Hi Jim,

There are a few different schools of thought. Here in New England, I use a
very simple GPS locked 10 MHz oscillator. It's based on the Jupiter GPS
series with 10 kHz out. It drives a Qualcomm 1152 MHz board. This board
generates harmonics through 24 GHz.

I don't lock any of my rigs up to 24 GHz. From 47 GHz and up (78, 122, 241
GHz) the rigs are locked mostly using Axtal Axiom 75 series OCXO's and GPS
via VE1ALQ reflock boards.

For my rigs up to 24 GHz, I use the Qualcomm board to generate accurate
markers. From there, I can adjust my IF to compensate. The simple GPSDO
driving the Qualcomm is accurate to about 2 Hz/GHz, so even at 24 GHz, I'm
typically within about 50 Hz - well within my narrow CW filter on the IF.

The added benefit of using the GPS is I have it hooked to an ON4IY RoverBox.
This gives me grid square, and Sun position based on location/time for
aligning my dish (if the sun is out...)  Plus the simple GPSDO is locked and
running in about 3 minutes or so from a cold start.

I find I don't need "ultimate" accuracy from 24 GHz down. And not messing
with the crystals in LO's keeps phase noise down etc...

We've started to add Panadpaters to our IF rigs lately using the FUNCube SDR
initially and now the cheaper sticks. These let us see 96 kHz or better at
once, so finding the signal can be pretty easy.

Remember, locking is great if both ends are locked. If not, you're back to
tuning around to find the signal (that's where the panadapter is great!)

KT1J and I did our first 2 km 122 GHz contact a while back. We were both
locked and had rifle scopes calibrated for aiming. It was amazing to point
our dishes and start sending and instantly hear the other station with no
tuning!

73,
Mike, N1JEZ
"A closed mouth gathers no feet"

----- Original Message -----
From: "Jim Lux" jimlux@earthlink.net
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Sunday, March 10, 2013 10:23 AM
Subject: [time-nuts] frequency reference for portable operation

Hi

Remember - OCXO's are going to be acceleration sensitive. As you bump about on back roads, the oscillator is likely moving around by a few ppb. If you are after a hertz at 10 GHz, that's a lot.

Your GPS will be off by a fairly predictable amount based on it's idea of it's location. If you have a 3M position accuracy, then you will likely have 10 ns of time error. With a 10 second loop, you are right back to a ppb.

A good Rb should be able to hit a sub ppb accuracy at 10 minutes.

No perfect answer.

Bob

On Mar 10, 2013, at 10:23 AM, Jim Lux jimlux@earthlink.net wrote:

I agree with using an OCXO for amateur radio operation.

The main activity in the US is the 10 GHz and Up contest which takes
place over two weekends, mid-August and mid-September. I've been active
in most of them for the last 15 years. Full participation requires
operating something like 12 hrs each of the two weekend days.
Occasionally I have operated into the wee hours Saturday night. Some
people go to a high location like a mountain top and stay there all
weekend. Many rove, driving hundreds of miles, or some combination of
the two strategies. Very few have AC power available so operate on
batteries for the two days, though sometimes with the option of charging
batteries between stops or by just running your engine or a generator.
By contest rules, to count as a new contact, at least one of the two
stations needs to move 10 miles or more from a previous location --
hence the rover strategy.

To reduce your levels of uncertainty in making contacts, two things
matter, antenna pointing accuracy and frequency accuracy and stability.
Being exactly on frequency is nice, but being off a few hundred Hz at 10
GHz is usually in the radio passband and good enough. So in my
experience a good OCXO is fine. It is accurate enough, very stable for
many minutes of contact operation, has good phase noise and moderate
power consumption. I think most people in the contest are using OCXOs. I
never checked exact accuracy, but I think, even with a lot of driving
and temperature extremes my rig stays within a few tens of Hz at 10 GHz.
A few operators use rubidiums. To my thinking, the extra accuracy is not
really needed and the extra power consumption is not worth going that
way if you are running off batteries. Usually, the ones available tend
to have a bit worse phase noise than a good OCXO too, though I'm not
sure if enough worse to matter in real contacts.

I have thought about taking a rubidium along to power on occasionally
and calibrate the OCXO but never found my OCXO frequency to be an issue
so never bothered to take the rubidium. There are applications like
microwave EME where very weak signals are extracted by post-processing
the data of a long contact in the noise level. In that case rubidium
accuracy is needed for very narrow bandwidth contacts.

You mentioned operating while driving. A few people have the omni
antennas to do that and I have worked some of them. For that the
frequency accuracy becomes moot. At freeway speeds the doppler shift at
10 GHz is very significant in the audio range. Because of that, the
mobile-while-moving contacts are usually made in FM mode with wide
bandwidths and no need for very accurate frequency. That mode can't do
the long distances of dishes and narrow SSB or CW but it has worked
better than I would have expected. Also, the 10-mile rule tends to make
the FM mobile less useful and it usually happens as an experiment while
someone is driving home at the end.

One side note about doppler. Often several guys roam in small packs. To
begin a contact, often one station will put up a steady carrier for the
other end stations to find. Often the rovers are set up near a freeway
and a station receiving near the guy sending steady carrier will hear
whoops in the steady tone caused by doppler bouncing of the signal off
the freeway traffic.

In my view, using GPS locked oscillators has the same disadvantage of
power consumption as the rubidiums. If you are in one location (a
mountain top, etc.) for long periods of time, it might work, especially
if you have AC available, but for roving, with the many locations, I
would think it would either not give you much accuracy or would cause a
big operational time penalty for multiple surveys. I'm not aware of any
roving operators around here using GPS (except for location, which
almost everyone now uses to determine their operating location).

A few operators, get by with an only moderately stabilized frequency.
This might be a "brick" oscillator with its so-so internal oven. It's
better to get out than to stay home thinking about better options. The
poor frequency control is usually on new operator's rigs and was more
common 10 years back. If the frequency is fairly stable, then the
drifting can be reined in by going out with other hams who have good
frequency to calibrate from, or if you have a beacon in range that you
can find to establish your offset.

So, yes, OCXO is the way most hams lock their mobile microwave rigs.

-Rex, KK6MK

On 3/10/2013 7:23 AM, Jim Lux wrote:

Hi

If you go with one of the better DOCXO's on eBay (spend the full $30 not $15) you should get something that will hold < 0.3  ppb for 48 hours. You would have to do a few things:

1. Keep it on power for a couple weeks ahead of time.
2. Keep it on power the whole weekend.
3. Make sure it's always in the same orientation (base down or whatever)
4. Put it in something like a cooler to keep the drafts off of it
5. Regulate the supply and efc tightly.

You might have to buy three and sort them, but I suspect not.

One ppb at 10 GHz would be 10 Hz. The carefully minded DOCXO should keep you within 3 Hz.

Bob

On Mar 10, 2013, at 6:24 PM, Rex rexa@sonic.net wrote:

I haven't researched this yet, but the SDR crowd is using GSM and LTE
for tuning their sample clocks.

I have no idea what kind of accuracy you can get from a tower.

Hi

With GSM you may have 10 ppb, you may get a lot better. With LTE you should get better, but you don't always. That was the downfall of the Symmetricom 2700. You are very much at the mercy of the cellphone outfit. They may or may not be paying attention.

Bob

On Mar 10, 2013, at 8:37 PM, gary lists@lazygranch.com wrote:

I do the stuff on your list that is easy. Controlling the environment --
orientation, ambient  temp, etc. -- is not worth the extra effort. Much
of what I discussed is that perfection isn't necessary.

The quality of the oscillator probably matters a good bit, but recently
I am learning, don't trust word of mouth or "specifications". Over the
last few years I have accumulated many "decent" OCXOs. But  I don't have
a base standard 10 MHz reference that I consider pristine
phase-noise-wise and until recently had no way that I trusted to make
any kind of phase noise measurements. Recently I got into a project
where I built a board that uses a LMX2541 chip to make 3600 MHz using a
10 MHz reference with a pretty wide loop bandwidth. The board multiplies
the 10 MHz reference up to a point where a good SA can see the phase
noise. Measuring that on my 8566 SA with John Miles software, I learned
a couple things.

First - My measurement setup doesn't give real accurate phase noise
measurements compared to passing my DUT and sources on to someone with
real quality instrumentation.

Second - Comparing my measurements to the very good equipment, it is
clear that my measurements give a close approximation to the good one's,
only not exact across 10 Hz to 10 KHz. But my measurements are good for
a qualitative feel within, say 5 dB, and certainly good for relative
comparison of the contribution from different 10 MHz reference sources.

So, I have bought a lot of 10 MHz OCXOs from eBay over the last few
years. The best phase noise baseline reference I have found so far is my
Z3805. I have lots of OCXOs in the 2x2x1.5 inch size. Many had good
specs pointed at by the listings or word of mouth. When I used them with
my board and SA most were pretty crappy compared to the Z3805. A couple
of the ones I bought were Morion MV89As. - supposedly good, but what I
saw didn't look very great. One of the best ones I have is a small
2x2x.75 inch Wenzel I bought a few years back. It has a custom part
number of 500-11935. But don't buy by name. I recently picked up a 1x1x3
Wenzel 10 MHz with sma output connector and its phase noise looks pretty
horrible.

Some old Isotemps were decent, but not as good as the Z3805 and I
haven't measured some 10811s and 10554s I have in the back of my box
because they are harder to feed DC-wise.

My point is, I collected a lot of OCXOs that are not nearly as good as I
thought they would be. But all would probably make reasonable references
for frequency stability. Not sure if the level of not-great phase noise
from many of them would be noticeable if they were used to lock a good
10 GHz radio. If I was younger, I'd probably go for that experiment. --
In reality, don't know if I ever will.

-Rex

On 3/10/2013 5:15 PM, Bob Camp wrote:

On 10/03/13 16:17, Bob Camp wrote:

If you keep the OCXO on the Rb heated, you should get quicker on the
mark. Then it is mostly the heatup of the physical package which
consumes power and time then.

Then again, pre-heating with the car cuts down warm-up times.

Cheers,
Magnus

Hi

I would not depend on the 10 MHz reference for my phase noise floor at microwaves. I'd do as Bert suggested and lock up a quiet 100 MHz oscillator to the 10 MHz. The "best of the best" 10 MHz OCXO isn't going to be able to beat a 100 MHz over a 100 Hz to >=10 KHz bandwidth. The 100 MHz likely will beat the 10 MHz by 10 to 20db.

Bob

On Mar 11, 2013, at 4:25 AM, Rex rexa@sonic.net wrote: