bg7tbl gpsdo master reference

[jpb]

Here is a review (and some measurements) of the distribution amplifier:
http://www.reeve.com/Documents/Articles%20Papers/Reeve_10MHzDist.pdf
It seems to be good but a cheaper option is to use a video distribution amplifier and modify it from 75 ohm to 50 ohm (or not bother modifying it and just use 75 ohm cable):
https://xdevs.com/doc/Extron/Extron_ADA_6_modifications_for_use_as_10MHz_distribution_amp_STEINMETZ.pdf
https://www.jackenhack.com/reference-10-mhz-distribution-amplifier/

There has been quite a lot of discussion on these forums on the latter approach and there is a lot of material on-line
https://www.eevblog.com/forum/testgear/diy-10-mhz-distribution-amplifier-requirements-gathering/
Gerry Sweeney built his reference into the distribution amplifier as there is a lot of spare space:
https://gerrysweeney.com/build-a-10mhz-rubidium-frequency-standard-and-signal-distribution-amp-for-my-lab/

[Electro Fan]

Another day, another impression.

There is something soothing about seeing all the 0s on the counters and seeing waveforms on scopes that are just rock solid with no drift.

I'm starting to think that if I were to purchase any more test equipment I'd probably make sure there is a 10 MHz input included with almost anything I'd consider.

Given the overall investment in test equipment the price of a GPSDO as a frequency and clock source seems like a very good investment.

[jpb]

Another day, another impression.

There is something soothing about seeing all the 0s on the counters and seeing waveforms on scopes that are just rock solid with no drift.

I'm starting to think that if I were to purchase any more test equipment I'd probably make sure there is a 10 MHz input included with almost anything I'd consider.

Given the overall investment in test equipment the price of a GPSDO as a frequency and clock source seems like a very good investment.

It is amazing that for very little money you can acquire a source with an accuracy far higher than most expensively equipped  labs a few decades ago. An accuracy of 10^-11 means the error over one second is the time span than light takes to travel just 3 mm!

[Electro Fan]

Another day, another impression.

There is something soothing about seeing all the 0s on the counters and seeing waveforms on scopes that are just rock solid with no drift.

I'm starting to think that if I were to purchase any more test equipment I'd probably make sure there is a 10 MHz input included with almost anything I'd consider.

Given the overall investment in test equipment the price of a GPSDO as a frequency and clock source seems like a very good investment.

It is amazing that for very little money you can acquire a source with an accuracy far higher than most expensively equipped  labs a few decades ago. An accuracy of 10^-11 means the error over one second is the time span than light takes to travel just 3 mm!

Roger that.  Taped to one of my pieces of gear is a chart that reminds me that in 1ps light travels .3mm
But now that we have a taste it's hard not want some more ......

So, more rookie questions:

I'm thinking about deploying the distribution amp with several identical length and terminated coax cables for the various pieces of test equipment.  I have a line on some good coax but each cable might need a specific (SMA to BNC) adapter to keep everything uniform.  I'm not too worried about the ~0.1 ? dB loss from the adapters, but I'd like to avoid taking the fairly precise setup that is evolving and introduce unwanted reflections, waveform distortions, and maybe/especially any out of phase conditions.  I'll do some experimenting but any advice on how to keep things pristine would be appreciated.  For example, there will be a connection from the GPSDO to the distribution amp and then connections from the distribution amp outputs to each piece of TE.  The TE interfaces will generally be BNC (but as mentioned my cables will require a SMA to BNC adapter); more to the point, each piece of TE will have a 50 Ohm BNC interface.  Am I going to want to come up with ~6-8 50 Ohm through terminators (one for each dist amp output connector), and is it also necessary to then put 50 Ohm through terminators on the GPSDO 10 MHz output and the distribution amp input?

Thanks!

[Electro Fan]

In Lady Heather is there a way to figure out which of the colorful satellite symbols displayed on the satellite positions map are associated with the specific PRNs in the PRN/AZ/EL/SNR table?  (I've read the LH manual but might have missed it.)

(It would be cool if you could click on a PRN and have it highlight the satellite on the map, and/or vice versa.)

Update:  problem solved.  Just needed to stare at it and think about it longer.  The colored numbers in the four corners pretty much spell it out.  I was initially trying to see those numbers as the SNRs for the satellites in their respective quadrants but that didn't make sense given the relative sizes of each satellite (which I knew did/does reflect their relative signal strengths); in fact, the colored numbers in the corners are the PRNs, and the arrows (if present) show the direction being taken by each satellite. 

And I think if you have a Signal strength vs as/el graphic that shows a wide range of signal strengths that you will find the strongest areas of coverage correlate to where you generally see the "strongest" ("biggest") satellites on the satellite position map. 

The math and code (not to mention the understandings of physics and technology) that was required to develop LH must be very very cool.

[FransW]

The shortest time step is the time that a photon needs to travel the Planck distance.

This timestep is 10^(-44) seconds. We have to wait a while for this barrier to show up.
Or fill the gap with more phyisical knowledge. Wiki:

time unit is the time required for light to travel a distance of 1 Planck length in a vacuum, which is a time interval of approximately 5.39 × 10 −44 s.[1] The unit is named after Max Planck, who was the first to propose it.


The Planck time is defined as:[2]

{\displaystyle t_{\mathrm {P} }\equiv {\sqrt {\frac {\hbar G}{c^{5}}}}} {\displaystyle t_{\mathrm {P} }\equiv {\sqrt {\frac {\hbar G}{c^{5}}}}}
where:

ħ = ​h⁄2π is the reduced Planck constant (sometimes h is used instead of ħ in the definition[1])
G = gravitational constant
c = speed of light in vacuum
Using the known values of the constants, the approximate equivalent value in terms of the SI unit, the second, is

{\displaystyle 1\ t_{\mathrm {P} }\approx 5.391\,16(13)\times 10^{-44}\ \mathrm {s} ,} {\displaystyle 1\ t_{\mathrm {P} }\approx 5.391\,16(13)\times 10^{-44}\ \mathrm {s} ,}
where the two digits between parentheses denote the standard error of the approximated value.

https://en.wikipedia.org/wiki/Planck_time

Frans

[texaspyro]

In Lady Heather is there a way to figure out which of the colorful satellite symbols displayed on the satellite positions map are associated with the specific PRNs in the PRN/AZ/EL/SNR table?

The next version lets you put the mouse cursor in the satellite map and it shows the info for the nearest satellite below the map.  Also if you display the satellite map as a full screen zoomed image, the satellite system identifier letter is shown on the satellite.

Also the next version will show the signal level at the mouse cursor below the signal level map (as long as there is enough space on the screen to show it).  Also the algorithm for coloring in the signal level map has been improved.

[Electro Fan]

In Lady Heather is there a way to figure out which of the colorful satellite symbols displayed on the satellite positions map are associated with the specific PRNs in the PRN/AZ/EL/SNR table?

The next version lets you put the mouse cursor in the satellite map and it shows the info for the nearest satellite below the map.  Also if you display the satellite map as a full screen zoomed image, the satellite system identifier letter is shown on the satellite.

Also the next version will show the signal level at the mouse cursor below the signal level map (as long as there is enough space on the screen to show it).  Also the algorithm for coloring in the signal level map has been improved.

Cool!

[Electro Fan]

The shortest time step is the time that a photon needs to travel the Planck distance.

This timestep is 10^(-44) seconds. We have to wait a while for this barrier to show up.
Or fill the gap with more phyisical knowledge. Wiki:

time unit is the time required for light to travel a distance of 1 Planck length in a vacuum, which is a time interval of approximately 5.39 × 10 −44 s.[1] The unit is named after Max Planck, who was the first to propose it.


The Planck time is defined as:[2]

{\displaystyle t_{\mathrm {P} }\equiv {\sqrt {\frac {\hbar G}{c^{5}}}}} {\displaystyle t_{\mathrm {P} }\equiv {\sqrt {\frac {\hbar G}{c^{5}}}}}
where:

ħ = ​h⁄2π is the reduced Planck constant (sometimes h is used instead of ħ in the definition[1])
G = gravitational constant
c = speed of light in vacuum
Using the known values of the constants, the approximate equivalent value in terms of the SI unit, the second, is

{\displaystyle 1\ t_{\mathrm {P} }\approx 5.391\,16(13)\times 10^{-44}\ \mathrm {s} ,} {\displaystyle 1\ t_{\mathrm {P} }\approx 5.391\,16(13)\times 10^{-44}\ \mathrm {s} ,}
where the two digits between parentheses denote the standard error of the approximated value.

https://en.wikipedia.org/wiki/Planck_time

Frans

Future generations of EEVbloggers will need a better "GPSDO", some successor to texaspyro's version of Lady Heather, a better "distribution amp", and cables with a really good Velocity Factor to handle this.

And the level of precise machine synchronicity (and possibly human thinkronicity) that might come with this would probably be astonishing.  I think we might be in a golden era of "it's inexpensive enough for lots of people to have access to amazing stuff by historical standards and there's room for lots of freedom and creativity to learn, apply, and enjoy it" - without being completely chained to it.  Hopefully the golden era will just expand and get better.  Knock on wood.

[Electro Fan]

Near the center top of Lady Heather's main screen are values for PDOP, HDOP, and VDOP.  In the pdf manual there is one lonely reference to PDOP and none (that I found) for HDOP and VDOP.

In case anyone wants some info on these, here you go:

https://www.anatumfieldsolutions.com/What-is-PDOP-And-Why-its-Obsolete_b_43.html

Spoiler alert:  according to the article it seems we should be somewhat happy with less than 4 for PDOP. I've been watching to see if there is any correlation between PDOP (and HDOP and VDOP), the DOP line in the graph, and the proximity of the LCD readout to 10 MHz (10,000,000.0000Hz).  Only briefly (for a few seconds) have I seen a PDOP slightly over 4, mostly it's in the high 1s to lower 2s, and currently it's 1.97 - and it's 4 zeros to the right of the decimal point in the LCD Hz counter.  Oops, PDOP just went to 1.91 (lowest I've seen) but the LCD Hz counter went to .0001 briefly, and then back to .0000.

Lowest HDOP I've seen is 1.04 and lowest VDOP is 1.59 - but I've only been watching for about 30 minutes (an hour ago I didn't know these things even existed.)  Just hit 1.90 on PDOP but it's very slowly toggling between .0001 Hz and .9999 Hz.  I'm guessing the GPS and the OCXO are chasing each other around.  Now it's back to .0000 at a PDOP of 1.99.  Guess we have some more #s to watch.   

So within the category of Time Nuts might be a category of PDOP Nuts.  I'm thinking there is probably a cure for this or it's just temporary.  (until something new to measure comes along)

Update:  new best #s below but I'm guessing given my limited antenna view that guys with better antennas and much better views of the sky are getting much lower numbers.

Update:  My bad, should have done a search around here first.  Turns out people have been talking about these #s since Dec 2014. But I'm still feeling sort of like Columbus discovering a new land (when people were already living there).

Speaking of sailing, time, lat, and long if you haven't seen this it's pretty amazing (not the video quality but the content):

[texaspyro]

Depending upon the device type there can be HDOP, VDOP, PDOP, TDOP, GDOP, EDOP, and NDOP (alas, no BeBopDOP).  The DOP plot is the average of all the various DOP values the receiver sends.  Probably the most relevent DOP for GPSDOs is TDOP.

[Electro Fan]

Roger that and Thanks texaspyro.

Now that I know TDOP is a thing, I was able to search on it along with some of it's other relatives

https://the5krunner.com/2018/07/18/gps-accuracy-hdop-pdop-gdop-multipath-the-atmosphere/
https://www.gsat.us/support/glossary/tdop
https://nptel.ac.in/courses/Webcourse-contents/IIT-KANPUR/ModernSurveyingTech/lectureB_11/B_11_3GDOP.htm
https://en.wikipedia.org/wiki/Dilution_of_precision_(navigation)
http://www2.unb.ca/gge/Resources/gpsworld.may99.pdf

[Electro Fan]

First Question:  Is 37 the currently correct value for the UTC leap second offset in a GPSDO?
Does manually changing that field have any impact on any of Lady Heather's calculations, or anything else with respect to GPSDO use?

---
https://en.wikipedia.org/wiki/Leap_second

A leap second is a one-second adjustment that is occasionally applied to civil time Coordinated Universal Time (UTC) to keep it close to the mean solar time at Greenwich, in spite of the Earth's rotation slowdown and irregularities. UTC was introduced on January 1, 1972, initially with a 10 second lag behind International Atomic Time (TAI). Since that date, 27 leap seconds have been inserted, the most recent on December 31, 2016 at 23:59:60 UTC, so in 2018, UTC lags behind TAI by an offset of 37 seconds.[1]

https://www.timeanddate.com/time/leapseconds.html

Last Leap Second in December 2016

The last leap second was added on December 31, 2016, at 23:59:60 UTC. The difference between UTC and International Atomic Time (TAI) increased from the 36 seconds to the current 37 seconds.

Next Leap Second

Based on current predictions, the next leap second should be added on June 30, 2020. However, since the speed of the Earth's rotation is subject to unpredictable short-term variations, the date may still change. Once it is officially announced, we will publish it in our Time Zone News.
---

Second question:  is the main/big digital clock just a repeat in larger size of the local time in the upper left hand corner of Lady Heather, or is the main/big digital clock intended to display UTC?  If so, how do you get the local time in the upper left hand corner and UTC in the main/big digital clock?  I have studied the /tz= instructions and I can't figure out how to make both local and UTC time display - I can get one or the other to show in both locations, but I can't figure out how to make LH show both.  I know, kind of duh..... 

[Electro Fan]

Looks like I found the answer to question 1:

It's 18 seconds vs GPS.  The 37 seconds offset is vs atomic clock time.
https://endruntechnologies.com/support/leap-seconds
http://leapsecond.com/java/gpsclock.htm

LH does seem to take into account the specified offset for GPS time when T is set for GPS.

It's not so clear what LH is doing to the UTC displayed time when the offset is adjusted.  ??

Still curious to know if it's possible for LH to display both local time in the upper left and UTC on the big/main digital clock.

[Electro Fan]

I compared the antenna that came with the BG7TBL (which in my case was a C & Q GPS DAMA1575AT41 from CAIQIN TECHNOLOGY http://www.cai-qin.com) with a ANN-MB-00-00 antenna from ublox (https://www.u-blox.com/en/product/ann-mb-series) and it was a little different than what I expected.

Disclaimer - nothing at all about this was scientific, it was just first impressions.

The ublox is larger and heavier and feels very sturdy, but after trying both in the same dozen or so similar locations it seems that they were either tied or the C & Q won each time.  In general, in my tree challenged window location, they either both produced one antenna bar next to the antenna symbol on the GPSDO LCD or the ublox produced one bar and the C & Q produced 2 bars.

I tried to take some screen shots of the satellite SNRs in LH each time I changed out the antennas but it's kind of hard to determine much because the satellites are moving of course.  However, it did seem from the SNRs and maybe also the number of satellites that had green values vs yellow values that the C & Q was producing either comparable or better results.

They both say RG174 on their attached cables but the ublox cable is a bit thicker (~3.02mm vs ~2.97mm).  Maybe there is some attribute that makes the C & Q a better match with the BG7TBL or maybe it's just a better antenna.  Or maybe my testing is just a misleading first impression.  YMMV

[texaspyro]

is the main/big digital clock just a repeat in larger size of the local time in the upper left hand corner of Lady Heather, or is the main/big digital clock intended to display UTC?

The Heather displays whatever time the receiver is sending, possibly adjusted for your specified time zone.  If you set Heather to display GPS time (TG command) there will be no leapsecond adjustment.  If set to display UTC time (TU command), the time will be UTC.  The TZ time zone command lets you set your time zone offset (and time zone name(s)) If a time zone has been set, the UTC time will be adjusted for your time zone.   Heather can also automatically adjust for daylight savings time.

The big clock, small clock, and analog watch always show the same time.

[FransW]

Coax cable attenuation listings (incl RG316):

http://rfelektronik.se/manuals/Datasheets/Coaxial_Cable_Attenuation_Chart.pdf

Frans

[Involute]

Photos of BG7TBL 2019-03-25 (1/2)

[Involute]

Photos of BG7TBL 2019-03-25 (2/2)

[Fennec]

Did you check the "UBlox" modul on version and firmware? And the VFC Voltage at the OCXO?
Because I have changed mine to an UBlox M8T and a brand new UCT 108663-01 double oven.
Works way more stable than the original crap (compared with 3 Rubidium standards).

[kj7e]

Did you check the "UBlox" modul on version and firmware? And the VFC Voltage at the OCXO?
Because I have changed mine to an UBlox M8T and a brand new UCT 108663-01 double oven.
Works way more stable than the original crap (compared with 3 Rubidium standards).

Fennec, just to confirm, you found the M8T to be both a direct replacement and better performance than the M7?

https://www.u-blox.com/sites/default/files/products/documents/NEO-LEA-M8T_ProductSummary_%28UBX-16000801%29.pdf

https://www.digikey.com/product-detail/en/NEO-M8T-0/672-1010-1-ND/6150673/?itemSeq=295031672

[kj7e]

Decided to play around a bit, changed out the Ublox NEO-M7 for the NEO-M8T.  Took a bit of playing with U-Center to get the MCU to run again and provide data on the LCD display.  Had to update the NMEA Protocol Main Talker ID to GPS then everything started working again.  Unfortunately, I really have no way to see if this made an improvement or not.  Mostly changed out the chip just to play with it.

[Electro Fan]

Slightly off topic but after using a BG7TBL I'm trying a Star 4+.  Seems like a number of differences (as reflected in the info displayed by LH) but at least one that I can't figure out is the PRNs.  On the BG7TBL I don't think I ever saw a number higher than about 51 but in any event I'm almost certain I never saw a three digit PRN.  With the Star 4+ I'm getting PRNs in the 130s.  The antennas look to be somewhat similar in design and they are sitting about an inch from each other.  Any reason/significance the BG7TBL doesn't see higher PRNs (or that the Star 4+ does)?

PS, what's a typical timeframe for the Self Survey to complete upon first startup?

[texaspyro]

PRNs around 130 are SBAS/WAAS (augmentation) sats in geosynchronous orbit.

[Electro Fan]

cool Thx

BG7TBL doesn’t notice them because a different receiver is used?