Phase-sequence networks?
Don Lancaster
In section 5-16 of Horowitz's ART OF ELECTRONICS is a phase sequence
network. Which appears to be an elegantly simple and obviously superior
replacement for the "Weaver" method of generating analog audio signals
differentially 90 degree phase shifted over a wide bandwidth.
AOE gives no futher clue, analysis looks painful, and web searching
hasn't yet been helpful. Does anyone here have or know of the horses'
whatever source paper or person for this method?
Basically, its a bunch of resistors and capacitors in a round-and-round
network. that directly outputs broadband phases of 0, 90, 180, and 360
degrees to high accuracy.
Super cute, even.
Please also reply via email.
--
Many thanks,
Don Lancaster
Synergetics Press 3860 West First Street Box 809 Thatcher, AZ 85552
Voice phone: (520) 428-4073 email: d...@tinaja.com
Visit my GURU's LAIR web site at http://www.tinaja.com
Know your acronymns: url = utterly rancid location
net = not entirely true
www = world wide wait
John Woodgate
In article <34909724...@tinaja.com>, Don Lancaster
<d...@tinaja.com> writes
>In section 5-16 of Horowitz's ART OF ELECTRONICS
Sorry, who is Horowitz (;-)?
--
Regards, John Woodgate, Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
OOO - Own Opinions Only. It is useless to threaten a strong man - he will
ignore you. It is dangerous to threaten a weak man - he will kill you if he can.
Leon Heller
>network. Which appears to be an elegantly simple and obviously superior
>replacement for the "Weaver" method of generating analog audio signals
>differentially 90 degree phase shifted over a wide bandwidth.
>
>AOE gives no futher clue, analysis looks painful, and web searching
>hasn't yet been helpful. Does anyone here have or know of the horses'
>whatever source paper or person for this method?
>
>Basically, its a bunch of resistors and capacitors in a round-and-round
>network. that directly outputs broadband phases of 0, 90, 180, and 360
>degrees to high accuracy.
>Super cute, even.
I first saw this circuit many years ago in Pat Hawker's Technical Topics
column in RadCom.
Leon
--
Leon Heller: le...@lfheller.demon.co.uk http://www.lfheller.demon.co.uk
Amateur Radio Callsign G1HSM Tel: +44 (0) 118 947 1424
See http://www.lfheller.demon.co.uk/rcm.htm for details of a
low-cost reconfigurable computing module using the XC6216 FPGA
Jim Thompson
For those of us who don't have ART OF ELECTRONICS could someone please
post this schematic or place on a web page? Sounds VERY interesting!
...Jim Thompson
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice: (602)460-2350 | Brass Rat |
| ana...@primenet.com Fax: (602)460-2142 | 1962 |
HEY, LOOK! IT'S NOT MY FAULT; IT'S SOME GUY NAMED "GENERAL PROTECTION".
(Ratbert, the Consultant, speaks, in Dilbert, by Scott Adams)
For proper E-mail replys remove "numeric" from E-mail address.
Winfield Hill
Jim Thompson, <ana...@primenet.com> said...
>
>> Don Lancaster <d...@tinaja.com> writes
>>> In section 5-16 of Horowitz's ART OF ELECTRONICS is a phase sequence
>>> network. Which appears to be an elegantly simple and obviously superior
>>> replacement for the "Weaver" method of generating analog audio signals
>>> differentially 90 degree phase shifted over a wide bandwidth.
>>>
>>> network. that directly outputs broadband phases of 0, 90, 180, and 360
>>> degrees to high accuracy. Super cute, even.
>
> post this schematic or place on a web page? Sounds VERY interesting!
> ...Jim Thompson
Jim, I'll be happy to FAX you the pages in question,
but you should get yourself a copy, you'll like it.
>| James E.Thompson, P.E. |
>| Analog Innovations, Inc. |
>| Fax: (602)460-2142 |
--
Winfield Hill hi...@rowland.org _/_/_/ _/_/_/_/
The Rowland Institute for Science _/ _/ _/_/ _/
Cambridge, MA USA 02142-1297 _/_/_/_/ _/ _/ _/_/_/
_/ _/ _/ _/ _/
http://www.artofelectronics.com/ _/ _/ _/_/ _/_/_/_/
Walter Gray
In article <66upuv$n...@fridge.shore.net>, hi...@rowland.org (Winfield Hill) writes:
:Jim Thompson, <ana...@primenet.com> said...
:>
:>> Don Lancaster <d...@tinaja.com> writes
:>>> In section 5-16 of Horowitz's ART OF ELECTRONICS is a phase sequence
:>>> network. Which appears to be an elegantly simple and obviously superior
:>>> replacement for the "Weaver" method of generating analog audio signals
:>>> differentially 90 degree phase shifted over a wide bandwidth.
:>>>
:>>> Basically, its a bunch of resistors and capacitors in a round-and-round
:>>> network. that directly outputs broadband phases of 0, 90, 180, and 360
:>>> degrees to high accuracy. Super cute, even.
:>
:> For those of us who don't have ART OF ELECTRONICS could someone please
:> post this schematic or place on a web page? Sounds VERY interesting!
:> ...Jim Thompson
:
: Jim, I'll be happy to FAX you the pages in question,
: but you should get yourself a copy, you'll like it.
Man! That is _salesmanship_!!!
Walter
Disclaimer: My employer is not responsible for this stuff.
Winfield Hill
Jim Thompson, <ana...@primenet.com> said...
>
>> Don Lancaster <d...@tinaja.com> writes
>>> In section 5-16 of Horowitz's ART OF ELECTRONICS is a phase sequence
>>> network. Which appears to be an elegantly simple and obviously superior
>>> replacement for the "Weaver" method of generating analog audio signals
>>> differentially 90 degree phase shifted over a wide bandwidth.
>>>
>>> Basically, its a bunch of resistors and capacitors in a round-and-round
>>> network. that directly outputs broadband phases of 0, 90, 180, and 360
>>> degrees to high accuracy. Super cute, even.
>
> For those of us who don't have ART OF ELECTRONICS could someone please
> post this schematic or place on a web page? Sounds VERY interesting!
> ...Jim Thompson
OK, Jim, per your request, since you're such a nice fellow!
Phasing-style SSB generators need 90-degree phase-shifted signals for both the
RF and audio inputs. The RF stage is easy, because it's a single frequency.
But phase shifting a range of frequencies, as is required for audio signals,
is difficult. Phase-sequence filters are sometimes used to perform this task.
Practising my ASCII art now, here's the phase-sequence circuit in AoE page 241,
fig 5.41. In the filter all the R's are the same, while the caps scale with
each section. C in the 1st section, C/2 in the 2nd, and C/4, C/8, ... in
successive sections. Four stages are shown, but 6 or 8 stages give much
better bandwidth.
-IN o------------------------+----------,
,----------+--------- | -------- | ---------,
| ,--- | -------- | -------- | -------- | ----,
+IN o--+- R -+-||-+- R -+-||-+- R -+-||-+- R -+-||-+ | C
balanced | ,--- | -------- | -------- | -------- | ----,
inputs '- R -+-||-+- R -+-||-+- R -+-||-+- R -+-||-' | C/2
| ,--- | -------- | -------- | -------- | ----,
C/4 '- R -+-||-+- R -+-||-+- R -+-||-+- R -+-||-' |
| | | | |
C/8 '- R -+-||-+- R -+-||-+- R -+-||-+- R -+-||-'
| | | |
o 0 deg o 90 deg o 180 deg o 270 deg
| | | |
| ,------------------' |
| | | ,------------------'
| | | |
o o o o
+ - + -
balanced drive outputs: 0-deg 90-deg
As several have mentioned, this circuit, while difficult to analyze
analytically, is very easy to characterize using SPICE modelling.
Doing so, it's interesting to observe that the circuit has an ideal 90-degree
phase shift over many decades for the differences between adjacent quadrature
outputs. Sadly, this doesn't translate into a 90-degree phase shift between
the two pairs of balanced-drive outputs, over the same frequency range. This
is because the adjacent quadrature signals have widely diverging amplitudes
beyond the center frequency decade, with one rapidly dropping to zero, and the
other rising to equal the input.
This table shows the how the filter's operational bandwidth is increased by
adding stages. As stages are added, the upper frequency range of the filter
is entended (the next capacitor is C/16 and then C/32 etc.), increasing the
center frequency. For example, a 6 stage filter using C=0.1uF would require
R = 10k for a center frequency of 850Hz, and would operate from 135 Hz to
5.4kHz, providing a constant 90-degree phase shift with two +0.9-degree ripple
peaks and two -0.4-degree dips.
stages 4 6 8
--- --- ---
f_center = k/RC 0.35 0.85 1.8 parameter k
bandwidth ratio 11.5 40 150 phase ripple range
phase ripple 2.5 1.3 1.3 degrees ripple p-p
Hmm, my SPICE modelling didn't confirm our 100:1-range claim in AoE,
for 6 stages.