Did anyone ever do a comparison between their bg7tbl gpsdo and a different reference to see if the output frequency is 10MHz? I'm using mine (version 2017-03-29) to feed a clock (date + time). Because that clock wanders compared to an ntp synchronised clock I measured the output frequency and it turns out my bg7tbl gpsdo is 1.5Hz too low!
Needless to say this thing starts to look more and more like a waste of money.
BG7TBL has several different GPSDOs. Only the first version had a known frequency error and it was off in the 1E-9 range. If you are 1.5 Hz, something is wrong... and that something could be the reference you are comparing it to.
Is that an NTP synchronized clock thats synchronized to a GPS 1pps via kernel discipline, or is it running off the GPSDO 10 MHz?
They would likely be different it sounds -
With the caveat that I have still been too lazy to recompile my Raspberry Pi's kernel to get an extra two orders of magnitude with my plain GPS time server, I fully do expect that when i do, I will get much more accurate time from my GPSDO and its 1PPS than I do from my GPS. Even though its a good one.
However, lacking the so called kernel discipline its still accurate to better than a microsecond so I cant complain.
NTP connected clocks do have error. Various kinds of wander occur and can be characterized. So what you see is / are likely the combined effect of a number of different sources of error common to computers in that situation. Including clock and network latency. Thats far more likely.
I think "Chrony" is better than NTP in that respect. I'm planning to hook my GPSDO up to a machine I have that has a real ethernet card not a USB one, and real GPIOs and I'm also planning to switch my RPI to Chrony one of these days. Then I'll have two or when my desktop is running three sources of network time to serve as sanity checks for one another.
Did anyone ever do a comparison between their bg7tbl gpsdo and a different reference to see if the output frequency is 10MHz? I'm using mine (version 2017-03-29) to feed a clock (date + time). Because that clock wanders compared to an ntp synchronised clock I measured the output frequency and it turns out my bg7tbl gpsdo is 1.5Hz too low! Needless to say this thing starts to look more and more like a waste of money.
Did anyone ever do a comparison between their bg7tbl gpsdo and a different reference to see if the output frequency is 10MHz? I'm using mine (version 2017-03-29) to feed a clock (date + time). Because that clock wanders compared to an ntp synchronised clock I measured the output frequency and it turns out my bg7tbl gpsdo is 1.5Hz too low! Needless to say this thing starts to look more and more like a waste of money.
Are you sure the NTP clock is as good? Usually network time is only good down to milliseconds or hundreds of microseconds because of the time it takes to query the server and the differing travel times - networks that need very precise network time usually have a local stratum1/2 reference and then are compensate for cable run length.
If you've got a real stable reference (and even a good OCXO will not be absolutely accurate without some sort of disciplining because of drift and aging) and a frequency counter to match up against, then I'd trust the measurement, but less than that.... I doubt the error is with the GPSDO (unless it's not GPS locked, in which case it can certainly drift).
Did anyone ever do a comparison between their bg7tbl gpsdo and a different reference to see if the output frequency is 10MHz? I'm using mine (version 2017-03-29) to feed a clock (date + time). Because that clock wanders compared to an ntp synchronised clock I measured the output frequency and it turns out my bg7tbl gpsdo is 1.5Hz too low! Needless to say this thing starts to look more and more like a waste of money.
Unless my maths is wrong (wouldn't be the first time) 1.5Hz is 0.15PPM or about 13ms a day. How long a period did you measure it over and how did you compare it against your NTP synced clock?
Did anyone ever do a comparison between their bg7tbl gpsdo and a different reference to see if the output frequency is 10MHz? I'm using mine (version 2017-03-29) to feed a clock (date + time). Because that clock wanders compared to an ntp synchronised clock I measured the output frequency and it turns out my bg7tbl gpsdo is 1.5Hz too low! Needless to say this thing starts to look more and more like a waste of money.
Are you sure the NTP clock is as good?
Yes, it is one from pool.ntp.org. The clock wanders off by several seconds within a couple of minutes and NTP and the RTC crystal driving the other clock are not that bad. OTOH I now realise that 1.5Hz on 10MHz is around 1ppm so the whole situation doesn't make much sense.
That's so much error it doesn't make sense. Even my plain $7 GPS would have much less error than that.
Did anyone ever do a comparison between their bg7tbl gpsdo and a different reference to see if the output frequency is 10MHz? I'm using mine (version 2017-03-29) to feed a clock (date + time). Because that clock wanders compared to an ntp synchronised clock I measured the output frequency and it turns out my bg7tbl gpsdo is 1.5Hz too low! Needless to say this thing starts to look more and more like a waste of money.
Are you sure the NTP clock is as good?
Yes, it is one from pool.ntp.org. The clock wanders off by several seconds within a couple of minutes and NTP and the RTC crystal driving the other clock are not that bad. OTOH I now realise that 1.5Hz on 10MHz is around 1ppm so the whole situation doesn't make much sense.
It turns out other parts of the software on the system are messing with the system clock every now and then. I didn't expect that.
Once you have to look across an Ethernet network you lose precision because of the inherent latency of using a network. If you want precision you are better off distributing your 1PPS or 10 MHz signals via some simpler non-networked means.
FWIW Leo Bodnar's unit from SDR-kits is about the same price and provides a choice of two output frequencies. Very useful for things that don't accept a 10 MHz input such as the SDRplay RSP2. Probably my next purchase after my VNWA 3EC arrives.
Of the units listed in the OP, which would be the most desirable?
Are any much better than the others?
I'm looking for a GPSDO and really don't know much about them.
Would these BG7TBL units be considered the best bang for the buck or should I look elsewhere for a ready to go box.
Hopefully when someone purchases the new unit; they can do a teardown.
I got one of his Star-4 based units a year ago. It did not have the management interface connected to the serial port. I added an external RS-232 interface and had been using it that way. Last week I went in and re-wired the internal serial port to the management interface.
Here is the ADEV, etc measurements on my Star-4 based unit. I used a TAPR TICC clocked by an HP-5071 cesium beam oscillator. The room temp varied over 2 degrees C and had no effect on the frequency output... the 8663 oscillator using in them is a very good oscillator and the "ATDC" versions of the Star-4 use their highest spec'd model.
So that is substantially better than the TruePosition, right?
Are the ADEV graphs you posted of the TP on TimeNuts last year directly comparable to this one (were the conditions similar enough to be directly comparable?)
I could post them but since they are yours...
There were two that
I saw. The Star 4 based unit looks like it could be considerably better if those two charts are directly comparable to this one.
Are the ADEV graphs you posted of the TP on TimeNuts last year directly comparable to this one (were the conditions similar enough to be directly comparable?)
I could post them but since they are yours...
The setup was the same. I don't remember how long the Truposition had been powered up... the plot showed 6 hours of data whereas the Star had 13 hours and had been on for a day.
I got one of his Star-4 based units a year ago. It did not have the management interface connected to the serial port. I added an external RS-232 interface and had been using it that way. Last week I went in and re-wired the internal serial port to the management interface.
Here is the ADEV, etc measurements on my Star-4 based unit. I used a TAPR TICC clocked by an HP-5071 cesium beam oscillator. The room temp varied over 2 degrees C and had no effect on the frequency output... the 8663 oscillator using in them is a very good oscillator and the "ATDC" versions of the Star-4 use their highest spec'd model.
After I saw your message about the unit, I went and purchased that one. I'm been very happy with the unit. I would agree that the oscillator is very good.
There is one on the auction site which has a blue lcd screen and not listed in the OP.
I guess this must be the latest installment in this series of GPSDO's.
Has anyone done a teardown of this unit yet?
How is your Lady Heather 6 beta coming along, are you ready to release your code to the public?
First i've heard of this! Any available info links?
Hello,
New GPSDO 10 Mhz made in BG7TBL :
I bought this GPSDO 10 Mhz in aluminum box via EBay (or aliexpress) for +/- 98$ and I am very satisfied.
It can provide a square or sinusoidal wave by moving a component on the output.
Pins 14-15-16 Neo7 are open, but they can be arranged to vary the baud rate.
Edit : After verification, this new GPSDO - PLL bought $98 contains an OCXO, not a VCXO.
The metal case Ocxo becomes very hot after a few minutes of operation.
The deal is excellent !!
----
-
https://www.eevblog.com/forum/testgear/new-gpsdo-bg7tbl/Diabolo
I'm using Ver 5 of Lady Heather with a Thunderbolt GPS. When I first invoked the L/H it came up with co-ordinates for probably KE5FX so in addition to changing the time zone for Perth in Western Australia I successfully changed the co-ordinates at which stage L/H now says there are no sats usable. Why - is it ?perhaps the fact that KE5's IP address still shows? How do I get rid of it as I'm connected to my PC via Com1.
I know my GPS antenna is OK as on another piece of equipment it shows 8 satellites currently available.
Before I made any changes L/H was showing 6 satellites OK.
Any help appreciated Merv VK6BMT
Hi Merv,
Your lady heather configuration should not show an ip address when you are connected via a COM port.
When the program is active type ? and you will see all of the startup commands available (a lot of them!), you can change the COM port by simply typing /1 in your case, that should clear things up.
You are probably running the example configuration that connects to John Miles net connected Thunderbolt in Seattle.