Efratom FRK-L Rb Frequency Standard

[notfaded1]

I picked up an Efratom FRK-L Rb recently.  I was planning on building it into a case with linear PS and some nice outputs plus maybe a lock signal LED.  These units are neat because they don't have a bunch of complicated electronics in them.  If I'm reading it correctly some version of these made their way into space on some early satellites.  It locked up pretty quickly less than 5 minutes.  It went through the usual up and down sweep then locked on 10MHz and stuck.

Bill

[Mr. Scram]

Very cool and definitely a win. Thanks for sharing!

[notfaded1]

Here's the frequency 30 minutes with 53131 ext reference BG7TBL.

Bill

[notfaded1]

I bid on two SR620's Friday but missed on both of them... they were both not powering up but one looked really nice.  I guess it's back to other ways to get more than 10 digits now...  I think I'm going to put the Cs 4040A back on the bench and work on it some more now that Corby posted the manual for the FTS4060!!!  Thanks again brother!  That's going to be HUGE help for me because I have some real info that very closely matches my unit.  I also just noticed the page you posted on my other thread about the monitor cable pin outs.  I'm tempted to bid on the 5061B.  It looks REALLY nice!

I want one of those Thuderbolt monitor kits when they're available too Corby!

Bill

[notfaded1]

Here's the BG7TBL fed ext ref with my Thunderbolt measuring the Efratom FRK-L.

Bill

[notfaded1]

Since it's stopped raining in the desert here in AZ finally I may work on moving my GPS antenna up on the roof finally... that would give me some much better view of the sky finally!  I can't wait till the desert valley heat comes back to help kill off all the COV-19 virus that might have made it to AZ.

[ArthurDent]

I guess it's back to other ways to get more than 10 digits now... 
Bill

Check your counter manual about the math function. Here's a photo of a Fluke/Philips PM6680B 10 digit counter using the math function to subtract 10 Mhz from the 10 Mhz input signal and displaying the difference for a longer time period and displaying just the difference of .00003 Hz. There are a few high resolution counters out there that might work this way that I might have posted photos of before. My CNT-81/PM6681 has a timebase multiplier and can display the same information as the PM6680 in 1/10 the time. All the counters shown are using an external GPSDO for the time base.

On the FRK, The 'L' suffix is the lower stability/aging version. I have an FRK-H which is 3 times better stability wise and also has the 'LN' suffix (FRK-HLN) which is low noise. Mine has the Winchester power connector instead of the individual pins. It locks in 4 minutes 15 seconds and anything in the 5 minute range I would consider quite good so your Rb should have a lot of life left. Notice mine was made in 1993 and still works great. The date code on yours is 1991.

[notfaded1]

What would be neat would be figuring out the best way to compare my 3 Rb standards and the two GPSDO's.  I figure the thunderbolt once it's optimized will be better than the BG7TBL but it would be neat to find out more information about it for sure.

[notfaded1]

I did some better measurements using the SRS Rb as my reference for a Pendulum CNT-90 running in 83132A mode so it reads more easily with timelab.  I plugged the SRS into a medical isoMATE to help hopefully with noise.  It's still a switched mode power supplying the 24V and I'm planning on switching it over to a linear power supply in the near future for all of these measurements.  The Pendulum counter is pretty neat.  I now have an 53131, 53132 and the CNT-90.  My next plan is to do phase noise measurements with the TURN Dynamic Systems FSA3011.  It's an interesting device and the updated version hopefully works pretty well.  I'll know soon.  It looks like my Efratom is a little off high and might need a little adjusting.

Bill

[notfaded1]

some more pics of the setup...

[notfaded1]

pics

[testpoint1]

some more pics of the setup...

2 problems:
a. you may set the CNT-90 meastime to 1 Second then get 13 digits;
b. the receiving level why so low? actually a normal output at least 2V (Vpp).