Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Rubidium Standars-do they crap out eventually?  (Read 2145 times)
W8AAZ
Member

Posts: 358




Ignore
« on: August 31, 2010, 02:39:41 PM »

Something for you ham engineers to answer for me.  I bought a standalone Rubidium based ref. osc. unit (not a module) pretty cheap and powered it up. Seems to be functioning fine.  The lock-on light appears after a prescribed warmup period per the manual. My counter says it is certainly 10 Megs out to it's resolution.   Now I know xtal oscillators have lifetimes beyond ours, but do Rubidiums eventually croak over?  I have no idea how many hours this has, but I would assume any lifespan depends on operational hours?  Generally are we talking hundreds or thousands of hours or what?  I found the mfr of the actual Rub. module inside it but they do not list a "tech questions or support" email address.   Thanks.  (Yep, the edit page will not let me fix the title!)
Logged
K3GM
Member

Posts: 1824




Ignore
« Reply #1 on: August 31, 2010, 04:42:32 PM »

The very few Rb oscillators I've dealt with in work had a conservative MTBF of >200,000 hours, but you could easily get 10, 15, even 20 years out of it.
Logged
W8AAZ
Member

Posts: 358




Ignore
« Reply #2 on: September 01, 2010, 05:33:12 PM »

Thanks. All I had to compare it to is a 5345 with a calibrated time base and it shows itself to be seemingly playing well for only 85$.  Looks hardly used so I hope that translates to low time on the circuitry.  Hope that 200k hours means that if I only use it occasionally, letting it reach its accuracy spec in a couple hours at a time that I will get all the use I need before it goes pfffft. Now to find someone with a GPS timebase to really try to confirm it better, as most hams cannot afford Cesium!  No cal labs around the corner, around here!   
Logged
K3GM
Member

Posts: 1824




Ignore
« Reply #3 on: September 02, 2010, 05:06:57 AM »

My friend uses a GPS timebase for all of his transverters. I checked with him, and he sent me this link:  http://www.rabel.org/archives/General_GPS_Info/A%20Simple%20GPS%20Stabilized%2010%20MHz%20Oscillator.pdf  He added that the design had an incompatible logic level flaw that was simple to fix.  It's pretty cool to watch on a waterfall as the frequency drifts about until the unit locks on the satellites and then snaps into a dead stable line.
Logged
AA4HA
Member

Posts: 1583




Ignore
« Reply #4 on: September 02, 2010, 08:40:07 PM »

Actually they do have a life expectancy that is shorter than you think. Many Rb oscillators have provisions for you to monitor the lamp voltage. As they age the lamp voltage can drift.

The light is an RF excited plasma discharge running at around 110 C with a high RF and magnetic field. There is an effect called the hyperfine transition state means that the Rb gas is oscillating at 6.834687 GHz. This is done by "optical pumping" and a darkening effect.

On some Rb oscillators you can periodically (every few years) trim up the voltage to the lamp to offset aging. There is a monitoring point and usually a pot that you can get to.

If you are using an Efratom Rb source I can send you the operations/ maintenance manual that explains the physics package, adjustments and outputs. Just shoot me an email or drop me a PM.

Tisha Hayes
Logged

Ms. Tisha Hayes, AA4HA
Lookout Mountain, Alabama
KA5N
Member

Posts: 4380




Ignore
« Reply #5 on: September 03, 2010, 05:04:53 AM »

Of course everything has a lifetime.  With a rubidium freq. standard there are electronic parts that can and do fail.  The lamps can fail to ignite after much use.  The gas cell can leak, etc. etc.
The problem is: if you suffer a failure other than a simple electronic fault how do you go about repairing it?   These devices were never considered to be field repairable.  If you don't fully understand how the device works and have a source of repair parts, you will be lost. 
Back in the 1960's I was an engineer at a company that built Rb. Stds.  In those days the devices were about ten times as large as those being sold today.  If I was to need to work on a modern unit I would need a few weeks to catch up to todays devices.
So if you have one that works just use and enjoy it.  When it craps outs, get another surplus unit and don't worry it will be cheaper on the pocketbook and your brain.
Allen
Logged
W8AAZ
Member

Posts: 358




Ignore
« Reply #6 on: September 03, 2010, 07:09:04 PM »

Thanks!( I thought I checked the block for this dang thing to send me emails when answers were posted)  Well I only got 85$ tied up in it. The listing did not say it was a Rb std. They subsequently claimed they did not know.   I thought it was OXCO or something from the description.  Guess that was a nice bonus. Till it dies, if ever.  Another seller selling one says it contains a Frequency Electronics FE-5650A as the Rb component.   FE has a phone number but no tech support email, just administrative stuff like HR. I guess another downside is the garbage content of a Rb std. is higher than a good OXCO setup but I am not sure how that will affect anything practically for me.  Actually ebay sellers are selling the Rb module for alot more than the whole unit cost me.  And they are used also, apparently.  Guess I will use it sparingly. It has no on off switch, it is on whenever it is plugged in.  So I guess I can plug it in sometimes and use it occasionally to check the TXCO in my little HP counter and such after sufficient stabilisation.   
Logged
AA4HA
Member

Posts: 1583




Ignore
« Reply #7 on: September 03, 2010, 08:16:34 PM »

To get the higher accuracy levels of any Rb frequency standard they should never be turned off. Many of the designs even suggest running the unit off of a 24 volt battery to keep them at a very stable temperature.

Just firing one up you may be 1x10 -5 stability but over the hours and days that will increase up to 1x10 -8 or -9. You could get 1x10-6 just by zero beating against WWV.

I use a Rb standard as a constant reference for my HP oscilloscope and HP frequency counter as they have aux reference inputs.

Generally it is a PITA (pain in the "amplifier") to keep a Rb reference running 24/7. To do it right there are a dozen different factors that can all contribute to inaccuracy but it is cheaper than hauling gear out to a metrology lab for a frequency calibration sticker.

Tisha Hayes
Logged

Ms. Tisha Hayes, AA4HA
Lookout Mountain, Alabama
W8AAZ
Member

Posts: 358




Ignore
« Reply #8 on: September 04, 2010, 04:03:35 PM »

My HP counter and Fluke synth. RF gen. accept an external 10 Megs, but a scope?  I have a multirange scope calibrater that has a TXCO in it, but other than aligning the scope sweep with that device, I cannot imagine needing a scope sweep to be super ultra accurate?
Logged
AA4HA
Member

Posts: 1583




Ignore
« Reply #9 on: September 04, 2010, 06:20:50 PM »

It's a giant lab grade HP four channel scope and has a 10 meg reference input so I use it.

It also helps offset one of the BITE error codes that pop up regarding the timebase oscillator stability on the scope.

It has been a long time since I needed that sort of resolution on a digital scope. Nowadays I will use something with stability if making jitter measurements on a microwave link and when I do that, I just run the link into a TTC/Acterna T-BERT.

Tisha Hayes
Logged

Ms. Tisha Hayes, AA4HA
Lookout Mountain, Alabama
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!