Anyone have ideas regarding how to implement a voltage-variable phase shifter useable at 210MHz? ...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et |
It depends on how many degrees of phase shift you want, but you need varact ors. For say 45 degrees try a PI network with two varactors across the top and an inductor to ground. Play around with the values and you can keep a t ight VSWR.
For more phase shift use a 90 degree hybrid ( 4 port device). The input po rt goes to two 90 degree outputs, and for infinite VSWR it will bounce off and go to the output port. Use matched varactors on the two 90 degree por ts and you will be able to get a very good phase shifter to almost 90 degre es. If you want more then add hybrid devices together.
ctors. For say 45 degrees try a PI network with two varactors across the t op and an inductor to ground. Play around with the values and you can keep a t ight VSWR.
port goes to two 90 degree outputs, and for infinite VSWR it will bounce o ff and go to the output port. Use matched varactors on the two 90 degree p orts and you will be able to get a very good phase shifter to almost 90 deg rees. If you want more then add hybrid devices together.
oops, T network , not PI
Do you remember the MXR Phase 90? It was a guitar effects pedal that used a chain of op amp all-pass filters:
as a phase shift network and a JFET as a voltage variable resistance in parallel with the resistor to ground to vary the phase.
Could one build something similar just for a higher frequency, maybe? 500 MHz GBW op amps are not terribly expensive these days; use PHEMTs as the variable resistance elements?
--
----Android NewsGroup Reader----
Do you want to buy it as a part, or build it into one of your IC designs? How much phase shift do you need?
There are tons of analog IQ modulator chips around, some with a built-in 90 degree phase shifter. They can do full the 360 degrees if you apply sin/cos multiplier inputs.
Small phase shifts can be done with a varicap RC or LC network. Or an RC or LC shifter and a multiplier or two.
Extreme: an inductor-varicap variable-delay transmission line. The HP8133A had several, insanely long ones.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement
I've used a variation of the normal op amp all-pass filter for this, except with a high frequency diff amp:
*---------------* | | 0-----*-----RRRR-------*----| | | | | | | R | Diff amp | | Gain of 0.5 R | | | R | *-----Out | | | | | | | | | GND | | | | | *-----RRRR--*----*----| | | | | | | | *---------------* | L | L | L | L L | L | L | L --- Varactor | ^ | | | *---RRRR----0 V_phi | | | | CCC | CCC big C | | GND GNDYou pick the voltage divider to have a gain of 0.5; the series L resonates the varactor just past the low-voltage end of V_phi; and the shunt inductor resonates the series combination just past the high end of V_phi. You can get about 150 degrees in one stage, with linearity of +- a few degrees. Good medicine.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant
It would be helpful if you indicated whether your input signal was a constant sinusoid, square wave or some arbitrary signal (in which case it would also be helpful if you indicated the signal bandwidth, as this makes a difference). It would also be useful to know whether the phase shift should be proportional to the voltage, or the time delay through the circuit should be proportional to the voltage, or whatever the control function should be. Also the linearity of the control input would be an important spec. as well as any requirement on the phase noise introduced (e.g. if there is noise added to the phase control signal).
That's what the customer is currently using except for a capacitance multiplier made with an AD834(?) ...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et |
Crikey! Why didn't I think of that... marvelous! Thanks for the pointer!
...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et |
Has someone already said opamp all-pass and varactor/diode?
George H.
For small phase shifts, you could bias a jfet or something, with a d-g feedback cap, and vary Vgs (ie, transconductance) to change the phase shift. Dirt cheap. An inductor somewhere might be interesting.
People used to make FM modulators like that, with toobs. Pre-emphasis made phase shift into frequency shift. Reactance tube modulator, something like that.
-- John Larkin Highland Technology, Inc picosecond timing laser drivers and controllers
My go-to toob book, "Electron Tube Circuits", by Samuel Seeley, second edition 1958
-- ----Android NewsGroup Reader---- http://usenet.sinaapp.com/
My go to toob book, "Electron Tube Circuits", by Samuel Seeley, second edition 1958, has a section on reactance tube modulators:
-- ----Android NewsGroup Reader---- http://usenet.sinaapp.com/
Sorry, something weird going on with my Android newsgroup reader. Anyway, I was going to say that the book has a section on reactance tube modulators:
-- ----Android NewsGroup Reader---- http://usenet.sinaapp.com/
The Analog Devices AD834 is four quadrant analog multiplier good to 500MHz. It certainly struck me as the obvious tool for the job.
Jim is probably suffering from non-invented here. Barry Gilbert's analog multipliers were a lot better than Jim's designs for Motorola (and lot more expensive).
-- Bill Sloman, Sydney
Order a packaged voltage controlled phase shifter from Lorch or Mini-Circuits and be done by Lunch?
Steve
The 90 degree phase shifter which is built into the IQ modulator is not a variable phase shift. It is shifting the sample frequency of the IQ modulator by 90 degrees by using a master clock which is 4X the ssample rate.
Perhaps the IQ modulator will work, but it does require two very precisely controled analog voltages to achieve the desired phase shift.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.