I haven't spent a lot of time looking at your very detailed description of the problem. Indeed, the question is pushed out of focus by the sheer amount of information here.

.

I suggest you should bring it back into crisp sharpness again, explaining as few words as possible precisely what it is that you need to measure, and why, and what barriers to progress are you encountering.

.

Barrie

Barrie,

Thanks to the Coach ! 

Have updated the website. 

Glen

Glen:

.

You have done a nice job in your explanations

of how a set of different phases in the harmonic

decomposition of a illustrative square-wave will

add-up improperly when reconstructed after its

journey through one or more filters. I invariably

find transient analysis (in simulation) provides

more immediate and usually more satisfactory

insights into many filter types.

.

The font sizes in many parts of your web-site

presentation appear very small on my lap-top.

Do they on your computer?

.

Cheers

.

Barrie

edited: 2018-10-13

purpose: clearer definition of Target

based on "Triple-Signal" and "Narrow-Group-Delay"

Barrie,

My goal is to better understand the concept of phase-delay

and build guidelines for designing filters such that

the 'mythical' group-delay distortion can be controlled,

aproximately.

Given a Triple-Signal behaves almost like a single frequency ,

having no odd harmonics,

it may be that Phase-Delay and Group-Delay 

are about the same in their effect. 

To be clear, quoting your comment, 

"""...will add-up improperly when reconstructed after its 

journey through one or more filters.""" 

is not what I intended to show. 

Fourier Transforms show that well enough.

Seems that in the real analog world,

square wave pulses are hard to generate,

always having a measurable rise/fall time,  over-shoot, etc., 

and are interesting phenomenom in their own right.

( HiFi Wide-Band Audio frequencies are filtered out,

thus are defined 'out-of-range' for this project ) 

The redefined Group-Delay becomes "Narrow-Group-Delay".

* Given: the "Triple-Signal" of 650Hz and 700Hz and 750Hz ,  

are the only signals that will pass through the filter stages,

* Then: the GOAL is to see if usable data can be provided by SPICE,

such that some GUIDELINES could be formed

for band-pass filter design, based on the realistic "Triple-Signal".

In re-thinking my goal, you are moving my thinking closer

to a clearer definition of this experiment.

If the Triple-Signal is analogous to a Single-Frequency  ,

then Phase-Delay and Group-Delay may be the same, practically.

* Given: the limits of the C.W. communication scenario

( narrow passbands of Q=5 )

* Then: this "Triple-Signal" is the "Target" to be investigated. 

* Therefore: this project will continue investigating

phase-relationships based on

the "Triple-Signal" Narrow-Group-Delay.

Glen

Barrie,

Perhaps to answser my own question:

General Summary : 

Measurements of Phase-Delay on the Triple-Signal  

do not produce measurements  

which link Signal-Phase-Delay with Group-Delay ,  

in the field of C.W. communication audio narrow pass-band signal. 

(1) It is found that the prior stage filtering attenuates the 1/3 and 3rd odd harmonic of the 700Hz target signal, such that no audio square-wave pulse characteristic is practically measured. 

(2) It is found that the Triple-Signal (650,700,750)  

mixes with itself, causing 'beat frequency' distortion.

Signal phase shift on each of the three signals

is in the order of 50uS after 12mS time, Q=10,

Phase-Shift depends on Q of the filter

with Q=3 being less and Q=10 being more Phase-Shift.   

HiFi wideband audio and C.W. communications 

are different scenarious.  

Group delay is widely reported and corrected in HiFi audio.

Group delay is casually reported in C.W. communications, 

with no experimental evidence. 

This experiment shows, in detail ,

that the Signal-Phase-Delay of 6mS ,

produced by the Q10 filter itself ,  

is a major contributor to the effects

of overall Signal Delay Distortion.  

Of concern to a C.W. operator,  working at speeds of 10wpm,

( where 'dit'=60mS, 'dah'=180ms ),

this phase-delay is only 10% of the shortest 'dit' signal

 and only 3% of the 'dah' signal.

(1) In practice, at 10wmp this delay is very tolerable,

since the sine-shaped pulse is long enough to detect by ear. .

(2) In practice, at 40wmp this delay is very destructive,

since the sine-shaped pulse is indistinct by ear. .

For example some real-time measurements:  

Given 10wpm and  'dit' as 120mS, then 6mS rise and 6mS fall takes away 10% of the 'dit. 

Given 20wpm and  'dit' as  60mS, then 6mS rise and 6mS fall takes away 20% of the 'dit'. 

Given 40wpm and  'dit' as  30mS, then 6mS rise and 6mS fall takes away 40% of the 'dit'. 

Results here are that  signal-phase-delay ( alone )

can distort the rise/fall times of a C.W. audio signal,

and can 'smother' a really fast CW signal.  

Results here are that Signal-Phase-Delay 

does not indicate Group-Delay, 

and Signal-Phase-Delay is adequate to cause distortion. 

Further investigation need to be done 

to provide more understanding of this phenomenom

This experiment is published on my website 

click down to "Signal Phase-Delay Analysis" bottom of page. 

Glen

http://www.geocities.ws/glene77is/articles_ET_files/article_ET_AFX_PhaseD.html

Is it possible to measure Group Delay, using SPICE ?

Not exactly, but by analogy "yes". 

With reference to the prior posts: 

Results at this stage are that Signal-Phase-Delay 

does not indicate Group-Delay exactly.  

But,  Triple-Signal Phase-Delay is adequate

to cause a phase distortion

which is analogous to Group-Delay. 

Further observation leads to these considerations : 

(0) Filtering practically removes sub/supra Harmonics. 

(1) the Triple-Signals are the only and basic signal for use.

(2) the Triple-Signals are representative of C.W. signals.

(3) the Phase-Delay for the Triple-Signals do cause distortion.

(4) the distortion is analogous to Group-Delay distortion. 

Further investigation and observation is in process  

to provide more understanding of this phenomenom

This experiment is published on my website 

click down to "Signal Phase-Delay Analysis" bottom of page. 

Glen

Is it possible to measure Group Delay, using SPICE ? - ResearchGate. Available from: https://www.researchgate.net/post/Is_it_possible_to_measure_Group_Delay_using_SPICE [accessed Aug 1, 2015].

http://www.geocities.ws/glene77is/articles_ET_files/article_ET_AFX_PhaseD.html

Barrie, 

I see that I now have a Research Project,

and Abstract and Description,

posted on Research Gate.

I am surrounded by exceptional research fellows

whose projects have titles I cannot spell, 

and whose descriptions  are obviously beyond the scope

of my past experience.  

What a very vast and wonderful world this is,

I am both elated and embarrased

that the thoughts of my old mind

could merit such formal presentation.  

Barrie, for the part you have played in coaching ,

I owe you 'thanks'.  

I must give credit where credit is due.

There is much re-wording to be done still, on the website,

and I may have to tweak the html for better viewing.  

But, here is the latest upgrade to my initial paragraphs.  

I hope that I am pulling out the portion that relates

more directly with an "abstract",

and pushing the remainder down into the "description" area.  

But, even as I write, I see too much description

creeping into the "abstract". 

The following quote is intended to be what I call a "Rationale" ,

a reason to exist,  which may be analogous to an Abstract:

*************************************************************************

Rationale:

My goal is to better understand the concept of phase-delay.  

Then build guidelines for designing filters

such that  the phase-delay and group-delay distortions

can be controlled in narrow-band C.W. Amateur Radio operations. 

Given that a C.W. narrow-band audio signal has no odd harmonics,  

it may be that Phase-Delay and InterModulation-Distortion

are the only effects for study.

For this study, the standard signal is f(700),

and odd harmonics for this are 233Hz and 2100Hz,

which are far outside the narrow bandpass of 100Hz.  

Thus, the Narrow-Band C.W. audio signal has no odd harmonics,

and the turn-on and turn-off shapes are of a sine-wave shape  

( not a square-pulse shape ).  This is termed "rise-time". 

It can be shown , via an O'scope, that the leading and trailing edges

of a C.W. audio signal have the characteristic sine-shape .  

The author's measurements in SPICE of the rise-time  

for a given 700Hz square-wave signal,

passing through a Q=10 band-pass filter is about 1.4 mS.  .  

Phase-Delay and Group-Delay are well defined in WIKI,  

as they are related to HiFi wide-band music audio.

In Wiki, neither concepts are developed relating to Narrow-Band Continuous-Wave Morse-Code type signals. 

Narrow-Bandwidth C.W. signals are about 100Hz wide

[ depending on the relation of BW=(Baud * K ) + Delta-F()  

which can apply to all digital pulsed signals. ]. This is an average of measurements/calculations for C.W. speeds varying

from 10 wpm to 40 wmp, that is from 30 Hz to 120 Hz band-width, containing the central signal and sideband IMD

passing through the filters. .  

The author adopts this 100Hz band-width

as typical and usable for study. 

In the real analog world,    square wave pulses are hard to generate, 

always   having a measurable rise/fall time, over-shoot, etc.,

and never being truly square.  

Square wave pulses of audio always require strong

Odd sub-and-supra Harmonics as a 'package'

in order to begin producing the 'square'ness

of the pulse wave form.  

When passed through narrow bandpass filters

( here read 100Hz BandWidth or Q=7 )

the C.W. audio signal will have a sine-shaped

leading/trailing edge ( No Squareness ).  

Given that this study is with C.W. Narrow Band 

the author has adopted the "triple-signal",

comprised of 600Hz and 700Hz and 750Hz,  

which is all that will pass through from earlier bandpass filter stages.  

This can be described as a data signal and adjacent background noise signals. 

The GOAL is to see if usuable data can be provided by SPICE,  

such that some GUIDELINES could be formed for band-pass filter design. 

If the Triple-Signal is analogous to a Single-Audio-Signal ,  

then Phase-Delay and Inter-Modulation Distortion

may be a proper study in C.W. audio signals. 

It can be shown that the Triple-Signal,

passing through a narrow band-pass filter,

has Phase-Shifting and Intermodulation Distortion

sufficient to alter a C.W. pulsed signal  

without reference to Group-Delay concepts. 

SPICE has been a valuable tool

in Realizing / Measuring these thoughts. 

Is it possible to measure Group Delay, using SPICE ?   

Available from:  https://www.researchgate.net/post/Is_it_possible_to_measure_Group_Delay_using_SPICE  

*************************************************************************

 I realize that this is still not a single paragraph abstract. 

My mind is not in tune with single paragraphs, 

and it is nly with many iterations 

that I can be so succinct.   

I still do electrical projects ( they call it 'engineering' )

at a large chuch facility, 

which seems all too basic, needed and time-consuming,

so in a few days I will return

to run through this "rationale" again.  

Glen

http://www.geocities.ws/glene77is/articles_ET_files/article_ET_AFX_PhaseD.html

This (free) tool will calculate group delay directly as one of the SPICE post-processing functions.

http://www.designsim.com.au/vspice3.html

It runs on Linux or Sun SPARC.

Mark,

Thank you for introducing me to your studies and developments.

Obvsiously, you have been around this tree several times,

... your writing gives the evidence of a seasoned professional.

...

I must mention that Barrie Gilbert , PhD, Director, Analog Devices Inc. ,

Beaverton, OR, was most encouraging

and offered many good insights into my way of thinking

and provided many inventive and thoughtful research papers.

...

"Group Delay" is a broad term from the hi-fi audio field,

and refers to the broad spectrum including at least

-1 Octave up to +1 Octave to capture sub/supra harmonic effects

... at least, that is the way it seemed to me.

...

My experiments, developments and working circuits

continue to involve Amateur Radio operations ,

where we are monitoring, receiving, decoding and understanding

the messages from Morse Code signals.

This has been my ultimate goal ; ie , radio "CW" operations.

...

My issues with Group Delay,

and having only PartSIM.com ngSpice to use via internet browser,

was resolved by carefully defining Group Delay in my experiments

so that the term Group Delay only included frequencies +/- 50 Hz

from my target frequency ( f(700) ) in the audio spectrum.

Given an Radio Frequency signal of (say) 7.050 MHz ,

our receivers can produce a f(center) of 700 Hz from a CW signal.

This is what a CW radio operator hears and must understand.

...

Using this narrow definition I was able to legitimately

derive useful phase-delay measurements

from the "Transient" function of the ngSpice plots.

...

From there, my method involved injecting three signals

( typical in Radio CW operations ) of 650, 700, 750 Hz,

and measuring effects.

Emphasis became on Phase Delay effects at +/- 50 Hz of 700 Hz.

...

From there, I proceeded to develop a method of mixing

low Q and high Q samples generated from the same 700 Hz signal.

...

The result was a simple circuit to shape and mix phased signals

and produce two -80 dB notches

surrounding my f(700) target signal.

Project is on my website, www.GeoCities.WS/glene77is

and a sample attached below . Prime result is "AFV" circuit.

...

With some refinements, that has fairly-well completed the project.

I use this working circuit daily in my Ham Radio CW operations.

The circuit , at a 100 Hz BandWidth matches published DSP quality.

It is all Analog and does not require a computer for operation.

I should note, that my first interest in Ham Radio began in the 1950's

with my first license and first station.

I am impressed with DSP capabilities , having spent decades

of my career involved with computer relational database methods.

I am always amazed, but as a old man,

looking for "one last project" ,

I wanted to do a fully Analog CW Filtering System

and do this by exploring some radical (non traditional ) techniques.

...

Well, I am sure all that is a diversion from your interests,

it is just the way an old apprentice thinks,

...

still tinkering in the Master's Workshop for a few more years.

...

I wish you good things in your projects, Mark,

and will 'follow' on RGN.

Glen Ellis, K4KKQ

Hi Glen,

Thanks for the input, sounds like a fascinating project.

We calculate group delay as (d.phi/d.f) across the frequency band, and plot it as a continuous trace. Please feel free to download all or any of the tools on the website, including the component libraries and modelling tools.

edited: 2018-10-13

purpose: redefine the target "Group-Delay"

based on a "Triple-Signal" and "Narrow-Group-Delay"

( HiFi Wide-Band Audio frequencies are filtered out,

thus are defined 'out-of-range' for this project ) 

The redefined Group-Delay becomes "Narrow-Group-Delay".

* Given: the "Triple-Signal" of 650Hz and 700Hz and 750Hz ,  

are the only signals that will pass through the filter stages,

* Then: the GOAL is to see if usable data can be provided by SPICE,

such that some GUIDELINES could be formed

for band-pass filter design, based on the realistic "Triple-Signal".

In re-thinking my goal, you are moving my thinking closer

to a clearer definition of this experiment.

If the Triple-Signal is analogous to a Single-Frequency  ,

then Phase-Delay and Group-Delay may be the same, practically.

* Given: the limits of the C.W. communication scenario

( narrow passbands of Q=5 )

* Then: this "Triple-Signal" is the "Target" to be investigated. 

* Therefore: this project will continue investigating

phase-relationships based on

the "Triple-Signal" Narrow-Group-Delay.

Glen

edited: 2018-10-13

To date: This project has been constructed

and tested in real-time conditions for more than three years

on U.S.A. Amateur Radio Bands for CW (morse) receiving

with good results.

The measurable (O'scope) observation is that the signal side-band energy

which is filtered (attenuated) is captured (contained) in the base-band noise

and can be measurably observed as sub/supra harmonics.

Because the experimental part of the project utilizes the "Triple-Signal"

( in place of the common music "Wide-Band-Audio" )

it is easy to observe this activity.

* Use of the "Triple-Signal", as realistic representation of typical audio

present in CW (morse) Communication signals,

has enabled the use of SPICE to show the phase-delay of these audio signals.

Therefore, to answer my initial "Question" ,

* SPICE can be used to analyze "Group-Delay" and "Phase-Shift"

IF a "Narrow-Band-Audio-Signal" is used as the sample signal.

Examples of the developmental circuitry and examples of the Spice Plots

are presented on my website: https://www.geocities.ws/glene77is

Glen

I'm not sure what the question is exactly.

SPICE can do .AC analysis, and thereby produce phaseshift. In, e.g., LTspice, group-delay can be plotted instead of phaseshift. For real-time signals, one would go for the FFT, or possibly the MFT (momentary Fourier transform).

That is from the project, (APF) which is now finished, and written.

At that time, I was experimenting in uncharted territory.

There were no clear definitions of phase-shift distortion

surrounding the band-pass peak, in a resonant narrow audio filter.

I researched several major band-pass narrow audio filters.

The circuit I had settled on was the the Deliyannis-Friend-Multiple-FeedBack filter, which contains both Low-Pass and High-Pass circuitry, sharing a common current source.

...

The first project (AFX) was prototyped and used for four years

in my amateur radio station (K4KKQ), very successfully.

The phase induced dual-notches were very effective.

...

You read the APF project :

https://www.researchgate.net/post/We_examine_a_Variant_All-Pass_Band-Pass_Narrow_Filter_circuit

...

That was the second project (APF), exploring radical phase filtering emphasizing the simple "All-Pass filter".

...

The second project (APF) was based on exploration/experimenting

on the use of All-Pass Filters, and worked very well.

I received comments from several Ph.D. and E.E. engineers, including

Dr. Barrie Gilbert of Analog Designs, Inc., Beaverton, Or, USA.

...

The APF was a very unusual design,

involving many sequential All-Pass filters

to produce a polynomial progression of notched signals ,

which I followed with several Resonant Band-Pass filters

to select a span of two notches around a 700 hertz peak.

The sequential all-pass filter design was not found in any literature,

but the APF design produced phase induced dual-notches

which were very effective. You must see the plots to understand.

This All-Pass project was very unlike the usual single-stage all-pass filter which is used for audio work. Very different, radical design.

...

Your suggestions are correct.

...

You may check my website

http://www.geocities.ws/glene77is/

I enjoyed many years of projects which were personally interesting

during my tenure at our local university.

I am long retired, now, and am involved in a good project.

I hope you find an interesting project !

to Marcel, "Thank you."

===