Hi

Your budget is going to make things tough.

A Thunderbolt needs a triple supply (+12, -12, and +5) to operate. It also needs a GPS antenna of some sort.

The Rubidiums all need a supply and a heat sink. Some of them like the FRS-A need unusual connectors.

They all need to be operated continuously to get the best accuracy. That makes powering them with a shared lab supply impractical.

The Thunderbolt is accurate as a stand alone device because it uses GPS. The rubidiums are all fairly accurate by design, but they would need to be calibrated against something else to be as accurate as the Thunderbolt.

Simple look at it:

If you need ~ 3x10-9 accuracy, then the rubidiums are all fine.

If you need < 2 x10-10 accuracy, then the Thunderbolt is your only choice.

In between those two limits, it depends on a lot of details like how long between calibrations, environment (like temperature), and sources of calibration. Even at the limits I've mentioned, what you are calibrating might impact the choice.

If you have a spare triple output supply already, Then the Thunderbolt is inside your budget. If you do not have a supply, you probably can pick up the pieces on individual auctions with some luck and patience. If you go with a rubidium, look for a dealer who will include the connectors for the device.

One last point: The Thunderbolt probably will run for a very long time. The rubidiums may also run for a long time, or they may die 5 or 10 years from now due to lamp wear out.

Lots to think about.

Bob

On Dec 27, 2009, at 4:12 PM, Giuseppe Marullo wrote:

And started down the dark path you have. You will not have to wait
long before you completely succumb to the dark side.

Jim Palfreyman

2009/12/28 Bob Camp lists@cq.nu:

Thanks a lot to all for your quick answer.

Rubinium should be good for my needs, but buying it surplus makes me
think I could get something very used (and abused) and it does not have
the self correcting thing thunderbolt has.
GPSDO gives me also the time, maybe with a supercool LCD display.

Do you know exactly the power requirement? On Ebay I read 15W then few mA on each branch (board only). Something is not clear to me.

The dark side of the noon already embraced I have....

Giuseppe

PS: I am experiencing mail problems for the first time in many years,
please anyone willing to contact me directly do cc copy also this other
email address: giuseppe.marullo@iname.com while my ISP gathers back all
the bits they lost (2 days of emails vanished from my IMAP account under
my eyes, literally)

Hi Giuseppe,

I think you would be better served with the Thunderbolt as it is both a GPS and an oscillator, that is it is a GPSDO.

Last I remember, an eBay user "Flukel" (the last character is a lower case "L") was spoken of, on this list, as being a reliable Chinese
vendor.  He had oscillators, Rubidiums and Thunderbolts.

That said, a power supply is easy to construct from many other dead chasis parts and if needed the regulator chips are pretty cheap
(well here in the USA anyway).  If you do go that way and have trouble scrounging some parts for the power suppply, perhaps some of us
here can help with that.

Bill....WB6BNQ

Giuseppe Marullo wrote:

Hi

The simple answer on the supply is that there are a couple of answers.

Very little current is pulled from the -12 supply.  I can not imagine a supply that would not keep up with the -12 requirement.

The +12 supply mainly runs the OCXO in the unit. Since it's running an oven, the power will drop off after warmup. It will also be lower at higher temperatures. Depending on exactly which OCXO you get in your unit, the power may be a little higher or a little lower.

A 5V supply that can furnish 300 ma should be adequate.

All of the supplies needed could be pulled off of a typical desk top PC power supply. If you do go that way, be sure to put some simple RF filtering in the supply leads.

The unit it's self has a fairly basic case. Enclosing even in a simple pine box should improve it's performance. It would also protect the leads from getting damaged in use. The enclosure does not need to be elaborate. You should be able to make it for free from scrap materials.

I have bought Thunderbolts from several dealers on eBay over the last month. All of them seem to be supplying the same product (same date codes, and firmware). They all seem to be from 2003 or 2004. Some dealers ship faster than others, some include a HF power splitter, some include a power connector. I have not had any of the units on power long enough to know if some are actually better than others.

Good Luck!

Bob

On Dec 27, 2009, at 7:56 PM, Giuseppe Marullo wrote:

Hi Giuseppe,
Welcome to the group.
If you are already happy constructing equipment and have basic test gear ('scope, DMM) and a small budget, I'd suggest a timing GPS as a starting point. While a Thunderbolt will give most of what you want, it is more expensive. A Timing version or Motorola's Oncore <(ebay item 300355981024 from flukel for instance) will give you an accurate 1 pulse per second output (you also need a power supply antenna and a PC). This and a 'scope will allow you to calibrate other instruments and a oven controlled crystal oscillator. A GPS based solution (including Thunderbolt) will give you good confidence of accuracy. Just buying a surplus Rubidium could result in you adjusting all your instruments to the wrong frequency! Like many of these decisions, it's a balance between time and money, with a lesser input from your capabilities and existing equipment.
 
Robert G8RPI.

--- On Mon, 28/12/09, Giuseppe Marullo giuseppe@marullo.it wrote:

From: Giuseppe Marullo giuseppe@marullo.it
Subject: Re: [time-nuts] Timenoob - Cheap and simple 10MHz reference
To: "Discussion of precise time and frequency measurement" time-nuts@febo.com
Date: Monday, 28 December, 2009, 0:56

Thanks a lot to all for your quick answer.

Rubinium should be good for my needs, but buying it surplus makes me think I could get something very used (and abused) and it does not have the self correcting thing thunderbolt has.
GPSDO gives me also the time, maybe with a supercool LCD display.

Do you know exactly the power requirement? On Ebay I read 15W then few mA on each branch (board only). Something is not clear to me.

The dark side of the noon already embraced I have....

Giuseppe

PS: I am experiencing mail problems for the first time in many years, please anyone willing to contact me directly do cc copy also this other email address: giuseppe.marullo@iname.com while my ISP gathers back all the bits they lost (2 days of emails vanished from my IMAP account under my eyes, literally)

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.

Giuseppe Marullo wrote:

A timing GPS receiver with a built-in 10MHz oscillator would fit
your budget nicely. The Navsync CW12-TIM looks like the obvious
choice. You would need the CW12-TIM module, a 3.3V DC supply and
a suitable aerial/antenna. Power consumption is well below 1W.
It has a 1PPS output with 30ns RMS accuracy and a NCO output
with a default output frequency of 10MHz.
http://www.navsync.com/docs/CW12-TIM_DS.pdf

I paid just over Eur50 for mine last year when the Euro was very
strong against foreign currencies.

--
Linux 2.6.30

Hi Robert,
Will the Motorola that You have mentioned give a 10.00 MHz output standard.
Thank You
 Best regards,

Sal C. Cornacchia
Electronic RF Microwave Engineer (Ret.)

From: Robert Atkinson robert8rpi@yahoo.co.uk
To: Discussion of precise time and frequency measurement time-nuts@febo.com
Cc: giuseppe.marullo@iname.com
Sent: Mon, December 28, 2009 5:14:49 AM
Subject: Re: [time-nuts] Timenoob - Cheap and simple 10MHz reference

Hi Giuseppe,
Welcome to the group.
If you are already happy constructing equipment and have basic test gear ('scope, DMM) and a small budget, I'd suggest a timing GPS as a starting point. While a Thunderbolt will give most of what you want, it is more expensive. A Timing version or Motorola's Oncore <(ebay item 300355981024 from flukel for instance) will give you an accurate 1 pulse per second output (you also need a power supply antenna and a PC). This and a 'scope will allow you to calibrate other instruments and a oven controlled crystal oscillator. A GPS based solution (including Thunderbolt) will give you good confidence of accuracy. Just buying a surplus Rubidium could result in you adjusting all your instruments to the wrong frequency! Like many of these decisions, it's a balance between time and money, with a lesser input from your capabilities and existing equipment.
 
Robert G8RPI.

--- On Mon, 28/12/09, Giuseppe Marullo giuseppe@marullo.it wrote:

From: Giuseppe Marullo giuseppe@marullo.it
Subject: Re: [time-nuts] Timenoob - Cheap and simple 10MHz reference
To: "Discussion of precise time and frequency measurement" time-nuts@febo.com
Date: Monday, 28 December, 2009, 0:56

Thanks a lot to all for your quick answer.

Rubinium should be good for my needs, but buying it surplus makes me think I could get something very used (and abused) and it does not have the self correcting thing thunderbolt has.
GPSDO gives me also the time, maybe with a supercool LCD display.

Do you know exactly the power requirement? On Ebay I read 15W then few mA on each branch (board only). Something is not clear to me.

The dark side of the noon already embraced I have....

Giuseppe

PS: I am experiencing mail problems for the first time in many years, please anyone willing to contact me directly do cc copy also this other email address: giuseppe.marullo@iname.com while my ISP gathers back all the bits they lost (2 days of emails vanished from my IMAP account under my eyes, literally)

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.

Robert,

Thank you and all the others, you are giving me a lot of informations. BTW, my provider retrieved my emails so I am fully operational (at least I hope).

while I did not originally envision building this kind of equipment, I
could do it while it lies well within the digital domain, I have some
test gear (frequency counter, DMM, analog scope, BF frequency generator,
LA and some FPGAs with 100+Msamples ADC) but...I will have to trust it
so I am back at square one, until I have a reliable source... I would
like a GPSDO, possibly ready to run, then I will start building
something that I could actually check against the GPSDO.

Maybe a GPSDO myself, who knows...wondering if this could be done all
digital, without using a DAC, without using a driving voltage for a OCXO.

Giuseppe