[time-nuts] fast switching quiet synthesizer
pete at petelancashire.com
Tue Oct 7 14:23:25 EDT 2014
What is your definition of low noise / purity ?
DDS and low noise take some pretty good engineering.
What is the min frequency ?
What is the step size ?
As to PTS's it took me around an hour to make a cable write the C code to
convert to its interface needs. If I had
not lost the documentation I'd send it to you.
If your looking for new, look on the web a Microwave Journal's web site
a couple of the military mags http://www.militaryaerospace.com/index.html
How many steps in a test run ?
An 8645A can do 4K step at a time. But alas it is a pretty old instrument,
Kesight recommends the E8257D but starting
price is over $33K
On Tue, Oct 7, 2014 at 10:32 AM, Magnus Danielson <
magnus at rubidium.dyndns.org> wrote:
> You should be able to use DDS test-boards and by timing your last write,
> you should be able to time the frequency jump.
> The STEL-1173 takes 6 bytes, but writing the last one latches all 6 bytes
> over to a single 48 bit word. I expect that other DDSes have the same
> distinct transfer-phase if you only look in the datasheet for the details.
> Some of the modern DDSes can take 10 MHz directly and step it up
> internally before hitting the DDS core, but it may be that you need to
> synthesize a higher clock from the 10 MHz first.
> On 10/07/2014 07:02 PM, Jim Lux wrote:
>> At work, I'm putting together a multichannel stepped frequency CW radar
>> breadboard, and I'm looking for something to serve as a source that I
>> can step quickly.
>> I'm looking at stepping every millisecond or so. Right now, I use a
>> Ardunino type microcontroller driving a serial DAC driving a VCO, but
>> that's a bit wonky and noisy, although it's easy to get the step timing
>> right on. The spectral purity is, shall we say, downright ugly.
>> Since I'm going to be doing precision ranging with this, the spectral
>> purity has to be reasonably good (not 1E-15 at 1000 seconds good,
>> I was thinking about a PTS synthesizers (beloved of time-nuts for all
>> kinds of reason), and they're nice because they are quiet, and switch
>> really fast (microseconds). However, they all seem to have BCD or GPIB
>> interfaces (only). Sure, I can code up something on an Arduino or other
>> microcontroller to drive the BCD on the PTS, but maybe there's something
>> else out there that might work as well? And is already off the shelf.
>> I could hook a Prologix on the back of a PTS with GPIB, and hit it over
>> the ethernet, but I'm not sure I'd be able to get the steps to occur
>> when I want them (ethernet and determinism do not go well together).
>> Maybe some DDS in a box product? That will take my nice clean 10 MHz
>> Ultimately, I'm looking at output frequencies in single digit GHz, but
>> something that can be mixed/multiplied up will work just fine.
>> I'm looking for something that is off the shelf-ey as much as possible.
>> Using surplus gear is ok, because I really only need 3 or 4 channels and
>> that might be scroungeable, but spending hours wiring up weird adapters
>> or locating connectors that haven't been made since 1943 is something
>> I'd like to avoid.
>> time-nuts mailing list -- time-nuts at febo.com
>> To unsubscribe, go to
>> and follow the instructions there.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> and follow the instructions there.
More information about the time-nuts