pll active filter - Electronic Circuits Projects Diagrams Free

jovabiot · 12th March 2004, 09:16 PM

I have design a pll type II with an active filter for obtain the frequency and the phase of a signal, but for some phases of this signal, my exit is at this frequency but with a 180º in phase. How can i do for obtain a 0º phase?? thanks a lot

Roff · 13th March 2004, 12:07 AM

I've designed a lot of PLLs, but I don't understand your question.

Russlk · 13th March 2004, 01:50 AM

What PLL circuit are you using? Type II PLL only indicates to me that you are using an operational integrator as filter. What is the phase comparator?

jovabiot · 13th March 2004, 07:23 AM

ok. my circuit is: in the phase detector i have a multiplier, and after that i have the active filter( with operational and RC) for obtain a final phase 0º. My problem is that i have recovered the frequency but the phase, not always is zero, sometimes is 180º, and it depends of the phase in the entry. i want recover the frequency and i want to have phase 0º always.Thanks a lot.

Russlk · 13th March 2004, 03:24 PM

If you are using a multiplier as phase detector, the VCO should be 90 degrees relative to the input. You are doing something strange to get 0 or 180 degrees. I suggest you use CD4066 or 74HC4066 and use the type 2 phase detector. That will give you zero degrees from the VCO.
Another approach (which I have used) is to run the VCO at 4X then divide by 4 with a Johnson counter. The Johnson counter output has 0 and 90 degrees available.

Roff · 13th March 2004, 07:20 PM

Russ, I was trying to come up with a scenario that would cause this. If this were a digital loop (as in no sine waves), and you divided both the input and the output by two before applying them to the multiplier, they would wind up (I think) consistently either in phase or 180 degrees out of phase, depending on the sign of the VCO transfer function. Now, if you took either the input or the output and divided it by two outside the loop, you could get 180 degree ambiguity. Sorta far-fetched, but it's all I can come up with. And why you would do this, .. :?:

jovabiot · 14th March 2004, 09:59 AM

the problem for use this pll (cd40*) is that i want to work at 50Mhz and it can't. Then, i am designing my own model of pll in pspice. I have recovered the frequency but with a RC filter i only recover the frequency; and with the operational filter(active) there are phase at entry that give me 180º in the output. i don't know why....
another problem, have anyone a model for the mixer sa602 or similary in pspice?? thanks

Roff · 14th March 2004, 01:59 PM

Quote:

Originally Posted by jovabiot

the problem for use this pll (cd40*) is that i want to work at 50Mhz and it can't. Then, i am designing my own model of pll in pspice. I have recovered the frequency but with a RC filter i only recover the frequency; and with the operational filter(active) there are phase at entry that give me 180º in the output. i don't know why....
another problem, have anyone a model for the mixer sa602 or similary in pspice?? thanks

Can you post your Pspice schematic here? Maybe we could answer your question if we could see what you are simulating.
As Russ pointed out, if you are using a multiplier as your phase detector, you should be getting 90º at the output, not 0º or 180º.

jovabiot · 15th March 2004, 08:56 AM

this is the model of pll.if i introduce a 100Mhz signal with a fase, i have in the output a 50Mhz signal but sometimes not with a 0º. thanks

Roff · 15th March 2004, 04:11 PM

Perhaps you already know this, but the frequency doubler in your feedback loop is the cause of your phase ambiguity. When you square a sine wave, the input can be 0º or 180º and you will still get the same double-frequency waveform. What are you really trying to accomplish? There may be another way. Are you simply trying to divide the input frequency by two?

jovabiot · 15th March 2004, 10:29 PM

yes. i am going to recover the modulated signal of a bpsk, then i need the frequency in the input divide by two, but in phase. do you know another thing i can do for this?
And do you know a comercial multiplier for do the bpsk?? (multiplie a digital and an AC signal)
thanks for all

Russlk · 16th March 2004, 07:01 PM

Is the 100 mHz input signal BPSK? If so, what is the PLL doing? If you are trying to compare the phase of the 50 mHz signal with the 100 mHz input, you can't do that. One cycle is just like any other! Why not run the VCO at 100 mHz and divide by two at the output?

jovabiot · 16th March 2004, 10:03 PM

how can i divide after the pll by two the frequency and obtain a good signal in frequency ?i tried with flip-flop but i didn't obtain a good signal.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

12th March 2004, 09:16 PM	(permalink)
jovabiot	pll active filter I have design a pll type II with an active filter for obtain the frequency and the phase of a signal, but for some phases of this signal, my exit is at this frequency but with a 180º in phase. How can i do for obtain a 0º phase?? thanks a lot

13th March 2004, 12:07 AM	(permalink)
Roff	I've designed a lot of PLLs, but I don't understand your question.

13th March 2004, 01:50 AM	(permalink)
Russlk	What PLL circuit are you using? Type II PLL only indicates to me that you are using an operational integrator as filter. What is the phase comparator? __________________ see my website: www.geocities.com/russlk

13th March 2004, 07:23 AM	(permalink)
jovabiot	active filter ok. my circuit is: in the phase detector i have a multiplier, and after that i have the active filter( with operational and RC) for obtain a final phase 0º. My problem is that i have recovered the frequency but the phase, not always is zero, sometimes is 180º, and it depends of the phase in the entry. i want recover the frequency and i want to have phase 0º always.Thanks a lot.

13th March 2004, 03:24 PM	(permalink)
Russlk	If you are using a multiplier as phase detector, the VCO should be 90 degrees relative to the input. You are doing something strange to get 0 or 180 degrees. I suggest you use CD4066 or 74HC4066 and use the type 2 phase detector. That will give you zero degrees from the VCO. Another approach (which I have used) is to run the VCO at 4X then divide by 4 with a Johnson counter. The Johnson counter output has 0 and 90 degrees available. __________________ see my website: www.geocities.com/russlk

13th March 2004, 07:20 PM	(permalink)
Roff	Russ, I was trying to come up with a scenario that would cause this. If this were a digital loop (as in no sine waves), and you divided both the input and the output by two before applying them to the multiplier, they would wind up (I think) consistently either in phase or 180 degrees out of phase, depending on the sign of the VCO transfer function. Now, if you took either the input or the output and divided it by two outside the loop, you could get 180 degree ambiguity. Sorta far-fetched, but it's all I can come up with. And why you would do this, .. :?:

14th March 2004, 09:59 AM	(permalink)
jovabiot	pll the problem for use this pll (cd40*) is that i want to work at 50Mhz and it can't. Then, i am designing my own model of pll in pspice. I have recovered the frequency but with a RC filter i only recover the frequency; and with the operational filter(active) there are phase at entry that give me 180º in the output. i don't know why.... another problem, have anyone a model for the mixer sa602 or similary in pspice?? thanks

15th March 2004, 04:11 PM	(permalink)
Roff	Perhaps you already know this, but the frequency doubler in your feedback loop is the cause of your phase ambiguity. When you square a sine wave, the input can be 0º or 180º and you will still get the same double-frequency waveform. What are you really trying to accomplish? There may be another way. Are you simply trying to divide the input frequency by two?

15th March 2004, 10:29 PM	(permalink)
jovabiot	pll yes. i am going to recover the modulated signal of a bpsk, then i need the frequency in the input divide by two, but in phase. do you know another thing i can do for this? And do you know a comercial multiplier for do the bpsk?? (multiplie a digital and an AC signal) thanks for all

16th March 2004, 07:01 PM	(permalink)
Russlk	Is the 100 mHz input signal BPSK? If so, what is the PLL doing? If you are trying to compare the phase of the 50 mHz signal with the 100 mHz input, you can't do that. One cycle is just like any other! Why not run the VCO at 100 mHz and divide by two at the output? __________________ see my website: www.geocities.com/russlk

16th March 2004, 10:03 PM	(permalink)
jovabiot	pll how can i divide after the pll by two the frequency and obtain a good signal in frequency ?i tried with flip-flop but i didn't obtain a good signal.