Author Topic: Reference oscillator for home lab question  (Read 3061 times)

0 Members and 1 Guest are viewing this topic.

Online 4thDoctorWhoFanTopic starter

  • Regular Contributor
  • *
  • Posts: 218
  • Country: us
Reference oscillator for home lab question
« on: December 31, 2017, 05:20:32 pm »
I would like your opinion on how many instruments can be connected to a 10Mhz reference oscillator before a distribution amplifier is needed.  Would the answer be different if you are using a rubidium osc versus a OCXO?
 

Offline DaJMasta

  • Super Contributor
  • ***
  • Posts: 2352
  • Country: us
    • medpants.com
Re: Reference oscillator for home lab question
« Reply #1 on: December 31, 2017, 05:56:41 pm »
One  :)


Especially if the instruments are high impedance loads, you could connect multiple to the same source without risking stressing the output amplifier for the reference, but you're also going to be distorting the waveform of the signal considerably - they're typically designed to use 50 ohm terminations, so if you use two (or more) in parallel the output amp has to drive twice the load (which is a lot of current for a little IC).


Luckily, distribution amplifiers are relatively simple, even if there aren't a huge number of 50 ohm ones designed for the application available.  You can DIY one of reasonable quality, there are PCB layouts available for free as well (at least one I know of cause I put it up there), and you can generally use 75 ohm distribution amps to good effect (used for audio/video distribution) which are cheap and widely available.
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3893
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Reference oscillator for home lab question
« Reply #2 on: December 31, 2017, 06:00:48 pm »
I just do not see the problem of connecting multiple high impedance input devices to a low impedance (50 ohm) source, if you terminate the end of the coaxial line properly. Use  BNC T directly on every high impedance input of the instrument.
 
The following users thanked this post: Johnny B Good

Online nctnico

  • Super Contributor
  • ***
  • Posts: 27811
  • Country: nl
    • NCT Developments
Re: Reference oscillator for home lab question
« Reply #3 on: December 31, 2017, 06:05:05 pm »
Also many instruments have a loop-through for the clock reference signal.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: Reference oscillator for home lab question
« Reply #4 on: December 31, 2017, 06:09:56 pm »
The correct answer is-it depends. The type of oscillator doesn't matter but the loading and termination on the output matters a great deal. I have a video distribution amp (DA) that I modified for 50 ohm rather than the video standard 75 ohm input and outputs. What determines whether the DA is 50 or 75 ohm is the resistor connected pin to ground on all the connectors. Rather than replace all the resistors it is easier to just solder a 150 ohm resistor in parallel with the 75 to get the 50 ohm needed. Video DAs are pretty common on Ebay at reasonable prices.

If you are driving multiple pieces of test equipment from one output (daisy chain) there are a few things to watch. First, check what the 10 Mhz voltage required for the instrument is. If you try to connect too many 50 ohm instruments to one line the voltage will drop as you add each one and you may go below the voltage required for the instrument to see the input. Second, make sure that the end of the line is terminated by 50 ohms to avoid reflections in the line that could cause false counting. Some instruments are internally terminated so you won't have to add an external 50 ohm termination if that is the last instrument in the line. Some instruments may be high impedance. 

Make sure that the waveform is correct for the instrument you are driving. Most use a sine wave but check to make sure. The video DA I modified also had a sync channel that would put out a 0-5V square wave if I need that. Another problem I ran into was that some instruments were poorly designed so that there was some crud being outputted from their ext ref in that would mess with other instruments on that line.  Some video DAs have one opamp driving several outputs thru resistors while others use a separate opamp for each output and give much higher isolation to interference and crud. It isn't obvious by looking at the listing on Ebay whether you are getting a really good one or a cheapy. 

I have one Rb standard that has a high quality 6 channel DA built in and I added a 1Ghz PLL 'brick' so I also had 1 Ghz for checking counters as well.   
« Last Edit: December 31, 2017, 06:15:06 pm by ArthurDent »
 

Online 4thDoctorWhoFanTopic starter

  • Regular Contributor
  • *
  • Posts: 218
  • Country: us
Re: Reference oscillator for home lab question
« Reply #5 on: December 31, 2017, 08:20:23 pm »
Thanks for the opinions.
Looks like a distro amp would be the proper way to go.  I'll have to do some research on a good video distro amp to modify.
 

Online 4thDoctorWhoFanTopic starter

  • Regular Contributor
  • *
  • Posts: 218
  • Country: us
Re: Reference oscillator for home lab question
« Reply #6 on: December 31, 2017, 09:13:12 pm »
I want to distribute the 10Mhz reference to about 5 or 6 different pieces of equipment.

I have a Lucent rubidium standard but since the rubidium does not last forever and it's expensive, I am NOT going to leave it on and connected to all the equipment 24/7. The plan is to use the rubidium standard to calibrate a precision OCXO and use the OCXO to distribute the 10Mhz and leave on 24/7.
 

Offline PTR_1275

  • Frequent Contributor
  • **
  • Posts: 561
  • Country: au
Re: Reference oscillator for home lab question
« Reply #7 on: December 31, 2017, 11:57:07 pm »
I bought a Racal Dana 9478 a few years ago for a really good price. It has its own internal reference oscillator (mine has the better spec option) and 9 outputs. There is a external reference inside as well as a frequency monitor on the front.

Since moving premises, I’ve not set up the frequency part of my rack yet, but there will be a HP GPSDO feeding into the racal dana, then feeding out from there. If the gpsdo drops signal, then the internal oscillator of the racal dana takes over. Complete overkill for my needs, but from memory I paid about $350 total for both the gpsdo and the racal dana.

In terms of cheap video distribution amps, a lot of people use Extron. Professional video stuff from an American company. Gerry Sweeney has a video where’s he added a oscillator into the extrons spare space in the case.
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: Reference oscillator for home lab question
« Reply #8 on: January 01, 2018, 02:22:37 am »
If you want a really good quality filtered output 4 port 10 Mhz DA and you're willing to assemble a p.c. board kit about the size of a credit card with surface mount parts, Down East Microwave, Inc. has a $25 kit that works really well. They also sell the unit assembled in a case for $75. If you check out their site there is a PDF that has all the information and the schematic. I've built a couple of these into units including one into a Trimble Thunderbolt GPSDO that I have and they are nice.

http://www.downeastmicrowave.com/product-p/10-4k.htm
 

Offline g0mgx

  • Regular Contributor
  • *
  • Posts: 79
  • Country: gb
Re: Reference oscillator for home lab question
« Reply #9 on: January 01, 2018, 06:15:43 am »
The owner of this site sadly passed in 2017, but the schematics are still on the site.

http://huprf.com/huprf/distribution-amplifiers-and-multi-couplers/8-channel-distribution-amplifiers/

I have one of these and it works very well.

Mark
G0MGX
 

Offline DaJMasta

  • Super Contributor
  • ***
  • Posts: 2352
  • Country: us
    • medpants.com
Re: Reference oscillator for home lab question
« Reply #10 on: January 01, 2018, 07:49:37 am »
Sure rubidium doesn't last forever, but the usual rated uptime is 10-20 years, so it's not like it will age-out quickly.  If your Lucent standard is a GPSDO, then that strategy will do, but if it's just a Rubidium oscillator, you don't really gain anything setting the OCXO to it and then relying on the OCXO, since the Rb's advantage is all in drift performance.

To toss my own design into the mix, here's the basic one I put together: https://circuitmaker.com/Projects/Details/DaJMasta/Seven-Output-10MHz-Distribution-Amplifier

Have that circuit on a different board built into my rubidium reference and it's performance has been good for the few months it's been running 24/7 now, and the real benefit is that the cost of the connectors is, by a significant margin, the most expensive part of the parts list.
 

Online 4thDoctorWhoFanTopic starter

  • Regular Contributor
  • *
  • Posts: 218
  • Country: us
Re: Reference oscillator for home lab question
« Reply #11 on: January 01, 2018, 02:53:30 pm »
Sure rubidium doesn't last forever, but the usual rated uptime is 10-20 years, so it's not like it will age-out quickly.

I understand what you are saying but most of the affordable used rubidium standards available are already about as old as you stated.  That's why they are on the used market.  Anyway, the way I look at, if the OCXO is good enough for multiple thousand dollar test equipment, it's good enough for my home bench.  I'll just periodically "cal" it to the rubidium standard.  Having said that, I will look into the GPSDO's.
 

Offline Chris56000

  • Frequent Contributor
  • **
  • Posts: 909
  • Country: gb
Re: Reference oscillator for home lab question
« Reply #12 on: January 01, 2018, 04:00:00 pm »
Hi!

Quote
Sure rubidium doesn't last forever, but the usual rated uptime is 10-20 years, so it's not like it will age-out quickly.

I can't remember where I found it, but there is a site that explains how to extend the life of a Rubidium Source by warming the lamp with hot-air and spinning it at low-speed at the same time to recover the rubidium that vaporises on the inside walls of the lamp and helps it settle by the lamp electrodes, which gives the lamp a life-extension!

I do remember the writer of the article said he cured an Eratom unit that completely failed to produce a lock by this means!

Can anyone give me a link to the repair posts on Rubidium please? I'm thinking of getting one! Any recommendations?

(More T.E.A. Symptoms!)

Chris Williams
It's an enigma that's what it is!! This thing's not fixed because it doesn't want to be fixed!!
 

Offline Tony_G

  • Frequent Contributor
  • **
  • Posts: 956
  • Country: us
  • Checkout my old test gear channel (link in sig)
    • TGSoapbox
Re: Reference oscillator for home lab question
« Reply #13 on: January 01, 2018, 05:02:56 pm »
I want to distribute the 10Mhz reference to about 5 or 6 different pieces of equipment.

I bought a used Symmetricom 6502B fairly cheaply for my rack. I'm also currently using one of the BG7TBL units.

Sometime in the future I'd like to move to a proper rack mounter HP/Symmetricom one.

TonyG

Offline jpb

  • Super Contributor
  • ***
  • Posts: 1771
  • Country: gb
Re: Reference oscillator for home lab question
« Reply #14 on: January 01, 2018, 10:28:35 pm »
Hi!

Quote
Sure rubidium doesn't last forever, but the usual rated uptime is 10-20 years, so it's not like it will age-out quickly.

I can't remember where I found it, but there is a site that explains how to extend the life of a Rubidium Source by warming the lamp with hot-air and spinning it at low-speed at the same time to recover the rubidium that vaporises on the inside walls of the lamp and helps it settle by the lamp electrodes, which gives the lamp a life-extension!

I do remember the writer of the article said he cured an Eratom unit that completely failed to produce a lock by this means!

Can anyone give me a link to the repair posts on Rubidium please? I'm thinking of getting one! Any recommendations?

(More T.E.A. Symptoms!)

Chris Williams
This, I think, is the document on rejuvinating rubidiums:
http://www.vk3um.com/Rubidium%20rejuvination.pdf

here is a repair guide for the LPRO
https://xdevs.com/doc/EFRATOM/Efratom_LPRO_101_Repair_Guide.pdf
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf