EEVblog #456 - CSIRO Rubidium Frequency Standard

[EEVblog]

Dave uses his CSIRO National Measurement Institute rubidium frequency standard to calibrate and adjust his Agilent 53131A frequency counter.
With bonus teardowns.

[mswhin63]

I love those racks, reminds me of the old railway days working on the radio and signalling systems 

[olsenn]

0.6ppm is nothing to sneeze at for a standard OTC crystal oscillator. Aging will be greater than that for a single year, so don't get too fond of that accuracy... that's what the Rb Std. is for. Speaking of which, do you still have your FE-5680A? How about adding one Rb. Std to the 10MHz in of your new counter and seeing that the other is a perfect 10MHz as well?

[EEVblog]

0.6ppm is nothing to sneeze at for a standard OTC crystal oscillator. Aging will be greater than that for a single year, so don't get too fond of that accuracy... that's what the Rb Std. is for.

Yes, that's the idea of course.
The stock oscillators are crap, it'll drift more than the SS Minnow

Speaking of which, do you still have your FE-5680A? How about adding one Rb. Std to the 10MHz in of your new counter and seeing that the other is a perfect 10MHz as well?

You won't see any difference with 10 digits

[free_electron]

The reason for the delta supply is because of the oven and the other options.
This supply is always running and the switch on the front only powers the counter section. If the oven is installed it is kept hot.

The little board with the transformer is an AC power supply for the filament of the VFD. the filament needs AC voltage to prevent VFD burning. as the delta switcher only delivers DC they need this little inverter.

now, since you got the rubidium standard just feed the 10MHz from the Rubidium standard into the back of the counter 'EXT REF IN' bnc and tell the counter to use the external reference. now your counter will always be spot on.

hoop the rubidium standard up to an UPS and you are golden !

[lowimpedance]

If you want to see what the NMI actually use for there time standards check out the link below.

http://www.measurement.gov.au/Services/calibrationtesting/Pages/TimeandFrequencyFacilities.aspx

[tylerl]

According to the manual for the 53131A, you can connect your rubidium frequency standard up to the back of the frequency counter (search the PDF for the words "external timebase"), and suddenly the accuracy of your Agilent frequency counter should improve pretty noticeably.

So then why try to calibrate the 53131A using the trim pot?

[EEVblog]

According to the manual for the 53131A, you can connect your rubidium frequency standard up to the back of the frequency counter (search the PDF for the words "external timebase"), and suddenly the accuracy of your Agilent frequency counter should improve pretty noticeably.
So then why try to calibrate the 53131A using the trim pot?

Why? Because I could.

[chickenHeadKnob]

Hello All

When dave did the tour of the HP cal lab and looked at the cesium clock it was stated that the cesium source eventually wears out or is consumed as the atoms are boiled off. Do rubidium clocks suffer the same type of life-span limit? If no, how do they die? Those FE-5680 units can look thoroughly knackered from external appearance.
 

[ddavidebor]

0.6ppm is nothing to sneeze at for a standard OTC crystal oscillator. Aging will be greater than that for a single year, so don't get too fond of that accuracy... that's what the Rb Std. is for.

Yes, that's the idea of course.
The stock oscillators are crap, it'll drift more than the SS Minnow

Speaking of which, do you still have your FE-5680A? How about adding one Rb. Std to the 10MHz in of your new counter and seeing that the other is a perfect 10MHz as well?

You won't see any difference with 10 digits

You need to use a scope and measure with the watch the time the two signal need for drift of X degree

And make some calculation, don't know what calculation

[Bored@Work]

Hello All

When dave did the tour of the HP cal lab and looked at the cesium clock it was stated that the cesium source eventually wears out or is consumed as the atoms are boiled off. Do rubidium clocks suffer the same type of life-span limit? If no, how do they die? Those FE-5680 units can look thoroughly knackered from external appearance.

Yes, the rubidium tube or lamp wears out. Manufacturers specify a lifetime. Something like ten years is typical. It does not mean they immediately die after ten years, but the likehood of them dying starts to get significantly higher. One reason you see so many used rubidium oscillators on eBay and other sites it exactly this. Gear gets decommissioned because the rubidium oscillator is near its expected lifetime. (another reason is rubidium oscillators are used in mobile phone network gear in larger amounts, and phone companies constantly update and upgrade their mobile phone networks, throwing out older gear).

[Huluvu]

The reason for the delta supply is because of the oven and the other options.
This supply is always running and the switch on the front only powers the counter section. If the oven is installed it is kept hot.

@Dave   can you please look at the Delta PSU board for some PN (maybe I can check in SAP ......)   

[EEVblog]

When dave did the tour of the HP cal lab and looked at the cesium clock it was stated that the cesium source eventually wears out or is consumed as the atoms are boiled off. Do rubidium clocks suffer the same type of life-span limit? If no, how do they die? Those FE-5680 units can look thoroughly knackered from external appearance.

Yes, the Stanford unit specifies a guaranteed 20 year life.

[Ed.Kloonk]

0.6ppm is nothing to sneeze at for a standard OTC crystal oscillator. Aging will be greater than that for a single year, so don't get too fond of that accuracy... that's what the Rb Std. is for.

Yes, that's the idea of course.
The stock oscillators are crap, it'll drift more than the SS Minnow

[chuckle]   

[grumpydoc]

The stock oscillators are crap, it'll drift more than the SS Minnow

They're also quite difficult to adjust with any precision as Dave discovered - especially if there is any slack in the trimmer. Also, sometimes, any change can perturb the oscillator by orders of magnitude more than the adjustment you're trying to achieve. 

I'm surprised that the oscillator isn't voltage controlled with a DAC to generate the V_FC - that is not only a better way of doing it but lends itself to automatic calibration rather more easily than sticking a screwdriver into a trimmer. Especially as the tongue angle calibration on most units tends to be very poor

I suspect that these were intended to be used with the Rubidium standard rather than the inbuilt oscillator.

[firewalker]

Would it worth the trouble to replace that trim cap with something more "trimable"?

Alexander.

[c4757p]

I suspect that these were intended to be used with the Rubidium standard rather than the inbuilt oscillator.

Yeah, I don't have too much experience with "professional" equipment like this, but if has an external reference input I wouldn't be expecting much from the internal reference.

Of course, my own experience proves my suspicion wrong (my HP function generator with external ref input is 1ppm out from a Rb reference, and it's not even ovenized), so take that with a grain of salt... 

[grumpydoc]

Would it worth the trouble to replace that trim cap with something more "trimable"?

Probably not - the underlying drift (both temperaature and aging) will be such that you're only ever going to get 6 digits of accuracy from it.

There are some micro ovenised units available from the Far East via ebay. I've had a couple of these to replace failed oscilators in couters and sig gens and they're pretty OK and would be a *much* better replacement if you wanted to improve on the standard crystal oscillator - but, then, Dave has multiple Rubidium units to use now

[EEVblog]

Yes the stock oscillator is just a toy in such a good counter. They expect you to at least spring for the oven osc.

[Ed.Kloonk]

Could you feed one of the rb outputs to a clock input on the counter?

[EEVblog]

Could you feed one of the rb outputs to a clock input on the counter?

Of course, that's why all counters have external clock inputs.

[Ed.Kloonk]

Could you feed one of the rb outputs to a clock input on the counter?

Of course, that's why all counters have external clock inputs.

 


Boozy tonite mate sorry.

[bingo600]

Daummmm .... Dave
This is "cheating" & Not fair 

Did you get both the RB & Freq Counter for $260 

A used PRS10 is 750+ US$ on the 'Bay ... alone

Send it to me for "Looooooonnnngggg time" calibration testing ... 10 years or so ...
I have the GPS antenna for it 


/Bingo

[EEVblog]

Did you get both the RB & Freq Counter for $260 

Yep, both, plus the industrial PC, plus the APC PSU and batt.

A used PRS10 is 750+ US$ on the 'Bay ... alone

Yes, they are not cheap. I could sell both for a big profit, but the PRS10 is just so darn nice... 
I had an FS725 once, but sold that.

[kyndal]

if i had a GPS disciplined, rubidium frequency standard on hand
i would do the only right thing for a hacker to do...

take my egg timer apart. and replace the 32.768 khz crystal with a divide by 305.17578125... circuit
and feed it the 10mhz reference...

finally go cook some eggs up with a only a few ppm's worth of precision and consistency

challenge posted!.   


haha..


/Kyndal