Bit of IQ

On Fri, 31 May 2019 10:55:21 -0700, John Larkin wrote:

On Thu, 30 May 2019 22:47:18 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 8:34 PM, John Larkin wrote:
On Thu, 30 May 2019 18:41:01 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 5:30 PM, John Larkin wrote:
On Thu, 30 May 2019 16:46:03 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 4:40 PM, John Larkin wrote:
On Thu, 30 May 2019 17:25:05 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Thu, 30 May 2019 07:50:49 -0700) it happened John
Larkin <jjlarkin@highlandtechnology.com> wrote in
2brveelifcvcon0lbjbbstjejtsubbsj9m@4ax.com>:

On Thu, 30 May 2019 13:27:19 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

There is a very old way to generate SSB using 90 degrees phase
shifted audio and 90 degrees phase shifted RF.

Found old circuit on the web for 90 degrees audio phase shift,
modified it a bit:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_circuit_diagram_IMG_6958.JPG
1% capacitors, and 1% resistors made with trimpots:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_top_view_IMG_6962.JPG
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_bottom_view_IMG_6960.JPG

Stereo or mono in, + and - 45 phase shifted audio out.

That circuit, with design tables, is in the Williams filter
book.
Unfortunately, the outputs differ by 90 degrees but are not +-45
relative to the input. They squirm all over the place.

Yes, it shifts a bit, but for audio it makes litte difference I
think. Nice circles on the analog scope :)

For audio, or SSB generation, absolute phase doesn't matter. We
make an IQ modulator box to simulate eddy-current transducers, and
absolute phase "absolutely" matters.

The box that makes a true wideband 90 degree shift is a Hilbert
transform, which sadly is impossible. It's non-causal, like an
ideal lowpass filter; its output impulse response begins before
the input.




You can make narrowband ones that work pretty well. The major
problem with wideband ones is that both the central spike and the
long tail of the impulse response have infinities.

Cheers

Phil Hobbs

Like a lowpass, you can approximate the ideal transfer function, but
it adds delay. That's an oxymoron for a phase shifter.

We could cheat and add a PLL or something to shift our "RF" input,
assuming that a transducer excitation is going to be pretty
constant.


Or else calibrate the filter and fix it in software.

Cheers

Phil Hobbs

One of the things I like about our I/Q box is that it's 100% analog.
That's nice once in a while.

About 50% of our output is all-analogue, but that number is going down a
bit on account of auto-tweaking and interface issues. We really like
the LPC804 and LPC845 Cortex M0+ for low-level stuff. The 804 has a
reasonable amount of FPGA fabric--about the same as a CPLD costing twice
as much, not even counting the M0+ core.

Cheers

Phil Hobbs

There's not much logic in that one!

We're about ready to pick a new generation of uPs and FPGAs, and there
are lots of opinions here. I'm leaning towards using ZINQs. We already
use the 7020, a $125 monster with two ARM cores that usually runs Linux.
We need a lower-end part, and I'm thinking Zynq 7007, 400 balls, single
760 MHz ARM core and 23K logic cells. That's under $50 in quantity.
Hardware float is nice.

The '7007s is actually the same die as the (nominally larger) '7010.
They use fuses to disable one of the ARM cores, and it's only possible to
access about 70% of the FPGA fabric. This limitation is actually
enforced by the development software; the chip has 100% of the FPGA
fabric present (and tested).
The FPGA bitstream header contains an identifying code though, which
stops you from building a binary for the '7010 and loading it into the
'7007s.
The upside is that you can fully utilise "100%" of the '7007s fabric
without running into problems (as "100%" is really only about 70% of the
fabric, which is a fairly modest utilisation).

Other Xilinx Zynq-7 chips nobbled this way:
'7020 => '7014S
'7015 => '7012S

More recent families (e.g. Ultrascale Plus) do a similar thing, e.g.
XCVU5P and XCVU7P have the same pair of dies but very different prices.

Regards,
Allan
 
On Sat, 01 Jun 2019 06:39:17 -0500, Allan Herriman
<allanherriman@hotmail.com> wrote:

On Fri, 31 May 2019 10:55:21 -0700, John Larkin wrote:

On Thu, 30 May 2019 22:47:18 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 8:34 PM, John Larkin wrote:
On Thu, 30 May 2019 18:41:01 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 5:30 PM, John Larkin wrote:
On Thu, 30 May 2019 16:46:03 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 4:40 PM, John Larkin wrote:
On Thu, 30 May 2019 17:25:05 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Thu, 30 May 2019 07:50:49 -0700) it happened John
Larkin <jjlarkin@highlandtechnology.com> wrote in
2brveelifcvcon0lbjbbstjejtsubbsj9m@4ax.com>:

On Thu, 30 May 2019 13:27:19 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

There is a very old way to generate SSB using 90 degrees phase
shifted audio and 90 degrees phase shifted RF.

Found old circuit on the web for 90 degrees audio phase shift,
modified it a bit:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_circuit_diagram_IMG_6958.JPG
1% capacitors, and 1% resistors made with trimpots:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_top_view_IMG_6962.JPG
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_bottom_view_IMG_6960.JPG

Stereo or mono in, + and - 45 phase shifted audio out.

That circuit, with design tables, is in the Williams filter
book.
Unfortunately, the outputs differ by 90 degrees but are not +-45
relative to the input. They squirm all over the place.

Yes, it shifts a bit, but for audio it makes litte difference I
think. Nice circles on the analog scope :)

For audio, or SSB generation, absolute phase doesn't matter. We
make an IQ modulator box to simulate eddy-current transducers, and
absolute phase "absolutely" matters.

The box that makes a true wideband 90 degree shift is a Hilbert
transform, which sadly is impossible. It's non-causal, like an
ideal lowpass filter; its output impulse response begins before
the input.




You can make narrowband ones that work pretty well. The major
problem with wideband ones is that both the central spike and the
long tail of the impulse response have infinities.

Cheers

Phil Hobbs

Like a lowpass, you can approximate the ideal transfer function, but
it adds delay. That's an oxymoron for a phase shifter.

We could cheat and add a PLL or something to shift our "RF" input,
assuming that a transducer excitation is going to be pretty
constant.


Or else calibrate the filter and fix it in software.

Cheers

Phil Hobbs

One of the things I like about our I/Q box is that it's 100% analog.
That's nice once in a while.

About 50% of our output is all-analogue, but that number is going down a
bit on account of auto-tweaking and interface issues. We really like
the LPC804 and LPC845 Cortex M0+ for low-level stuff. The 804 has a
reasonable amount of FPGA fabric--about the same as a CPLD costing twice
as much, not even counting the M0+ core.

Cheers

Phil Hobbs

There's not much logic in that one!

We're about ready to pick a new generation of uPs and FPGAs, and there
are lots of opinions here. I'm leaning towards using ZINQs. We already
use the 7020, a $125 monster with two ARM cores that usually runs Linux.
We need a lower-end part, and I'm thinking Zynq 7007, 400 balls, single
760 MHz ARM core and 23K logic cells. That's under $50 in quantity.
Hardware float is nice.

The '7007s is actually the same die as the (nominally larger) '7010.
They use fuses to disable one of the ARM cores, and it's only possible to
access about 70% of the FPGA fabric. This limitation is actually
enforced by the development software; the chip has 100% of the FPGA
fabric present (and tested).
The FPGA bitstream header contains an identifying code though, which
stops you from building a binary for the '7010 and loading it into the
'7007s.

It is nice to be able to drop the bigger part onto the PC board, even
if we have to pay more.

One 700 MHz ARM core is usually overkill, especially with an FPGA
attached.

The upside is that you can fully utilise "100%" of the '7007s fabric
without running into problems (as "100%" is really only about 70% of the
fabric, which is a fairly modest utilisation).

Other Xilinx Zynq-7 chips nobbled this way:
'7020 => '7014S
'7015 => '7012S

More recent families (e.g. Ultrascale Plus) do a similar thing, e.g.
XCVU5P and XCVU7P have the same pair of dies but very different prices.

Regards,
Allan

We have used some Altera chips with different specified numbers of
CLBs and different prices that turn out to be identical; either runs
the bit stream compiled for the other. Ditto speed and temperature
grades, apparently.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics
 
søndag den 2. juni 2019 kl. 02.26.43 UTC+2 skrev Phil Hobbs:
On 6/1/19 12:04 PM, John Larkin wrote:
On Sat, 01 Jun 2019 06:39:17 -0500, Allan Herriman
allanherriman@hotmail.com> wrote:

On Fri, 31 May 2019 10:55:21 -0700, John Larkin wrote:

On Thu, 30 May 2019 22:47:18 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 8:34 PM, John Larkin wrote:
On Thu, 30 May 2019 18:41:01 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 5:30 PM, John Larkin wrote:
On Thu, 30 May 2019 16:46:03 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 4:40 PM, John Larkin wrote:
On Thu, 30 May 2019 17:25:05 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Thu, 30 May 2019 07:50:49 -0700) it happened John
Larkin <jjlarkin@highlandtechnology.com> wrote in
2brveelifcvcon0lbjbbstjejtsubbsj9m@4ax.com>:

On Thu, 30 May 2019 13:27:19 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

There is a very old way to generate SSB using 90 degrees phase
shifted audio and 90 degrees phase shifted RF.

Found old circuit on the web for 90 degrees audio phase shift,
modified it a bit:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_circuit_diagram_IMG_6958.JPG
1% capacitors, and 1% resistors made with trimpots:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_top_view_IMG_6962.JPG
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_bottom_view_IMG_6960.JPG

Stereo or mono in, + and - 45 phase shifted audio out.

That circuit, with design tables, is in the Williams filter
book.
Unfortunately, the outputs differ by 90 degrees but are not +-45
relative to the input. They squirm all over the place.

Yes, it shifts a bit, but for audio it makes litte difference I
think. Nice circles on the analog scope :)

For audio, or SSB generation, absolute phase doesn't matter. We
make an IQ modulator box to simulate eddy-current transducers, and
absolute phase "absolutely" matters.

The box that makes a true wideband 90 degree shift is a Hilbert
transform, which sadly is impossible. It's non-causal, like an
ideal lowpass filter; its output impulse response begins before
the input.




You can make narrowband ones that work pretty well. The major
problem with wideband ones is that both the central spike and the
long tail of the impulse response have infinities.

Cheers

Phil Hobbs

Like a lowpass, you can approximate the ideal transfer function, but
it adds delay. That's an oxymoron for a phase shifter.

We could cheat and add a PLL or something to shift our "RF" input,
assuming that a transducer excitation is going to be pretty
constant.


Or else calibrate the filter and fix it in software.

Cheers

Phil Hobbs

One of the things I like about our I/Q box is that it's 100% analog..
That's nice once in a while.

About 50% of our output is all-analogue, but that number is going down a
bit on account of auto-tweaking and interface issues. We really like
the LPC804 and LPC845 Cortex M0+ for low-level stuff. The 804 has a
reasonable amount of FPGA fabric--about the same as a CPLD costing twice
as much, not even counting the M0+ core.

Cheers

Phil Hobbs

There's not much logic in that one!

We're about ready to pick a new generation of uPs and FPGAs, and there
are lots of opinions here. I'm leaning towards using ZINQs. We already
use the 7020, a $125 monster with two ARM cores that usually runs Linux.
We need a lower-end part, and I'm thinking Zynq 7007, 400 balls, single
760 MHz ARM core and 23K logic cells. That's under $50 in quantity.
Hardware float is nice.

The '7007s is actually the same die as the (nominally larger) '7010.
They use fuses to disable one of the ARM cores, and it's only possible to
access about 70% of the FPGA fabric. This limitation is actually
enforced by the development software; the chip has 100% of the FPGA
fabric present (and tested).
The FPGA bitstream header contains an identifying code though, which
stops you from building a binary for the '7010 and loading it into the
'7007s.

It is nice to be able to drop the bigger part onto the PC board, even
if we have to pay more.

One 700 MHz ARM core is usually overkill, especially with an FPGA
attached.

The upside is that you can fully utilise "100%" of the '7007s fabric
without running into problems (as "100%" is really only about 70% of the
fabric, which is a fairly modest utilisation).

Other Xilinx Zynq-7 chips nobbled this way:
'7020 => '7014S
'7015 => '7012S

More recent families (e.g. Ultrascale Plus) do a similar thing, e.g.
XCVU5P and XCVU7P have the same pair of dies but very different prices..

Regards,
Allan

We have used some Altera chips with different specified numbers of
CLBs and different prices that turn out to be identical; either runs
the bit stream compiled for the other. Ditto speed and temperature
grades, apparently.



Of course that might not continue to be the case, especially if there's
some yield problem in future--the mfr is within its rights to exceed
specs if it likes. ;)

a place I worked eons ago used a cheaper reduced memory version of a DSP,
someone had made a mistake in a linker file so few bytes of the memory that
wasn't supposed to be there was used, it worked just fine because it was the same die as the bigger chip. Until, the volume got so high the manufacture
made a new die that actually had less memory and the production ground to halt with everything failing and much panic
 
On 6/1/19 12:04 PM, John Larkin wrote:
On Sat, 01 Jun 2019 06:39:17 -0500, Allan Herriman
allanherriman@hotmail.com> wrote:

On Fri, 31 May 2019 10:55:21 -0700, John Larkin wrote:

On Thu, 30 May 2019 22:47:18 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 8:34 PM, John Larkin wrote:
On Thu, 30 May 2019 18:41:01 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 5:30 PM, John Larkin wrote:
On Thu, 30 May 2019 16:46:03 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/30/19 4:40 PM, John Larkin wrote:
On Thu, 30 May 2019 17:25:05 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Thu, 30 May 2019 07:50:49 -0700) it happened John
Larkin <jjlarkin@highlandtechnology.com> wrote in
2brveelifcvcon0lbjbbstjejtsubbsj9m@4ax.com>:

On Thu, 30 May 2019 13:27:19 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

There is a very old way to generate SSB using 90 degrees phase
shifted audio and 90 degrees phase shifted RF.

Found old circuit on the web for 90 degrees audio phase shift,
modified it a bit:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_circuit_diagram_IMG_6958.JPG
1% capacitors, and 1% resistors made with trimpots:
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_top_view_IMG_6962.JPG
http://panteltje.com/pub/
audio_90_degrees_phase_shifter_PCB_bottom_view_IMG_6960.JPG

Stereo or mono in, + and - 45 phase shifted audio out.

That circuit, with design tables, is in the Williams filter
book.
Unfortunately, the outputs differ by 90 degrees but are not +-45
relative to the input. They squirm all over the place.

Yes, it shifts a bit, but for audio it makes litte difference I
think. Nice circles on the analog scope :)

For audio, or SSB generation, absolute phase doesn't matter. We
make an IQ modulator box to simulate eddy-current transducers, and
absolute phase "absolutely" matters.

The box that makes a true wideband 90 degree shift is a Hilbert
transform, which sadly is impossible. It's non-causal, like an
ideal lowpass filter; its output impulse response begins before
the input.




You can make narrowband ones that work pretty well. The major
problem with wideband ones is that both the central spike and the
long tail of the impulse response have infinities.

Cheers

Phil Hobbs

Like a lowpass, you can approximate the ideal transfer function, but
it adds delay. That's an oxymoron for a phase shifter.

We could cheat and add a PLL or something to shift our "RF" input,
assuming that a transducer excitation is going to be pretty
constant.


Or else calibrate the filter and fix it in software.

Cheers

Phil Hobbs

One of the things I like about our I/Q box is that it's 100% analog.
That's nice once in a while.

About 50% of our output is all-analogue, but that number is going down a
bit on account of auto-tweaking and interface issues. We really like
the LPC804 and LPC845 Cortex M0+ for low-level stuff. The 804 has a
reasonable amount of FPGA fabric--about the same as a CPLD costing twice
as much, not even counting the M0+ core.

Cheers

Phil Hobbs

There's not much logic in that one!

We're about ready to pick a new generation of uPs and FPGAs, and there
are lots of opinions here. I'm leaning towards using ZINQs. We already
use the 7020, a $125 monster with two ARM cores that usually runs Linux.
We need a lower-end part, and I'm thinking Zynq 7007, 400 balls, single
760 MHz ARM core and 23K logic cells. That's under $50 in quantity.
Hardware float is nice.

The '7007s is actually the same die as the (nominally larger) '7010.
They use fuses to disable one of the ARM cores, and it's only possible to
access about 70% of the FPGA fabric. This limitation is actually
enforced by the development software; the chip has 100% of the FPGA
fabric present (and tested).
The FPGA bitstream header contains an identifying code though, which
stops you from building a binary for the '7010 and loading it into the
'7007s.

It is nice to be able to drop the bigger part onto the PC board, even
if we have to pay more.

One 700 MHz ARM core is usually overkill, especially with an FPGA
attached.

The upside is that you can fully utilise "100%" of the '7007s fabric
without running into problems (as "100%" is really only about 70% of the
fabric, which is a fairly modest utilisation).

Other Xilinx Zynq-7 chips nobbled this way:
'7020 => '7014S
'7015 => '7012S

More recent families (e.g. Ultrascale Plus) do a similar thing, e.g.
XCVU5P and XCVU7P have the same pair of dies but very different prices.

Regards,
Allan

We have used some Altera chips with different specified numbers of
CLBs and different prices that turn out to be identical; either runs
the bit stream compiled for the other. Ditto speed and temperature
grades, apparently.

Of course that might not continue to be the case, especially if there's
some yield problem in future--the mfr is within its rights to exceed
specs if it likes. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com
 

Welcome to EDABoard.com

Sponsor

Back
Top