FA1 noise test was shown here:
http://lists.febo.com/pipermail/time-nuts_lists.febo.com/2019-August/097445.html

TAPR TICC noise test was shown here:
https://www.eevblog.com/forum/metrology/comparing-the-performance-of-17-different-gpsdos/?action=dlattach;attach=591979;image

Tom ran my TICC data through Timelab to get pretty plots:
http://leapsecond.com/u/sims/gpsdo17/

A frequency doubler and a frequency quintupler will be needed.

Options for the quintupler:

1. Convert to square wave and isolate the desired 25MHz output using filters. A bandpass filter is a poor choice if high phase stability is required. A high pass filter supplemented with series tuned shunt traps for the unwanted harmonics can provide higher phase stability

2. Wenzel's odd harmonic diode multiplier will also work:
   http://techlib.com/files/RFDesign2.pdf

Options for the doubler

1. The NIST JFET based frequency doubler

2. A BJT based variant of the above

3. A diode based frequency doubler (full wave rectifier plus output filter)

Bruce

Am 08.09.19 um 14:02 schrieb Paul Bicknell:

5 MHz -> 10 MHz is easy.

5 MHz -> 25 is harder. You can use a transistor biased to class B

with tuned circuits at the output, but to get an acceptable efficiency,

you'll have to adjust the collector current duty cycle, which is

f(input power, bias, etc) to keep Mr. Fourier happy.

You'll also need substantial filtering at the output.

Narrow filters are a natural enemy of phase stability.

You can find this & that useful here:
http://www.wenzel.com/documents/circuits1.htm

For 5->10 MHz, you can use something like the output amplifier

of my OCXO support board V2.0 and operate it in push-push doubler style.

See attachment. The amplifier/doubler is just the narrow strip below the
SMA

connector in the photo.

While I'm at it:

I've got the V2.0 boards just last week and tested it only with a Morion
MV89A

(on the back side) and the output amplifier in push pull without doubling.

With a 12V supply, it is easy to get +23 dBm out in nice quality. I have

also seen +26 dBm with some distortion. That could probably be healed

with adjusted bias, but for me +23 dBm is plenty. Enough to drive any

ring mixer or Timepod input even through a power splitter.

One could use a 1:4 splitter for 4 outputs with identical phase noise

and still have substantial signal power.

Watch your power meter or spectrum analyzer input. They might burn.

RF Transformers are $1.99 or so at Digikey.  No DIY required.

There are 2 crystal or LC notches to reduce unwanted frequencies.

The Morion MV89 doubles internally from 5 to 10 MHz; on mine I can see

the 5 MHz fundamental on the scope. Every other 10 MHz sine is somewhat

higher or lower. Gives an unsteady scope picture. Uncaring.  :-(

The board can lock the local VCXO to an incoming frequency or 1pps
reference,

produce a 1pps_out from the VCXO (3V3 CMOS level into 50 Ohms, good enough

for TTL) and indicate the signal level of the incoming ref frequency
(temp comp. diode

rectifier).

Supported oscillators are HP10811A, Morion, MTI-260, ECOC-2522-100.000-3FC

(affordable 100 MHz SC oven, avail. at DigiKey, abt. €100 ) and some el
cheapo others.

2FF-Phase detector and 1pps generation are in a $2 Xilinx Coolrunner II.
There are

strapping options for (5, 10, 100, x)MHz, Vtune sense etc. and optional
volt. regulators.

There are also lots of unused pins on a 100mil grid for experiments.

At $5 total for 10 boards + transport from PCBway it is cheap enough to
use it

just for mechanical mounting.

regards, Gerhard

Hi

A lot depends on what your end use is and what you have to work with. There are
Altera FPGA boards that have built in PLL’s that will accept a 5 MHz input. You can
generate just about anything with them. 10 and 25 MHz (plus 1 pps) with multiple
outputs of each would be pretty easy. Phase noise / jitter might not be super duper,
but how good is your source? What are you driving? It may well be good enough.

Bob

Bert was looking at the ICS512 and have to agree the price is cheap. How
have you applied them. It seems really simple. Do you follow with
filtering. Looking at 5 > 10 MHz and 5 > 15...
regards
Paul
WB8TSL

On Mon, Sep 9, 2019 at 7:03 AM ew via time-nuts time-nuts@lists.febo.com
wrote:

Hi just got in thank you all for your help I will have to do some recherché
during the week  just got a rubidium to fix first

Regards Paul B

-----Original Message-----
From: time-nuts [mailto:time-nuts-bounces@lists.febo.com] On Behalf Of paul
swed
Sent: 09 September 2019 16:05
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] 5 Mhz to 10 Mhz and 25 Mhz

Bert was looking at the ICS512 and have to agree the price is cheap. How
have you applied them. It seems really simple. Do you follow with
filtering. Looking at 5 > 10 MHz and 5 > 15...
regards
Paul
WB8TSL

On Mon, Sep 9, 2019 at 7:03 AM ew via time-nuts time-nuts@lists.febo.com
wrote:

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.8048 / Virus Database: 4793/15884 - Release Date: 08/14/18
Internal Virus Database is out of date.

You might look at the NB3N502  PLL mult chip. I've used this chip on my rubidium interface board and other projects.  It's cheap, current production and does a number of multipliers from 2X 2.5X 3X 3.333333x 4X and 5X  Available from Mouser and others.  I'm using one in a hybrid CW transmitter as a nearly coil-less scheme with a 6CL6 final.  all 5 bands from either a 80 meter or 40 meter crystal. Keys really clean too.
Bob, KE6F

-----Original Message-----
From: paul swed paulswedb@gmail.com
To: Discussion of precise time and frequency measurement time-nuts@lists.febo.com
Sent: Mon, Sep 9, 2019 9:01 am
Subject: Re: [time-nuts] 5 Mhz to 10 Mhz and 25 Mhz

Bert was looking at the ICS512 and have to agree the price is cheap. How
have you applied them. It seems really simple. Do you follow with
filtering. Looking at 5 > 10 MHz and 5 > 15...
regards
Paul
WB8TSL

On Mon, Sep 9, 2019 at 7:03 AM ew via time-nuts time-nuts@lists.febo.com
wrote:

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.

Have to say that chip looks identical to what Burt sent earlier. Settings
and all.

On Mon, Sep 9, 2019 at 3:01 PM Bob via time-nuts time-nuts@lists.febo.com
wrote: