phase shifter circuit

On a sunny day (Thu, 18 Oct 2007 12:34:34 -0000) it happened snipped-for-privacy@gmail.com wrote in :

So, comparator - 131 cycle delay - | 1MHz clock

Solution: PIC. Use 8 pin PIC. 12F629 it has the comparator. pins: +, -, in, xtal in, xtal out, out: makes 7. One spare pin to break off.

But actually, if you want to use the DCR-10520 to convert to sine, then a PWM on the PIC can do that too.

"Jan Panteltje" snipped-for-privacy@gmail.com

Seems he wants to have his info before it arrives.

....... Phil

On a sunny day (Thu, 18 Oct 2007 23:47:14 +1000) it happened "Phil Allison" wrote in :

My wrong, delay one cycle - lead = 2049.180328 - 130.918542 = 1918.261786 say 1918 us. So delay 1918 clock ticks of 1MHz.

"Jan Panteltje"

Betcha.

..... Phil

On a sunny day (Fri, 19 Oct 2007 00:12:32 +1000) it happened "Phil Allison" wrote in :

Well, we will have to wait and see, I have hopes.

Not enough information. You **can** use a simple RC phase shifter with an amplitude follower and about four or five stages to give you 337 degrees of phase LAG, which is the same thing as 23 degrees of phase LEAD for the following cycle.

Jim

-- "If you think you can, or think you can't, you're right." --Henry Ford

any advice/books/links that might aid this design?

For a fixed frequency, you can get small amounts of phase lead easily with an RC highpass network. The phase of such a network is

tan(23 degrees) phi = atan(1/(2*pi*f*RC)), so RC = --------------- = 138 microseconds. 2*pi*(488 Hz)

You could use a 140k resistor and a 1nF capacitor. You'll want to use a decent cap, so that your phase doesn't vary everyplace, but your spec corresponds to about +- 25% capacitor tolerance, so something like a 10% film cap will do fine.

You can clean up the edges with a comparator.

Cheers,

Phil Hobbs

Seems like other posters so far haven't quite groked what a DRC-10520 is. You need to drive the differential reference input with a sinewave (not digital) at 3.4V RMS, and the input impedance is guaranteed within 0.5% of 26 kohms. If your input 6.1V is isolated (not ground referenced), it should be easy to use series capacitors and resistors to scale and phase shift. Phil Hobbs' posting gives you the basic info you need for the phase shift part...you just need to understand how much attenuation there is in the high pass filter so you know the voltage out of it, and then pick the load resistance such that the added resistors and the internal DRC-10520 resistance form a voltage divider to get you the rest of the way to 3.4 volts.

If your 6.1V is ground referenced, then I'm unsure whether you will be able to use that directly or not. The DRC-10520 data sheet is very vague on the common mode range of the reference input, and whether it must be driven differentially or if instead RL may be grounded. I'd call DDC about that one, to find out. If it can be grounded, then you can just use a single series RC from your 6.1V to get the phase shift and the attenuation. But DDC's data sheet for this part only shows the reference being driven from an isolated transformer winding, so it's not really clear how you're supposed to drive it single-ended. It's also unclear to me just what they mean by listing different single-ended and differential reference input impedances; perhaps you can drive it single-ended with RL floating. Again, get ahold of DDC and ask them! They should be happy to provide you with some applications info about the part. It's to their advantage to have you use the part successfully.

Cheers, Tom