active loop filter
Added after 5 hours 18 minutes:
please see the image
Why not ? Do you instead prefer a connection with the neg. input (with an additional phase shift of 180 deg) - with the result of instability ?
Additional when input is direct to amp
you have more gain
Noninvert G=1+R2/R1
Invert G=-R2/R1
Also impedance of the + is infite!
Those are all good reasons. But the fundamental reason that you might select the + input is that you have to choose either the + or - input depending on the tuning slope of the VCO used. Many VCOs tune higher in frequency the more positive the tuning voltage goes. Usually (but not always) PLL chips are set up so you want a positive op amp circuit gain for positive tuning slope VCOs. Some VCOs tune higher in frequency with a more negative going tuning voltage, and the opposite would then be true. In other words, if you want your control loop to work, it has to be set up for "negative feedback"--that is set up so that an error voltage coming out of the phase detector tunes the VCO in a direction that corrects that phase error.
As a secondary effect, you might not want to connect DIRECTLY to the op amp + terminal if there is a chance that the input differential volatge maximum can be exceeded. An op-27 can only stand +/-0.7 V input differential voltage max, and going over that might overload, or even damage, the op amp. You would want a current limiting resistor in series with the + pin if that were a possibility.
Rich
www.MaguffinMicrowave.com
Yes, quite right !
One additional comment to the circuit: Wouldn′t it be better to incorporate the last stage R3_C3 in the active part (i.e. to design a real active filter !).
In the circuit shown the R3_C3 lowpass is loaded by the VCO input and its pole frequency will be influenced !
Good point. It is a matter of personal preference. In this specfic crcuit, where C1, C2, and R1 form the dominant poles and zeros of the control loop, you could easily make the op amp circuit itself be a higher frequency lowpass flter. But you would lose out on one important safety feature--the op amp output voltage swing. Depending on unique or temporary circumstances (power supply sequencing, etc) the op amp output could momentarily end up at +5 or -5 or whatever voltage, until the PLL retunes the VCO tune voltage to be locked. This has the potential for blowing up the varactor diode in the VCO. As a rule of thumb, I always like to put at least a 100 ohm series resistor between the op amp output and the VCO tuning port for this reason. And if I am going to do that, I mght as well put it physically located right at the VCO, and add the shunt capacitor, to form a lowpass that helps to keep stray board noise (digtal, etc) from being picked up by the high impedance VCO tune port.
If it were a monolithic chip, it would not be an issue. But discrete components--its best to be safe.
And usually the chip capacitor used at the tune port input swamps out the internal shunt capacitances of the VCO circuit, and makes things behave a little more predictably from unit to unit.
Rich
www.MaguffinMicrowave.com
Thanks.
Could you tell me the advantage and disadvantage of that type and this type?
Please see the image next.