Rb & GPS DO comparison
Locked
 
|
Guys,
This may be of some use to those interested in locking the K3 to an external source. The issue of what kind of source to used has been the subject of a number of postings. Clifton labs did the above comparison. Jack's Rb standard came in at about .0004 Hz relative to a GPS DO at 10MHz. I was wondering how accurate the EBAY purchased LPRO-101 Rubidium oscillator was. (I wanted to use it as a master oscillator for my test equipment and perhaps later the K3.) Comparisons to WWV indicated that in all liklihood it was well within 1 Hz. There was no way to determine if it drifted within that range. Curiosity finally got the best of me. I purchased an EBAY Trimble Thunderbolt GSP Disciplined oscillator. Supposedly, it is more accurate than the Rb secondary standard. Assuming it's truth, the Rb secondary standard was evaluated via the phase drift method using a dual channel scope. Out of the box, the Rb standard came in at 0.0075 Hz (10 MHz) of the GPS-D0. With a little tweaking of the ten turn pot in the LPRO-101, the two are well within 0.001 Hz (nominally 0.0008Hz) at 10MHz. There is a small drift component of 0.0002Hz after 3 hours after lock. Tweaking isn't straight forward. The appears to be a few minute latency period after a tweak. One can't simply stop the phase angle from moving by turning the pot. Doing so results in an over correction a few minutes later. It's difficult to tweak the Rb standard beyond this degree of accuracy. It takes more than 20 minutes to determine a data point. To do better would take some automation and a phase detector running for several days to get sufficient statistics. Next project is building a phase detector..... I'm pleased. The Rb standard has lots of advantages: Price, minutes to locking on, portability and simplicity. The next question: How accurate is the GPS-DO? Anybody got a hydrogen maser handy? Now what to do with the GPS-DO? Remember when decent receiver dials had only 5 KHz markings? 73 de Brian/K3KO ----- No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1390 / Virus Database: 1518/3776 - Release Date: 07/20/11 ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
Hi Brian,
Interesting post. I've read of others using GPS this way, but what does this kind of accuracy buy you when maybe 1 in 100,000 other hams go to this kind of trouble? 73, Ken Alexander VE3HLS --- On Wed, 7/20/11, Brian Alsop <[hidden email]> wrote: > From: Brian Alsop <[hidden email]> > Subject: [Elecraft] Rb & GPS DO comparison > To: [hidden email] > Date: Wednesday, July 20, 2011, 2:41 PM > Guys, > > This may be of some use to those interested in locking the > K3 to an > external source. The issue of what kind of source to > used has been the > subject of a number of postings. > > Clifton labs did the above comparison. Jack's Rb standard > came in at > about .0004 Hz relative to a GPS DO at 10MHz. > > I was wondering how accurate the EBAY purchased LPRO-101 > Rubidium > oscillator was. (I wanted to use it as a master > oscillator for my test > equipment and perhaps later the K3.) > > Comparisons to WWV indicated that in all liklihood it was > well within 1 > Hz. There was no way to determine if it drifted > within that range. > > Curiosity finally got the best of me. I purchased an > EBAY Trimble > Thunderbolt GSP Disciplined oscillator. Supposedly, > it is more accurate > than the Rb secondary standard. Assuming it's truth, the Rb > secondary > standard was evaluated via the phase drift method using a > dual channel > scope. > > Out of the box, the Rb standard came in at 0.0075 Hz (10 > MHz) of the > GPS-D0. With a little tweaking of the ten turn pot in the > LPRO-101, the > two are well within 0.001 Hz (nominally 0.0008Hz) at > 10MHz. There is a > small drift component of 0.0002Hz after 3 hours after lock. > Tweaking > isn't straight forward. The appears to be a few > minute latency period > after a tweak. One can't simply stop the phase angle > from moving by > turning the pot. Doing so results in an over > correction a few minutes > later. > > It's difficult to tweak the Rb standard beyond this degree > of accuracy. > It takes more than 20 minutes to determine a data > point. To do better > would take some automation and a phase detector running for > several days > to get sufficient statistics. Next project is > building a phase > detector..... > > I'm pleased. The Rb standard has lots of > advantages: Price, minutes to > locking on, portability and simplicity. > > The next question: How accurate is the > GPS-DO? Anybody got a hydrogen > maser handy? > > Now what to do with the GPS-DO? > > Remember when decent receiver dials had only 5 KHz > markings? > > 73 de Brian/K3KO > > > > > > > ----- > No virus found in this message. > Checked by AVG - www.avg.com > Version: 10.0.1390 / Virus Database: 1518/3776 - Release > Date: 07/20/11 > > ______________________________________________________________ > Elecraft mailing list > Home: http://mailman.qth.net/mailman/listinfo/elecraft > Help: http://mailman.qth.net/mmfaq.htm > Post: mailto:[hidden email] > > This list hosted by: http://www.qsl.net > Please help support this email list: http://www.qsl.net/donate.html > Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
Hi Brian,
If you want to go straight to the top of the frequency measuring heap, I have a GPS-disciplined Rubidium for sale ;-) It's a NIB Symmetricom StarLoc II Plus, that gives you GPS accuracy when it's locked to the sats, and Rubidium accuracy when it's in holdover. I've been a ham long enough to remember when I was happy to know that I was inside the band! 73, geo - n4ua On Wed, Jul 20, 2011 at 2:52 PM, Ken Alexander <[hidden email]>wrote: > Hi Brian, > > Interesting post. I've read of others using GPS this way, but what does > this kind of accuracy buy you when maybe 1 in 100,000 other hams go to this > kind of trouble? > > 73, > > Ken Alexander > VE3HLS > > > --- On Wed, 7/20/11, Brian Alsop <[hidden email]> wrote: > > > From: Brian Alsop <[hidden email]> > > Subject: [Elecraft] Rb & GPS DO comparison > > To: [hidden email] > > Date: Wednesday, July 20, 2011, 2:41 PM > > Guys, > > > > This may be of some use to those interested in locking the > > K3 to an > > external source. The issue of what kind of source to > > used has been the > > subject of a number of postings. > > > > Clifton labs did the above comparison. Jack's Rb standard > > came in at > > about .0004 Hz relative to a GPS DO at 10MHz. > > > > I was wondering how accurate the EBAY purchased LPRO-101 > > Rubidium > > oscillator was. (I wanted to use it as a master > > oscillator for my test > > equipment and perhaps later the K3.) > > > > Comparisons to WWV indicated that in all liklihood it was > > well within 1 > > Hz. There was no way to determine if it drifted > > within that range. > > > > Curiosity finally got the best of me. I purchased an > > EBAY Trimble > > Thunderbolt GSP Disciplined oscillator. Supposedly, > > it is more accurate > > than the Rb secondary standard. Assuming it's truth, the Rb > > secondary > > standard was evaluated via the phase drift method using a > > dual channel > > scope. > > > > Out of the box, the Rb standard came in at 0.0075 Hz (10 > > MHz) of the > > GPS-D0. With a little tweaking of the ten turn pot in the > > LPRO-101, the > > two are well within 0.001 Hz (nominally 0.0008Hz) at > > 10MHz. There is a > > small drift component of 0.0002Hz after 3 hours after lock. > > Tweaking > > isn't straight forward. The appears to be a few > > minute latency period > > after a tweak. One can't simply stop the phase angle > > from moving by > > turning the pot. Doing so results in an over > > correction a few minutes > > later. > > > > It's difficult to tweak the Rb standard beyond this degree > > of accuracy. > > It takes more than 20 minutes to determine a data > > point. To do better > > would take some automation and a phase detector running for > > several days > > to get sufficient statistics. Next project is > > building a phase > > detector..... > > > > I'm pleased. The Rb standard has lots of > > advantages: Price, minutes to > > locking on, portability and simplicity. > > > > The next question: How accurate is the > > GPS-DO? Anybody got a hydrogen > > maser handy? > > > > Now what to do with the GPS-DO? > > > > Remember when decent receiver dials had only 5 KHz > > markings? > > > > 73 de Brian/K3KO > > > > > > > > > > > > > > ----- > > No virus found in this message. > > Checked by AVG - www.avg.com > > Version: 10.0.1390 / Virus Database: 1518/3776 - Release > > Date: 07/20/11 > > > > ______________________________________________________________ > > Elecraft mailing list > > Home: http://mailman.qth.net/mailman/listinfo/elecraft > > Help: http://mailman.qth.net/mmfaq.htm > > Post: mailto:[hidden email] > > > > This list hosted by: http://www.qsl.net > > Please help support this email list: http://www.qsl.net/donate.html > > > ______________________________________________________________ > Elecraft mailing list > Home: http://mailman.qth.net/mailman/listinfo/elecraft > Help: http://mailman.qth.net/mmfaq.htm > Post: mailto:[hidden email] > > This list hosted by: http://www.qsl.net > Please help support this email list: http://www.qsl.net/donate.html > Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
In reply to this post by alsopb
"Anybody got a hydrogen
maser handy?" I do. Not at home of course, but at work. We actually manufacture them. If you really want to test your GPS DO, let me know. This is the company that made LPRO as well. Here are some numbers for you: LPRO accuracy spec when it left the dock - better than 1e-11 (relative frequency. Multiply by 10,000,000 to get frequency offset in Hz), aging drift - better than 1e-11 per month. Useful life of LPRO - usually 15 years. Make your math how much it drifts over life. GPS satellites have either very good Rb clock or average Cs clock. The accuracy spec of recovered GPS clock is better than 1e-12. The drift is not important because the clocks up there are adjusted once in a while to keep accurate. Hg maser is mostly praised for frequency stability, not accuracy per se, but its accuracy is also impressive - better than 1e-14 after 1000 seconds averaging. By the way, if you have way to use GPS Do with your K3, I would prefer that over LPRO. You did capture pros of LPRO. Here are the cons: high power consumption (need to heat large Rb lamp and resonant cell to over 100C°) and limited life. The surplus one you get from eBay is probably have most of it useful life gone. 73, Igor, N1YX -----Original Message----- From: [hidden email] [mailto:[hidden email]] On Behalf Of Brian Alsop Sent: Wednesday, July 20, 2011 2:41 PM To: [hidden email] Subject: [Elecraft] Rb & GPS DO comparison Guys, This may be of some use to those interested in locking the K3 to an external source. The issue of what kind of source to used has been the subject of a number of postings. Clifton labs did the above comparison. Jack's Rb standard came in at about .0004 Hz relative to a GPS DO at 10MHz. I was wondering how accurate the EBAY purchased LPRO-101 Rubidium oscillator was. (I wanted to use it as a master oscillator for my test equipment and perhaps later the K3.) Comparisons to WWV indicated that in all liklihood it was well within 1 Hz. There was no way to determine if it drifted within that range. Curiosity finally got the best of me. I purchased an EBAY Trimble Thunderbolt GSP Disciplined oscillator. Supposedly, it is more accurate than the Rb secondary standard. Assuming it's truth, the Rb secondary standard was evaluated via the phase drift method using a dual channel scope. Out of the box, the Rb standard came in at 0.0075 Hz (10 MHz) of the GPS-D0. With a little tweaking of the ten turn pot in the LPRO-101, the two are well within 0.001 Hz (nominally 0.0008Hz) at 10MHz. There is a small drift component of 0.0002Hz after 3 hours after lock. Tweaking isn't straight forward. The appears to be a few minute latency period after a tweak. One can't simply stop the phase angle from moving by turning the pot. Doing so results in an over correction a few minutes later. It's difficult to tweak the Rb standard beyond this degree of accuracy. It takes more than 20 minutes to determine a data point. To do better would take some automation and a phase detector running for several days to get sufficient statistics. Next project is building a phase detector..... I'm pleased. The Rb standard has lots of advantages: Price, minutes to locking on, portability and simplicity. The next question: How accurate is the GPS-DO? Anybody got a hydrogen maser handy? Now what to do with the GPS-DO? Remember when decent receiver dials had only 5 KHz markings? 73 de Brian/K3KO ----- No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1390 / Virus Database: 1518/3776 - Release Date: 07/20/11 ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
In reply to this post by alsopb
On Wed, Jul 20, 2011 at 7:08 PM, Igor Kosvin <[hidden email]> wrote:
> ...aging drift - better than 1e-11 per month. Useful life of LPRO - usually > 15 years. > Make your math how much it drifts over life.... =========== An interesting question. If the drift is always in the same direction, you get the (apparently) obvious answer. If, at the other extreme, each month's drift is independent of the previous month's, then the expected frequency at the end will be described by a Gaussian distribution with mean = the starting point, and standard deviation scaled by the square root of the number of months. In practice, one would expect some serial correlation between monthly drifts, so a bias would be added to the expected mean. Tony KT0NY ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
The drift rate of 1e-11 per month or so is regarded after one month of conituous operation. The first month after turn-on is different. It usually takes few days, sometimes couple weeks to reach the final drift rate. After that it is pretty much constant slope. Every time you turn off the clock and then turn i t on again, depends on the duration of off time, the frequency will drift initially at higher rate. It takes time to Rubidium atoms to reach equilibrium position in the cloud inside the resonance cell. When you turn the unit off, the atoms of rubidium (which is metal alk ali by the way) will fall down (gravity) and form little metal drop on the bottom of the cell. When you turn it on, the cell is heated and the atoms evaporate and go into the cloud form. By the way, ambient temperature variations have much higher effect on frequency than just aging. Magnetic fields are also detrimental. Don't place it in close proximity to CRT monitors or dynamic speakers. Igor, N1YX ----- Original Message ----- From: "Tony Estep" <[hidden email]> To: [hidden email] Sent: Thursday, July 21, 2011 11:33:54 AM Subject: Re: [Elecraft] Rb & GPS DO comparison On Wed, Jul 20, 2011 at 7:08 PM, Igor Kosvin <[hidden email]> wrote: > ...aging drift - better than 1e-11 per month. Useful life of LPRO - usually > 15 years. > Make your math how much it drifts over life.... =========== An interesting question. If the drift is always in the same direction, you get the (apparently) obvious answer. If, at the other extreme, each month's drift is independent of the previous month's, then the expected frequency at the end will be described by a Gaussian distribution with mean = the starting point, and standard deviation scaled by the square root of the number of months. In practice, one would expect some serial correlation between monthly drifts, so a bias would be added to the expected mean. Tony KT0NY ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
«
Return to [HAM]
|
1 view|%1 views
| Free forum by Nabble | Edit this page |