Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

[SOLVED] Active Filter for PLL

Status
Not open for further replies.
K

kommanche

Member level 4
Joined
Mar 18, 2007
Messages
70
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
1,766
Hello,

I want to design a PLL based oscillator. I am using ADF4157 as synthesizer and AD797A as active filter OPAMP.

Schematic of the structure is as given in the attachment.

I operate the opamp by supplying Vs+: 15V and Vs-: GND ; Vin+: GND

I am sure that ADF4157 is working correctly but I see 14V at the tuning port of oscillator in other words my active filter does not work. When I decrease Vs+, I always see 1V below Vs+ value at the tuning port (i.e. Vt=12V when Vs+=13V...). Situation did not change when I leave Vin+ opened or gave a reference voltage.

View attachment untitled.bmp

Do you have any idea about this problem?
 

Two somewhat obvious things to try:
1) You should try reversing the polarity of the charge pump, since the op amp integrator inverts the polarity. There is probably a bit you can set in programming the chip to try the other polarity.
2) The R3 and C4 values look too big. Try R3 at 220 ohms, and try C4 at something like 100 pF. If that capacitor is too big, the phase shift from that additional pole will definately push the loop into an unstable region (no phase margin).

Let us know how it goes.
Rich
 
Last edited by a moderator:
Hi kommanche:

Three questions:

1.) Single or split supply?
2.) Is the signal input able to provide the necessary dc bias for the inv. opamp terminal?
3.) No dc feedback (necessary for dc stability)?

---------- Post added at 11:51 ---------- Previous post was at 11:46 ----------

to 3.): Not absolutely necessary, but to be preferred for most cases.
 
Yeah, good point, I have been having a lot of trouble with "single-supply" op amps doing uninteded things lately!!! THey have their own rules to live by.
 
First of all thank you,

to Biff44:
I will try your recommendations,

to LvW:
1) It is single supply
2) Yes
3) I don't have a DC feedback

I will inform you about the progress,
 

No, I don't know that but I also don't know why it is so complicated. Is the opamp selection problematic? Or I need extra circutry to use it? I also don't know split supply concept.

Can you please give an example?

Thanks.
 

kommanche said:
No, I don't know that but I also don't know why it is so complicated. Is the opamp selection problematic? Or I need extra circutry to use it? I also don't know split supply concept.
Can you please give an example?
Thanks.
Click to expand...

For symmetrical (dual) supply the designed operating point (opamp dc output) is approx. zero volts - thus your signal output can swing to both sides.
For single supply you must bias the opamp in order to have an operating point approx. at half of the supply voltage that allows again signal swing in both directions.
Remember the voltage divider at the base of a transistor in common emitter configuration.
For details you can find a lot of information via googling "single supply".

---------- Post added at 09:46 ---------- Previous post was at 09:44 ----------

But check if the whole circuit (PLL) works correctly with a quiescent filter dc output voltage.
 
Thanks LvW. I understand the configuration I attached from your words. If it is correct, I have already tried this configuration. I expect DC operating point of 5.2 V
and I bias the Vin+ at 3V. Unfortunately opamp pulls the output voltage to +Vcc.
View attachment untitled2.bmp


LvW said:
For symmetrical (dual) supply the designed operating point (opamp dc output) is approx. zero volts - thus your signal output can swing to both sides.
For single supply you must bias the opamp in order to have an operating point approx. at half of the supply voltage that allows again signal swing in both directions.
Remember the voltage divider at the base of a transistor in common emitter configuration.
For details you can find a lot of information via googling "single supply".



---------- Post added at 09:46 ---------- Previous post was at 09:44 ----------

But check if the whole circuit (PLL) works correctly with a quiescent filter dc output voltage.
Click to expand...
 

kommanche said:
Thanks LvW. I understand the configuration I attached from your words. If it is correct, I have already tried this configuration. I expect DC operating point of 5.2 V
and I bias the Vin+ at 3V. Unfortunately opamp pulls the output voltage to +Vcc.
View attachment 66087
Click to expand...

Yeah - no surprise. This happens because you have no dc feedback. You need 100% dc feedback - only then the dc gain is unity and transfers the dc bias at the pos. input to the output.
 
LvW said:
Yeah - no surprise. This happens because you have no dc feedback. You need 100% dc feedback - only then the dc gain is unity and transfers the dc bias at the pos. input to the output.
Click to expand...

Of course you are right. I understand the point. I now think of an inductive feedback, can you suggest a suitable feedback configuration?

---------- Post added at 11:26 ---------- Previous post was at 11:09 ----------

Will it work if I connect a resistor from Vin+ to Vout?
 

kommanche said:
Will it work if I connect a resistor from Vin+ to Vout?
Click to expand...

Are you joking?
I suppose if you are working with PLL's you know some opamp basics.
You ask for a "suitable configuration"?
At first, you have to clarify
(a) what kind of loop filter you need (are there alternatives?)
(b) if you really have to live with single supply
(c) the input requirements of the VCO (nominal control voltage zero volts or something else?), because I suppose that capacitive coupling between loop filter and VCO does not work.
 
Hi,

your loop filter seems to be wrong, check your loop filter calculation.
Otherwise disconnect the Active filter and put passive loop filter.

Regards,
Prabuddha



kommanche said:
Hello,

I want to design a PLL based oscillator. I am using ADF4157 as synthesizer and AD797A as active filter OPAMP.

Schematic of the structure is as given in the attachment.

I operate the opamp by supplying Vs+: 15V and Vs-: GND ; Vin+: GND

I am sure that ADF4157 is working correctly but I see 14V at the tuning port of oscillator in other words my active filter does not work. When I decrease Vs+, I always see 1V below Vs+ value at the tuning port (i.e. Vt=12V when Vs+=13V...). Situation did not change when I leave Vin+ opened or gave a reference voltage.

View attachment 66020

Do you have any idea about this problem?
Click to expand...
 

In my opinion DC feedback is not necessary.
In the past I used AD797A in a single supply configuration and I have had troubles. Even when the PLL was working, changing the op-amp with OP27 the performances were better.
Remeber to put a capacitor in parallel to Rb.
Of course, you are sure the frequency you want to tune is inside VCO frequency range ?
I think you should check if the amplitude of both reference anf RF feedback are high enough. To do this you can program the MUXOUT in order to see on this pin the N divider output and then the R divider output. You can check those signal by means of an oscilloscope. You should see in both cases a rectangular wave at the frequency you set the comparator.
 

LvW said:
Are you joking?
I suppose if you are working with PLL's you know some opamp basics.
You ask for a "suitable configuration"?
At first, you have to clarify
(a) what kind of loop filter you need (are there alternatives?)
(b) if you really have to live with single supply
(c) the input requirements of the VCO (nominal control voltage zero volts or something else?), because I suppose that capacitive coupling between loop filter and VCO does not work.
Click to expand...

I have used passive filters up to now. I am using this active filter first time.

If you have looked at the first post I have sent my filter.

I must use single supply.

I am not using a capacitive coupling between loop filter and VCO.

By the way this is a suggested filter in ADISim.
 

Another thing: don't forget to use decoupling capacitor for both RF feedback and ref in since they are self biased.
 

I did not know this. Actually I know what an opamp is but I am unfamiliar to usage of it. I was considering this structure as a standard thing that can be used easily. I am sorry about my very basic questions but do you have any suggestions about modification that can make this work?
 

kommanche said:
I did not know this. Actually I know what an opamp is but I am unfamiliar to usage of it. I was considering this structure as a standard thing that can be used easily. I am sorry about my very basic questions but do you have any suggestions about modification that can make this work?
Click to expand...

I am not a specialist for PLL design - some other forum members certainly have more experience.
However, for my opinion the first question to be answered is (as mentioned before): What is the nominal VCO control voltage that must be provided by the loop filter?
 
Last edited:
LvW said:
I a not a specialist for PLL design - some other forum members certainly have more experience.
However, for my opinion the first question to be answered is (as mentioned before): What is the nominal VCO control voltage that must be provided by the loop filter?
Click to expand...

It is approximately 5.2 V
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top