My preference would be for James Arthur to learn that bankers really can screw up. Once he has mastered this intellectual insight, I'd be happy to see him go on to learn how to ski as fast as John Larkin, or anyone else who wants spend a lot of money on breaking a limb or two.

-- Bill Sloman, Nijmegen

Provided that you remember to intialise it, and to check for false states.

-- Bill Sloman, Nijmegen

Yes folks, I did want a _sinewave_ drive as stated in the OP. The circuit is not intended to activate the relays. The coils are only used to create a rotating EM field, with continuous transition between coils.

I had been considering an op amp phase shift oscillator, but was unable to find the exact circuit.

Any suggestions regarding this would be appreciated.

Mark Weaver

Hmm, i've been watching this thread and there is one thing that I seem to be missing here..

It was asked to do 3 phase signals to operate reed relays ? This tells me that an overlap is desired during the change ?

Or did I miss something with this HC174 chip? I didn't know it could do overlapping effects ( late break effects in the switch world)..

Jamie

On a sunny day (Fri, 05 Nov 2010 22:56:15 GMT) it happened snipped-for-privacy@questron.com (Mark Weaver) wrote in :

This can still be done with a simple PIC by creating 3 software PWM signals and have the software put in values from sine wave table at 120 degrees point every excursion of the PWM routines. Lowpass the PWMs to get the 10 Hz. It is just a bit more art of programming.

An more ancient ways is to use a 555 oscillator, a 4040 binary counter,

It could also be done with a bucket brigade delay line from one 10 Hz wave, or some other phase shifter.

Could also be done with some ring oscillator: ftp://panteltje.com/pub/rgb.jpg

I guess you could use transformers too :-)

Or a rotating disk with electrostatic segments...

Motor, bunch of gears and pots.

There must be many ways, I'd try the PIC first myself. For 1000$ in advance I would have a go at the code.

He never said anything about reed relays, only referred to reed relay coils.

Best regards, Spehro Pefhany

Coil inductance/resistance? ...Jim Thompson

All the more reason to use a PIC--you get stable phase relationships, frequency, and stable amplitude, things less easily achieved with analog techniques.

PWM yielding 10Hz sinewaves is easily done in software.

-- Cheers, James Arthur

Show us some electronics you've designed lately.

John

Yes, I've done that before, years ago. the first Op was the (in your case 10hz osc), that signal was then used to drive a 90 degree shift and another RC shift with slightly different values so not to over shift into 180 land. Then, gain compensation was placed there to make up for loss.. This stage was the reference for the last stage that did the same thing with the same values.

I would suggest how ever, a PIC or AVR would be a much better approach because it would be more accurate and few components.. Since these chips wouldn't be very burden doing this little task, you could use basic PWM schemes in software. You simply do phase (A) and then have the other two follow behind in a buffer or mathematically for the phase B and C.

P.S. A look up table for the SINE PWM translation in the uC would be the best solutions. This way, you can easily generate the A,B and C phases just by starting B and C slightly into the table ahead of A phase and then just wrap around.. just make sure you pic the uC that has enough flash to hold the table and your code.

Yes, he stated he wanted a 3 phase rotating field.. He most likely is doing with with eddy currents.. 10 Hz is the magic freq for most eddy current fields of what I am thinking of. Maybe he is making a crack detector in materiel or the like.

Are you actually using these coils to drive relays? How long are the relays going to last at 10Hz?

Sylvia.