DDS questions...

[server]

On Fri, 12 Aug 2022 07:10:56 -0700, John Larkin
<jjlarkin@highlandtechnology.com> wrote:

On Fri, 12 Aug 2022 16:54:04 +0300, upsidedown@downunder.com wrote:

On Wed, 10 Aug 2022 15:42:00 -0700 (PDT), Ricky

<snip>

If the purpose is to create a variable _timing_ generator (not just
frequency generator), why mess with the DDS principle at all ?

Most of our customers expect to set an internal trigger frequency.
There are times when setting it to high resolution is valuable.


Using a divide-by-N counter clocked at say, 1 GHz, you can timing
intervals in 1 ns steps. With a 48 bit synchronous down counter, you
can get timing intervals of several days with 1 ns timing steps. Some
trickery is needed to avoid running all 48 stages at full ECL speed.

I can\'t do that in an FPGA. And resolution is mediocre around 10 MHz.

Use a fast 8 bit down counter followed by a slow 40 bit down counter.
If the fast counter is run at 1 GHz (1 ns), the slow counter only
runs at 4 MHz. When the slow down counter reaches 0x0000000000, it can
start reloading the preset value. At that time the fast counter is at
0xFF and it takes 256 ns before reaching zero and doing the preset. At
that time the slow counter has already been reloaded and it can start
counting as soon as the preset fast counter reaches 0 the next time.

<snip>

But the real question is, do you really need nanosecond step size in
minutes, hours or day time scale ?

A straightforward DDS will have tons of period jitter at low
frequencies, which is ugly.

Then do not use DDS directly for very low frequencies.

And some customers whine if we stop
triggering while we reprogram a DDS (or a synth chip) and a divisor.

Reprogramming DDS = loading a new addend value into the DDS. After
that the phase accumulator increases more or less rapidly, so quite
hard to even detect in a short time.

Admittedly, the 1 ns timing step is quite coarse at short pulses, inn
which only 1 ns, 2 ns, 3 ns, 4 ns and so on is available, so a DDS
might be justified to get 1 ps timing steps. But for longer times,
say 1 us (1 MHz) or 1 ms (1 kHz), why not put a divide-by-N divider
after the DDS ? Combining the DDS and divide-by-N programming, quite
strange periods can be obtained.

If you have a DDS for periods shorter than 1 s, you could add a
divide-by-N for longer periods. Each time the divide-by-N reaches 0,
you could enable the DDS addend loading, thus the timing would be
quite clean, even for a time sweep. Of course, this will require
precalculating the DDS addend and the divide-by-N before the divider
expires, but a very primitive CPU could do it before a sweep or other
predictable sequence.

 

[John Miles, KE5FX]

On Friday, August 12, 2022 at 9:08:27 AM UTC-7, Mike Monett wrote:

A New Frequency Synthesis Technique
. . .
The description of Rational Approximation Synthesis starts on page 151. A
block diagram is on page 156.

Amusing. That wasn\'t exactly \'new\' when I implemented it in 2002 or so.
( http://www.ke5fx.com/synth.html ) It was probably still covered
by one Qualcomm patent or another at the time, in retrospect.

I used a crystal filter rather than a narrowband VCXO, but same difference.

-- john, KE5FX

 

[Mike Monett]

\"John Miles, KE5FX\" <jmiles@gmail.com> wrote:

On Friday, August 12, 2022 at 9:08:27 AM UTC-7, Mike Monett wrote:
A New Frequency Synthesis Technique
. . .
The description of Rational Approximation Synthesis starts on page 151. A
block diagram is on page 156.

Amusing. That wasn\'t exactly \'new\' when I implemented it in 2002 or so.
( http://www.ke5fx.com/synth.html ) It was probably still covered
by one Qualcomm patent or another at the time, in retrospect.

I used a crystal filter rather than a narrowband VCXO, but same difference.

-- john, KE5FX

Amazing documentation. Thanks.

Mike



--
MRM

 

[John Miles, KE5FX]

On Saturday, August 13, 2022 at 8:18:00 PM UTC-7, Mike Monett wrote:
> Amazing documentation. Thanks.

Pretty awful synthesizer design, though.

-- john, KE5FX

 

[Mike Monett]

\"John Miles, KE5FX\" <jmiles@gmail.com> wrote:

On Saturday, August 13, 2022 at 8:18:00 PM UTC-7, Mike Monett wrote:
Amazing documentation. Thanks.

Pretty awful synthesizer design, though.

-- john, KE5FX

The performance is clearly superior to the ICOM. Congratulations.

I am absolutely amazed by the quantity and quality of your documentation.
This would take a normal human a year to complete.



--
MRM

 

[John Miles, KE5FX]

On Saturday, August 13, 2022 at 9:24:21 PM UTC-7, Mike Monett wrote:

I am absolutely amazed by the quantity and quality of your documentation.
This would take a normal human a year to complete.

Thanks! There was lots of room for improvement (the crystal filter/
comparator section is just embarrassing) but it performed well
on a $/dBc/Hz basis. And it did made a very nice article for QEX.

Nowadays, all that stuff fits on a single IC and provides 20 GHz of
coverage with far less noise of both white and flicker varieties. Of
course the best chips are unobtainium TI parts that have to be sourced
from Chinese scalpers at prices well into the 3-digit range, but that seems
to be the new way of the world.

-- john, KE5FX

 

[Mike Monett]

\"John Miles, KE5FX\" <jmiles@gmail.com> wrote:

On Saturday, August 13, 2022 at 9:24:21 PM UTC-7, Mike Monett wrote:
I am absolutely amazed by the quantity and quality of your
documentation. This would take a normal human a year to complete.

Thanks! There was lots of room for improvement (the crystal filter/
comparator section is just embarrassing) but it performed well
on a $/dBc/Hz basis. And it did made a very nice article for QEX.

Nowadays, all that stuff fits on a single IC and provides 20 GHz of
coverage with far less noise of both white and flicker varieties. Of
course the best chips are unobtainium TI parts that have to be sourced
from Chinese scalpers at prices well into the 3-digit range, but that
seems to be the new way of the world.

-- john, KE5FX

I\'m interested. Can you name some of the best and who sells them?




--
MRM

 

[John Miles, KE5FX]

On Saturday, August 13, 2022 at 11:04:08 PM UTC-7, Mike Monett wrote:
> I\'m interested. Can you name some of the best and who sells them?

LMX2820 (TI) and 8V97003 (Renesas) are two of the best integrated
PLL/VCO parts I\'ve seen. The LMX2595 has similar specs but the
LMX2820 has some nice advantages, such as lower flicker noise and
the ability to drive both phase detector inputs externally.

Look them up on octopart.com and you\'ll see that they are all
sitting in Chinese warehouses by the tens of thousands, waiting for
buyers who don\'t mind paying 4x-5x MSRP.

ADF4371 (62.5 MHz - 32 GHz) also looks good. Some stock left at
Mouser.

Of all these parts, I\'ve only messed with the LMX2820 in person.
I have a demo board for the Renesas part but haven\'t gotten
around to powering it up and trying it out yet.

-- john, KE5FX

 

[Gerhard Hoffmann]

Am 14.08.22 um 09:19 schrieb John Miles, KE5FX:

On Saturday, August 13, 2022 at 11:04:08 PM UTC-7, Mike Monett wrote:
I\'m interested. Can you name some of the best and who sells them?

LMX2820 (TI) and 8V97003 (Renesas) are two of the best integrated
PLL/VCO parts I\'ve seen. The LMX2595 has similar specs but the
LMX2820 has some nice advantages, such as lower flicker noise and
the ability to drive both phase detector inputs externally.

I have published a synthesizer based on LMX2594 (15 GHz) in
Dubus 1/22. I will put it on my web site sooner or later, but
the chip is currently vapourware.
<
https://www.flickr.com/photos/137684711@N07/51691780129/in/datetaken/ >

<
https://www.flickr.com/photos/137684711@N07/51519856398/in/dateposted-public/
>

There is 1 hittite HMC-451 output amplifier populated to drive
a HMC-220 ring mixer. The TI dongle & software cannot tell
it from an eval board.


There will be a companion clock generator with either a
100 MHz crystal oven or a CVHD-950. The oven/VCXO can be locked
to an external 10 MHz reference. It delivers
300 MHz for LMX2594 in fractional mode or
400 MHz for LMX2594 in integer mode.

400 MHz uses 2 SAW filters after the multiplier,
300 MHz uses 2*3pole LC, top coupled.
I could not find a nice 300 MHz SAW filter that is not EOL
or has less than 10 dB loss.

A previous version had some 10 MHz spurious from
the 1/10 prescaler. It also had only 300 MHz.

<
https://www.flickr.com/photos/137684711@N07/52284841519/in/dateposted-public/lightbox/
>

It could be used for a DDS also, but currently there is no
need to rush it because of the 2594.

Look them up on octopart.com and you\'ll see that they are all
sitting in Chinese warehouses by the tens of thousands, waiting for
buyers who don\'t mind paying 4x-5x MSRP.

ADF4371 (62.5 MHz - 32 GHz) also looks good. Some stock left at
Mouser.

At least it has some filtering after the output doubler.
The ADF5356 had a subharmonic that was horrible.

<
https://www.flickr.com/photos/137684711@N07/50403778976/in/dateposted-public/
>

If someone wants to play with these things, there are enough
boards. Hot air soldering is required, minimum.

cheers,
Gerhard DK4XP

 

[Mike Monett]

\"John Miles, KE5FX\" <jmiles@gmail.com> wrote:

On Saturday, August 13, 2022 at 11:04:08 PM UTC-7, Mike Monett wrote:
I\'m interested. Can you name some of the best and who sells them?

LMX2820 (TI) and 8V97003 (Renesas) are two of the best integrated
PLL/VCO parts I\'ve seen. The LMX2595 has similar specs but the
LMX2820 has some nice advantages, such as lower flicker noise and
the ability to drive both phase detector inputs externally.

Look them up on octopart.com and you\'ll see that they are all
sitting in Chinese warehouses by the tens of thousands, waiting for
buyers who don\'t mind paying 4x-5x MSRP.

ADF4371 (62.5 MHz - 32 GHz) also looks good. Some stock left at
Mouser.

Of all these parts, I\'ve only messed with the LMX2820 in person.
I have a demo board for the Renesas part but haven\'t gotten
around to powering it up and trying it out yet.

-- john, KE5FX

Very good info. Thanks



--
MRM

 

[Mike Monett]

Gerhard Hoffmann <dk4xp@arcor.de> wrote:

[...]

If someone wants to play with these things, there are enough
boards. Hot air soldering is required, minimum.

cheers,
Gerhard DK4XP

Very interesting. Thanks

Mike VE3BTI




--
MRM

 

[server]

On Sun, 14 Aug 2022 00:19:53 -0700 (PDT), \"John Miles, KE5FX\"
<jmiles@gmail.com> wrote:

On Saturday, August 13, 2022 at 11:04:08 PM UTC-7, Mike Monett wrote:
I\'m interested. Can you name some of the best and who sells them?

LMX2820 (TI) and 8V97003 (Renesas) are two of the best integrated
PLL/VCO parts I\'ve seen. The LMX2595 has similar specs but the
LMX2820 has some nice advantages, such as lower flicker noise and
the ability to drive both phase detector inputs externally.

Look them up on octopart.com and you\'ll see that they are all
sitting in Chinese warehouses by the tens of thousands, waiting for
buyers who don\'t mind paying 4x-5x MSRP.

ADF4371 (62.5 MHz - 32 GHz) also looks good. Some stock left at
Mouser.

Of all these parts, I\'ve only messed with the LMX2820 in person.
I have a demo board for the Renesas part but haven\'t gotten
around to powering it up and trying it out yet.

-- john, KE5FX

We use the LMX2571 as the internal trigger generator in one product.
Jitter is great. It was a huge pain to program, as TI doesn\'t reveal
the code that they use in the demo kit. Everybody has to figure it out
for themselves. The data sheet is 62 pages long and the app note is
another 31.

We have to suspend triggers while reprogramming the synth, which some
users don\'t like.