second

I'm doing some short-term runs on the ones I have around here, as well
(FE-5680A, PRS-10, LPRO).  Will be interesting to see how close ours are.

So far, the FE-5680A is doing well in the ADEV department, but its AD9832
DDS chip adds some substantial noise and spurs to the 10 MHz output.  Does
yours have that chip?  I understand that some of them didn't.

The overall SSB C/N ratio I'm seeing between 1 Hz - 100 kHz is pretty much
in line with Analog Devices' data sheet, as are the spur levels.

-- john

Once my first unit has warmed up properly I'll be curious to see how the adev compares (I can't measure phase noise.)  If any one is interested I've also got some adev data from my one working temex lpfrs Rb unit.  (I also have another unit that only locks once in a while.)    At approx 1/3 the price the 5680 seems like a better bargain especially if the 1pps output works.

Thanks in advance to those who can properly evaluate these units.

Regards Mark S

On 2012-02-03, at 7:19 PM, "John Miles" jmiles@pop.net wrote:

In the current 5680A units, the 10 MHz output comes from the 60 MHz VCXO (divided by 6 in CPLD) and not direct from a DDS. If my architecture understanding is right, the DDS signal output is mixed with the VCXO output only at the 114th harmonic of 60 MHz, and it's the PLL (looking at the optical signal from the Rb) that drives the VCXO to keep it lined up.  As I understand it, DDS phase noise should be divided by a factor of 6*114 by the time it appears at the 10 MHz output, and at larger frequency offsets the amplitude should also be (significantly) reduced by the PLL loop filter.

I don't have any phase noise measurement tools myself, so this is just an academic argument, but if there is significantly more noise on the 10 MHz than expected for a 60 MHz VCXO, I wonder if it's just inadequate filtering of an internal power rail.  Is the unit under test being driven by a linear, or switching supply?

output

optical

understand

amplitude

filtering of

or

What made me suspect the AD9832 is that the PN/spur behavior is so close to
what I'd expect from it.  Analog Devices' own plots contain a lot of strong
spurs, and the chip's actual SFDR spec is in the -70 dBc range that I'm
seeing (blue trace attached).  But yes, you're right, it would make a lot
more sense to use the DDS as an offset generator for the control loop,
rather than as an output device.

I tried it with a couple of different supplies, linear and otherwise, and
the noise didn't change meaningfully.  Since it's so far out of line with
FEI's specs, I have to believe that I either got a couple of bad examples --
although they lock very quickly and look essentially unused inside -- or I
don't have the pinout right, and am using the wrong ground or something like
that.  Seems pretty unlikely that this is normal operation.  Let's see what
JohnA comes up with.

-- john

On 2/4/2012 12:17 AM, John Miles wrote:

John,

From your comments, it sounds like you may be measuring one of the
earlier 5680A's that could be tuned over a large range with a DDS output.

These earlier ones had a 50.25 MHz internal osc which was locked after
multiplication to the rubidium frequency. Then there was a DDS on the
output that could be programmed over a wide frequency range.

The newer ones have a 60 MHz internal oscillator. This is multiplied up
to 6840 MHz and subtractively mixed with ~5.3125 MHz from a DDS to get
to the rubidium frequency. The feedback from the rubidium cell locks the
60 Hz which is divided by 6 for the 10 MHz output. No DDS on the output
generation. The DDS in the loop can digitally adjust the rubidium lock
frequency that tunes the 60 MHz, to fine tune the output only around 10
MHz.

Do you know which type of 5680A you are measuring?

That sounds plausible.  I haven't taken the time to bring myself 100% up to
speed with the many different variations and options for these little boxes,
but I could swear I'm looking at a low-resolution, unfiltered, uncleaned-up
DDS.

This one is marked S/N 0339-65969, purchased from
http://www.ebay.com/itm/260930018124 .  It seems to have the pinout
documented at http://vk2xv.djirra.com/tech_rubidium.htm for serial
0127-96634 (pin 1=V+, pin 2=GND, pin 3=+5V, pin 7=RF out.)  However, VK2XV
claims that s/n 96634 was non-programmable.  I haven't tried hooking up a
serial terminal to mine, but I did notice that there was some negative
voltage on one of the other pins (9?) that might correspond to an RS-232
signal level.

Can't complain too much at $40/each, anyway  I believe the seller's claim
that these are new or nearly so, even though they look like they've been
removed from an installation.  The internal foam insulation is usually
discolored from heat, while it looks great in the two units that I bought
from this seller.

-- john

Hello, John,

Your phase noise plot seems quite similar to that sent by Said some time
ago (4 Jan), that looks like a lot of spurs quite evenly separated. It
is likely from the price and p/n that your unit is one of the "newer"
ones with 60MHz xtal and the DDS inside the loop - and I would think
that the pn from the DDS would be removed by the loop. But since the
division from 60MHz is done by a CPLD, perhaps the signal is corrupted
by this CPLD, or perhaps it is corrupted by a switching regulator inside
the unit.

The 60MHz signal is available inside the unit, in a small coaxial
connector. Perhaps it would be interesting to also analyze it.

Regards,

Javier

El 04/02/2012 10:43, John Miles escribió:

On 04/02/12 04:19, John Miles wrote:

Consider that we have two basic generation of 5680A, one using the
50,255.. MHz VCXO and one using 60 MHz + DDS VCXO. The later can then
have a DDS output option.

So there can be three different 10 MHz qualities. It would be
interesting to see the ADEV, TDEV and phase noise of all three.

Cheers,
Magnus