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

Phase-shifter 90 degree

Status
Not open for further replies.
M

MacBam

Newbie level 6
Joined
Nov 11, 2014
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
86
Hi,

I am looking for a phase-shifter circuit, that can shift a signal with 90 degrees - over a frequency range (eg. 50kHz to 200 kHz). Any suggestion?

Best regards,
 

Differential 90 degree phaseshift can be implemented as an approximation with all-pass filter chains. Number of filter stages depend on acceptable phase ripple.
 

FvM said:
Differential 90 degree phaseshift can be implemented as an approximation with all-pass filter chains. Number of filter stages depend on acceptable phase ripple.
Click to expand...

Ok, do you have any link to some litterature.
 

If you're talking about sine waves, a capacitor automatically creates 90 degree phase difference, measuring voltage across it versus current through it. These are not necessarily in phase with the original signal.
 


MacBam said:
Hi,
I am looking for a phase-shifter circuit, that can shift a signal with 90 degrees - over a frequency range (eg. 50kHz to 200 kHz). Any suggestion?
Best regards,
Click to expand...
90 deg (exact !) over such a frequency range is impossible.
Therefore: What are your accuracy requirements?
Perhaps an opamp-based integrating stage?
The phase will be exact 90 deg at one single frequency only - however, the deviation (for smaller and larger frequencies) might be some degree only
 

An integrator (or differentiator) is basically not suited as wideband phase shifter due to frequency dependent magnitude. All-pass chains can achieve good accuracy (below 1 degree) with respective number of stages.
 
Hi,

Is "constant amplitude" a requirement?

Klaus
 

Yes the amplitude is important, I am going to use in a lock-in configuration. What about using Phase-lock loop is that possible?
 

Really depends on what you want to achieve. PLL can generate a signal phase shifted relative to a reference input, usually with fixed magnitude.
--- Updated ---

To be tracked by a PLL, the reference signal must be single frequency and quasi stationary.
 

Is this an analog or digital signal you want phase shifted ?

Allpass Network Shifts 90 Degrees over Wi | Maxim Integrated

Two wideband transconductance amplifiers produce an allpass network. Combining two such networks as shown in the illustration produces two outputs with a constant 90 degree phase shift (vs. frequency) between them.
www.maximintegrated.com www.maximintegrated.com

Wideband 90 deg phase shifter at low frequencies - Q&A - Amplifiers - EngineerZone

Hello! I design the radio receiver with an operating frequency range of 150...2300 kHz. It has several antennas and it is necessary to combine (summarize) two of
ez.analog.com ez.analog.com



https://community.cypress.com/message/142285#142285 Project using hilbert transform filter

Single chip approach -

1603235839390.png


Regards, Dana.
 
danadakk said:
Is this an analog or digital signal you want phase shifted ?

Allpass Network Shifts 90 Degrees over Wi | Maxim Integrated

Two wideband transconductance amplifiers produce an allpass network. Combining two such networks as shown in the illustration produces two outputs with a constant 90 degree phase shift (vs. frequency) between them.
www.maximintegrated.com www.maximintegrated.com

Wideband 90 deg phase shifter at low frequencies - Q&A - Amplifiers - EngineerZone

Hello! I design the radio receiver with an operating frequency range of 150...2300 kHz. It has several antennas and it is necessary to combine (summarize) two of
ez.analog.com ez.analog.com



https://community.cypress.com/message/142285#142285 Project using hilbert transform filter

Single chip approach -

View attachment 164992

Regards, Dana.
Click to expand...
Inetresting, it is an analog signal. Can I generate sine waves in PSoC5?
 

Yes on sine with WaveDAC component. In PSOC land a component
is an onchip resource. What freqs do you need ?

Attached is a component catalog for PSOC 5LP, there are lower end familes
with not as many on chip components. On chip you have logic, LUTs, DACs
OpAmps, Comparators, serial COM like USB, SPI, I2C, UART. DSP, PWM, Counters,
Timers, QuadDec, DDS.... Youc an do custom compoennts using the fabric as
well, I did a SIPO 64 bit shifter recently to learn. There is a community of folks
that have created some components not in standard IDE, like DDS, 74HC
type logic, CORDIC

**broken link removed**

The low end board is $ 10, complete with debug.


1603282713847.png


A board with more I/o, and higher cost - https://www.sparkfun.com/products/13714



IDE, PSOC Creator, and compiler free.

**broken link removed**


Regards, Dana.
 

Attachments

  • Component List (2).pdf
    183 KB · Views: 103
Last edited:

danadakk said:
Yes on sine with WaveDAC component. In PSOC land a component
is an onchip resource. What freqs do you need ?

Attached is a component catalog for PSOC 5LP, there are lower end familes
with not as many on chip components. On chip you have logic, LUTs, DACs
OpAmps, Comparators, serial COM like USB, SPI, I2C, UART. DSP, PWM, Counters,
Timers, QuadDec, DDS.... Youc an do custom compoennts using the fabric as
well, I did a SIPO 64 bit shifter recently to learn. There is a community of folks
that have created some components not in standard IDE, like DDS, 74HC
type logic, CORDIC

**broken link removed**

The low end board is $ 10, complete with debug.


View attachment 165007

A board with more I/o, and higher cost - https://www.sparkfun.com/products/13714



IDE, PSOC Creator, and compiler free.



Regards, Dana.
Click to expand...


MacBam
Newbie level 5
blevel2.gif

JoinedNov 11, 2014Messages8Helped0Reputation0Reaction score0Trophy points1Activity points47
Hi,
What is the resolution? I want to generate signal up to 300kHz with 100Hz steps. I already have the one at the top pictur
 

300 Khz is a push. Its predicated by DAC settling time and sample rate
of the sine wave. So if we used 32 samples/cycle for the sine, then sample
rate is 300 Khz x 32 =~ 10 Msps. The VDAC is limited to 1 Msps. The IDAC
6 Msps. Note this is full scale

That being said these are 50% full scale transitions for the VDAC, 100%
for the IDAC. Since the step response is between adjacent samples that
would produce a much faster possible rate.

If you want I will test at 300 Khz both VDAC and IDAC and produce a spectral
response to see what we get ?

Regards, Dana.
 

I think I can come close to 100 Khz, but 300 out of the question. Largely
due to rate I can trigger DMA to send out a sample. I did not look at
interleaved DMA which might help though.

Regards, Dana.
 

Simulated lissajous figure demonstrates, at all frequencies a capacitor produces 90 degree phase shift (with sine waves). This basic principle applies for all Farad values.
It changes phase from the original signal, so there's a question whether it can be put to work in your own project.

capacitor 90 degree phase shift lissajous.PNG
 

I did a work around to run the WaveDAC8 at ~ 12 Mhz sample rate, 16 samples/cycle,
and got ~ 144 Khz sine, third harmonic was 25 db down. Keep in mind the sample rate
is set by the DMA rate (and its cycles needed).

This violated the timing on the VDAC8 used inside the WaveDAC8, so is not really a reproducible
solution.

FYI.

Regards, Dana.
 
danadakk said:
I did a work around to run the WaveDAC8 at ~ 12 Mhz sample rate, 16 samples/cycle,
and got ~ 144 Khz sine, third harmonic was 25 db down. Keep in mind the sample rate
is set by the DMA rate (and its cycles needed).

This violated the timing on the VDAC8 used inside the WaveDAC8, so is not really a reproducible
solution.

FYI.

Regards, Dana.
Click to expand...
Ok thanks. 180kHz is unfortunately not enough for me.
--- Updated ---

FvM said:
Really depends on what you want to achieve. PLL can generate a signal phase shifted relative to a reference input, usually with fixed magnitude.
--- Updated ---

To be tracked by a PLL, the reference signal must be single frequency and quasi stationary.
Click to expand...

I am going with the all-pass filter. Do you have any suggestions for a good Op-Amp ( I am running from 50kHz to 400kHz). I am thinking the GBW should be 20 X400 kHz and with low slew-rate?
 

High rather than low slew rate, according to the maximum signal voltage. I think many of recent CMOS general purpose OP can do. Don't know if the dynamic range requirements are particularly high? What's the acceptable differential phase ripple?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top