Re: [time-nuts] Secondary phase noise standard & FE405

[Bob Camp]

Hi

The “sure fire” way to reduce the spurs is to go to a divisor that “fits” in to 
the DAC bit width. If you have a 12 bit dac, those points happen at Fclock / 
2^12.
For Time Nut sort of stuff that’s pretty coarse tuning. 

Bob

> On Nov 13, 2016, at 11:03 AM, Tim Shoppa  wrote:
> 
> Some AD DDS app notes give examples of spurs and choosing nearby (but not
> exactly on freq) numbers that are much less bad for spurs - or at least
> that move the spurs outside the cleanup filters/loops. I don't know of a
> general example or even code that does this in a general way.
> 
> One paper I like on the subject is this:
> http://ttcla.org/vsreinhardt/DDS%20spur%20reduction%20techniques.pdf
> 
> P.S. I just like saying "Wheatley Jitter Injector". Not for those who are
> embarrassed easily... Almost as good as "Wankel Rotary Engine".
> 
> One of
> 
> On Sat, Nov 12, 2016 at 8:13 PM, Hal Murray  wrote:
> 
>> 
>> kb...@n1k.org said:
>>> Yes, the FE-405 uses a DDS and a cleanup. Inside the cleanup loop the DDS
>>> spurs come  straight through. Since the FE-405 compensates for all sorts
>> of
>>> things, the DDS moves around a lot. Even a one bit change on a DDS will
>> move
>>> spurs around. With an ever changing  DDS, you have an ever changing
>> forest
>>> of “stuff” on the output.
>> 
>> Is there a web page or such telling me where the spurs will be on a DDS
>> for a
>> particular constant?
>> 
>> 
>> --
>> These are my opinions.  I hate spam.
>> 
>> 
>> 
>> 
>> ___
>> time-nuts mailing list -- time-nuts@febo.com
>> To unsubscribe, go to https://www.febo.com/cgi-bin/
>> mailman/listinfo/time-nuts
>> and follow the instructions there.
>> 
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Secondary phase noise standard & FE405

[Tim Shoppa]

Some AD DDS app notes give examples of spurs and choosing nearby (but not
exactly on freq) numbers that are much less bad for spurs - or at least
that move the spurs outside the cleanup filters/loops. I don't know of a
general example or even code that does this in a general way.

One paper I like on the subject is this:
http://ttcla.org/vsreinhardt/DDS%20spur%20reduction%20techniques.pdf

P.S. I just like saying "Wheatley Jitter Injector". Not for those who are
embarrassed easily... Almost as good as "Wankel Rotary Engine".

One of

On Sat, Nov 12, 2016 at 8:13 PM, Hal Murray  wrote:

>
> kb...@n1k.org said:
> > Yes, the FE-405 uses a DDS and a cleanup. Inside the cleanup loop the DDS
> > spurs come  straight through. Since the FE-405 compensates for all sorts
> of
> > things, the DDS moves around a lot. Even a one bit change on a DDS will
> move
> > spurs around. With an ever changing  DDS, you have an ever changing
> forest
> > of “stuff” on the output.
>
> Is there a web page or such telling me where the spurs will be on a DDS
> for a
> particular constant?
>
>
> --
> These are my opinions.  I hate spam.
>
>
>
>
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/time-nuts
> and follow the instructions there.
>
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Secondary phase noise standard & FE405

[Bob Camp]

Hi


> On Nov 13, 2016, at 2:55 AM, Gerhard Hoffmann  wrote:
> 
> Am 12.11.2016 um 23:52 schrieb Bob Camp:
>> Hi
>> 
>> Yes, the FE-405 uses a DDS and a cleanup. Inside the cleanup loop the DDS 
>> spurs come
>> straight through. Since the FE-405 compensates for all sorts of things, the 
>> DDS moves
>> around a lot. Even a one bit change on a DDS will move spurs around. With an 
>> ever changing
>> DDS, you have an ever changing forest of “stuff” on the output.
> I would not expect a lot of dynamics after a month of cont. power up. And the 
> spurs
> do not move and they don't average away, at least not within my limits of 
> boredom.
> 
> The PLL bandwidth seems to be 50 Hz, so the forest from 100KHz to 5MHz is
> far out. That seems somewhat weird and raised doubt together
> with the power level being out of spec.

The spurs you need to watch are under 50Hz. They are the ones from the DDS. The 
higher
stuff is from the microprocessor. 

Bob

> 
>> Not the best thing for a “standard” …..
> I've tried to kill 2 birds with one stone. The FE405 had collected dust for
> months and still was mostly untested.
> 
> The oscillator is not part of the "standard"; I could have used a BVA or
> a maser by just plugging in. I just needed a centre frequency with a
> few dBm for the E5052B to lock on.
> 
> All I want is some known phase noise reference reference lines
> for my private phase detector experiments. Recalibating the
> phase detector by modulating oscillators on different frequencies
> is no fun if you change the Schottkies every 2 minutes.
> 
> regards, Gerhard
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Secondary phase noise standard & FE405

[Gerhard Hoffmann]

Am 12.11.2016 um 23:52 schrieb Bob Camp:

Hi

Yes, the FE-405 uses a DDS and a cleanup. Inside the cleanup loop the DDS spurs 
come
straight through. Since the FE-405 compensates for all sorts of things, the DDS 
moves
around a lot. Even a one bit change on a DDS will move spurs around. With an 
ever changing
DDS, you have an ever changing forest of “stuff” on the output.
I would not expect a lot of dynamics after a month of cont. power up. 
And the spurs
do not move and they don't average away, at least not within my limits 
of boredom.


The PLL bandwidth seems to be 50 Hz, so the forest from 100KHz to 5MHz is
far out. That seems somewhat weird and raised doubt together
with the power level being out of spec.


Not the best thing for a “standard” …..

I've tried to kill 2 birds with one stone. The FE405 had collected dust for
months and still was mostly untested.

The oscillator is not part of the "standard"; I could have used a BVA or
a maser by just plugging in. I just needed a centre frequency with a
few dBm for the E5052B to lock on.

All I want is some known phase noise reference reference lines
for my private phase detector experiments. Recalibating the
phase detector by modulating oscillators on different frequencies
is no fun if you change the Schottkies every 2 minutes.

regards, Gerhard
___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Secondary phase noise standard & FE405

[Bob Camp]

Hi

It depends both on the DDS “firmware” and the DAC linearity. You can play games 
to 
come up with the firmware side of it. The normal approach is to design the part 
so the 
DAC dominates. More or less, "more firmware bits”  is cheaper than improving 
the DAC. 

Bob

> On Nov 12, 2016, at 8:13 PM, Hal Murray  wrote:
> 
> 
> kb...@n1k.org said:
>> Yes, the FE-405 uses a DDS and a cleanup. Inside the cleanup loop the DDS
>> spurs come  straight through. Since the FE-405 compensates for all sorts of
>> things, the DDS moves around a lot. Even a one bit change on a DDS will move
>> spurs around. With an ever changing  DDS, you have an ever changing forest
>> of “stuff” on the output.  
> 
> Is there a web page or such telling me where the spurs will be on a DDS for a 
> particular constant?
> 
> 
> -- 
> These are my opinions.  I hate spam.
> 
> 
> 
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Secondary phase noise standard & FE405

[Hal Murray]

kb...@n1k.org said:
> Yes, the FE-405 uses a DDS and a cleanup. Inside the cleanup loop the DDS
> spurs come  straight through. Since the FE-405 compensates for all sorts of
> things, the DDS moves around a lot. Even a one bit change on a DDS will move
> spurs around. With an ever changing  DDS, you have an ever changing forest
> of “stuff” on the output.  

Is there a web page or such telling me where the spurs will be on a DDS for a 
particular constant?


-- 
These are my opinions.  I hate spam.



___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] Secondary phase noise standard & FE405

[Bob Camp]

Hi

Yes, the FE-405 uses a DDS and a cleanup. Inside the cleanup loop the DDS spurs 
come 
straight through. Since the FE-405 compensates for all sorts of things, the DDS 
moves
around a lot. Even a one bit change on a DDS will move spurs around. With an 
ever changing 
DDS, you have an ever changing forest of “stuff” on the output. 

Not the best thing for a “standard” …..

Bob



> On Nov 12, 2016, at 4:24 PM, Gerhard Hoffmann  wrote:
> 
> I have made an experimental secondary phase noise standard, as F.Walls
> 
> would call it. In the end, it's inspired by his paper.
> 
> 
> There is not much to it. There is a Mini Circuits PSC2-1 splitter that is
> 
> fed from an oscillator or other source. The splitter divides it into a
> 
> clean output that goes to another SMA.
> 
> The other half is to be dirtyfied. It is fed into another PSC2-1 that is
> 
> used as a combiner. Its output goes to the signal source analyzer via
> 
> a third SMA.
> 
> 
> The dirt input is either  a 4th SMA or a noise generator. That is switched
> 
> by reed relays. The noise generator consists of a 60 Ohm resistor and
> 
> currently 2 LMH6702 stages for +20 and +40 dB. I'm not sure if the
> 
> 60R really defines the level, but the gain is at least flat from AF
> 
> to > 100 MHz. If not yet, it can be made so. I have already a
> 
> solid step attenuator so I can get rid of those reed relays.
> 
> 
> I checked it this evening with my FE405 osc I bought earlier this year
> 
> and that has enjoyed a good month of continuous running by now.
> 
> I observed
> 
> 1. that the FEI delivers only 3.5 dBm
> 
> 2. that there are lots of birdies far out. Is that normal, maybe from the 
> DDS?
> 
> 
> The addition of the noise reference lines seems to work properly.
> 
> oscillator:
> 
> < 
> https://www.flickr.com/photos/137684711@N07/30824608092/in/album-72157662535945536/
>  >
> 
> low calibration line:
> 
> < 
> https://www.flickr.com/photos/137684711@N07/30824607162/in/album-72157662535945536/
>  >
> 
> high calibration line:
> 
> < 
> https://www.flickr.com/photos/137684711@N07/30824607552/in/album-72157662535945536/
>  >
> 
> Later this evening I'll add the pic of the tin box. I need a USB cable...
> 
> :-)  Gerhard
> 
> 
> ___
> time-nuts mailing list -- time-nuts@febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

[time-nuts] Secondary phase noise standard & FE405

[Gerhard Hoffmann]

I have made an experimental secondary phase noise standard, as F.Walls

would call it. In the end, it's inspired by his paper.


There is not much to it. There is a Mini Circuits PSC2-1 splitter that is

fed from an oscillator or other source. The splitter divides it into a

clean output that goes to another SMA.

The other half is to be dirtyfied. It is fed into another PSC2-1 that is

used as a combiner. Its output goes to the signal source analyzer via

a third SMA.


The dirt input is either  a 4th SMA or a noise generator. That is switched

by reed relays. The noise generator consists of a 60 Ohm resistor and

currently 2 LMH6702 stages for +20 and +40 dB. I'm not sure if the

60R really defines the level, but the gain is at least flat from AF

to > 100 MHz. If not yet, it can be made so. I have already a

solid step attenuator so I can get rid of those reed relays.


I checked it this evening with my FE405 osc I bought earlier this year

and that has enjoyed a good month of continuous running by now.

I observed

1. that the FEI delivers only 3.5 dBm

2. that there are lots of birdies far out. Is that normal, maybe 
from the DDS?



The addition of the noise reference lines seems to work properly.

oscillator:

< 
https://www.flickr.com/photos/137684711@N07/30824608092/in/album-72157662535945536/ 
>


low calibration line:

< 
https://www.flickr.com/photos/137684711@N07/30824607162/in/album-72157662535945536/ 
>


high calibration line:

< 
https://www.flickr.com/photos/137684711@N07/30824607552/in/album-72157662535945536/ 
>


Later this evening I'll add the pic of the tin box. I need a USB cable...

:-)  Gerhard


___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.