On Mon, Jan 30, 2012 at 9:47 AM, Mark Sims holrum@hotmail.com wrote:

Inside "Packet 0x8F-AC" (supplemental timing packet) there are two
data fields named "PPSOset and 10MHZOset.  Each is a 4-bit signed
integer with units of nanoseconds for PPSOset and in pars per billion
for 10MHZOset.  They are used to report an estimate of the PPS and
10MHz signals offset from UTC.  Typically the values reported are not
constant and in the single digit range.  (I'm reading from section
A.10.31 of the Thunderbolt manual.)

This is what I meant when I said "sawtooth".  Maybe the wrong term.
But I think theses numbers can be used in a controller.

Chris Albertson
Redondo Beach, California

Mark,
If you feel adventurous, my GPSMonitor (written in C for the 8051) source
code is available. Development tools including C compiler are free.

Didier KO4BB

On Mon, Jan 30, 2012 at 11:47 AM, Mark Sims holrum@hotmail.com wrote:

On Mon, 30 Jan 2012 17:47:54 +0000
Mark Sims holrum@hotmail.com wrote:

I just wonder, whether one could take a GPS module, like the LEA-6T
and replace the TCXO they have with a VCXO that is phase locked
to the 10MHz reference. Or would that cause too much of the hanging
bridge problem?

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

On Tue, Jan 31, 2012 at 7:52 AM, Attila Kinali attila@kinali.ch wrote:

I'm trying to figure out your goal.  The above assumes one already
has a 10MHz reference.

Chris Albertson
Redondo Beach, California

On Tue, 31 Jan 2012 09:32:22 -0800
Chris Albertson albertson.chris@gmail.com wrote:

Let's say, i'm building a GPSDO with a high quality OCXO.
Wouldnt it then make sense to lock the reference clock of the GPS
receiver also to that OCXO?

		Attila Kinali

--
Why does it take years to find the answers to
the questions one should have asked long ago?

Exactly that is the appeal of the Tbolts. What frequency does the uBlox
6T TCXO have? Let us know your progress.

--

Björn

On Tue, 31 Jan 2012 18:50:08 +0100
bg@lysator.liu.se wrote:

Yes, but can this be replicated with a standard GPS module?
And does it have side effects?

48MHz as stated in the Timing AppNote.

My current progress is that the uC i wanted to use does not
do what i want. Can anyone recommend a uC with 32bit timers
and IEEE 1588 support?

The alternative would be to use an FPGA, but i'm reluctant to
do that as it makes the whole system a lot more complex.

		Attila Kinali

--
Why does it take years to find the answers to
the questions one should have asked long ago?

El 31/01/2012 20:43, Attila Kinali escribió:

You can have a look on these
http://www.ti.com/mcu/docs/mculuminaryfamilynode.tsp?sectionId=95&tabId=2597&familyId=1756
Some of them have IEEE-1588, like the LM3S9B96

Regards,

Javier

On Tue, 31 Jan 2012 20:43:35 +0100, Attila Kinali wrote:

Some of the ST's supports IEEE-1588
http://www.embeddedstar.com/weblog/2011/09/21/stm32-f4-mcu/
www.mouser.com/pdfdocs/BRSTM32F4_v7final.pdf

I'd suggest a $20 STM32-F4-Discovery (Digikey etc)
http://parts.digikey.com/1/parts/2484058-eval-kit-stm32f-discovery-
stm32f4discovery.html

And a Phy (The mac is in the F4) $10
http://www.ebay.de/itm/DP83848-Ethernet-Physical-Transceiver-RJ45-
connector-control-interface-Board-Kit-/260868624339

Some www-sw is here (German use google translate)
http://www.mikrocontroller.net/topic/237223#2491080

The F4 even supports "Single precision" Floats in HW ,
and has both ADC & DAC (with the ADC/DAC vref pin on external pin).
1G flash , 192k Ram , and it runs 168Mhz.
I think there is 2 x 32bit timers (Fclock/2) , and a fair amount of 16bit
timers.

It comes with build in debugger (proprietarary) , but the debug protocol
has been reversed ... (see texane project) , and afaik it is now suported
in openocd. So arm-gcc & arm-gdb can be used (ie codesourcery).

ST suggests Atollic as "gcc" toolchain , but that's imho "handicapped" in
the free version.

I have earlier suggested the F4-Discovery , as a "Tnut" GPSDO/OCXO/RBXO
controller board.

CFO

On Tue, 31 Jan 2012 21:06:04 +0100
Javier Herrero jherrero@hvsistemas.es wrote:

This was exactly the device i intended to use.
But it doesnt really have 32bit timers. They cascade two 16bit timers
to get 32bit, but then all kind of restrictions apply which make the
timers unusable. And when using 16bit timers, i'll get an overflow
every 800us (at 80MHz clock).

I browsed trough the internets and could find the following devices:

LPC18xx (NXP):
Look nice, but still in development or early production.
I especially like that they (will be) are available in a 200 pin QFP.
This would enable to use all the functions of the chip while still
being able to solder one by hand.

K60 (Freescale): Hell of a confusing documentation. Also quite new.
Have only 16bit timers. A big issue is that they have a crypto unit
on chip, which makes them export restricted. Ie the only way to buy
them is from a local distributor which makes them expensive.

STM32-F2/F4 (ST): ST doesn't want to give me the documentation to those.
(website fails w/o error message)

I havent found any other chips yet. Sofar the options i see are:

• wait for the LPC18xx become available in quanities (can take a year)
• Use a LM3S9B96 together with a small FPGA to implement the counter functions.

Any hints appreciated

		Attila Kinali

--
Why does it take years to find the answers to
the questions one should have asked long ago?

If you are willing to use external counters:

Dallas DS2423 Dual 32-bit Counter
Or, if you have an unlimited budget:
http://www.cwcelectronicsystems.com/dual32ct_modulario.html
the word "defense" is prominent here :-).
Or counter products from: http://www.lsicsi.com/
I use encoder interfaces from the last company and recommend them!
Don

Attila Kinali

--
"Neither the voice of authority nor the weight of reason and argument
are as significant as experiment, for thence comes quiet to the mind."
R. Bacon
"If you don't know what it is, don't poke it."
Ghost in the Shell

Dr. Don Latham AJ7LL
Six Mile Systems LLP
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
www.lightningforensics.com
www.sixmilesystems.com

In message deac00aacc21732c4705e34cd4097d7a.squirrel@www.webmail.montana.com,
"Don Latham" writes:

An suitable motherboard and an intel 82599 based ethernet card ?

The latter will set you back approx $700, but $1000 should get you far
in total.

The 82599 has some very interesting time-nuts features with respect
to the GPIO pins.  Unfortunately I have not been able to find out if
these pins are accessible on the Intel ethernet cards.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

If anyone is interested, I have just got hold of a PDF of the Technical Manual TM 5680-0211 for 5680A series Rubidiums.
Please contact me off list for a copy (1M, so too large to post on time-nuts@febo.com)

Rob Kimberley

-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On Behalf Of Attila Kinali
Sent: 31 January 2012 20:47
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] FE-.5680A trimming resolution

On Tue, 31 Jan 2012 21:06:04 +0100
Javier Herrero jherrero@hvsistemas.es wrote:

This was exactly the device i intended to use.
But it doesnt really have 32bit timers. They cascade two 16bit timers to get 32bit, but then all kind of restrictions apply which make the timers unusable. And when using 16bit timers, i'll get an overflow every 800us (at 80MHz clock).

I browsed trough the internets and could find the following devices:

LPC18xx (NXP):
Look nice, but still in development or early production.
I especially like that they (will be) are available in a 200 pin QFP.
This would enable to use all the functions of the chip while still being able to solder one by hand.

K60 (Freescale): Hell of a confusing documentation. Also quite new.
Have only 16bit timers. A big issue is that they have a crypto unit on chip, which makes them export restricted. Ie the only way to buy them is from a local distributor which makes them expensive.

STM32-F2/F4 (ST): ST doesn't want to give me the documentation to those.
(website fails w/o error message)

I havent found any other chips yet. Sofar the options i see are:

• wait for the LPC18xx become available in quanities (can take a year)
• Use a LM3S9B96 together with a small FPGA to implement the counter functions.

Any hints appreciated

		Attila Kinali

--
Why does it take years to find the answers to the questions one should have asked long ago?

time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

El 31/01/2012 21:47, Attila Kinali escribió:

I see... for edge count and edge time, only individual timers and not
concatenates. I've had a look to the LM3S9D96, and it is more or less
the same (the exception is that it is marketed as 4 32-bit timers that
can be used as 8 16-bit timers :) ), and I suspect that all the family
will have similar behaviout. A pity... since for Ethernet it includes
the LAN and PHY on-chip.

Regards,

Javier

On Tue, 31 Jan 2012 21:47:15 +0100, Attila Kinali wrote:

I have no probs with the ST site (Discovery-F4)
http://www.st.com/internet/evalboard/product/252419.jsp

You want these for the MCU
http://www.st.com/internet/mcu/product/252140.jsp

DataSheet
http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/
DATASHEET/DM00037051.pdf

Errata
http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/
ERRATA_SHEET/DM00037591.pdf

Reference Manual
http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/
REFERENCE_MANUAL/DM00031020.pdf

CFO

The Navsync FTS125 is an example where the GPS receiver engine (the CW25)
is driven by a 20MHz fixed OCXO. At the moment I don't know if the CW25 of
the FTS125 has a specific firmware for that but I suspect that it must be
so. In my opinion it is best to have a tunable OCXO (like the TBolt) to
have the best performance.

On Tue, Jan 31, 2012 at 10:19 PM, cfo xnews2@luna.kyed.com wrote:

On Tue, Jan 31, 2012 at 9:40 AM, Attila Kinali attila@kinali.ch wrote:

Yes, I see.  That is exactly what Trimble does in the Thunderbolt.
There is only one OCXO in the T-bolt that is used for both the output
10MHz and to run the receiver.  That system works well for them

Chris Albertson
Redondo Beach, California

On Tue, Jan 31, 2012 at 11:43 AM, Attila Kinali attila@kinali.ch wrote:

Does the system need to be small?  If not Generic PC hardware can
work.  Buy an Intel "Atom" main board.  For under $85 you get a
soldered down CPU and all the normal PC stuff, PCI bus and all.  The
board I bought does not have a CPU fan and burns all of about 5 watts.
Or you can re-cycle and old notebook computer.  If you run Linux
then yes it supports  IEEE 1588.  There are real-time versions of
linux that give you easy access to low level hardware much like with a
uP.

If you were going to mass produce these you could cut costs but for a
one-off using a $100 PC is reasonable.

Chris Albertson
Redondo Beach, California

Rob, I would be happy to put it on one of my web pages so people could
download it.

On Wed, Feb 1, 2012 at 8:14 AM, Rob Kimberley
robkimberley@btinternet.comwrote:

It's a design decision. Most GPSDO sold are made by companies
that buy an OEM GPS timing receiver and then create a GPSDO
with the usual assortment of OCXO, TIC, computer and DAC.

The exception would be if you are a GPS company and can
integrate the components, as Trimble did with the Thunderbolt.

But most companies selling GPSDO, and all amateurs building
homebrew GPSDO, have to use the separate component method.

Note the really high-end GPS timing receivers use an external
reference clock and don't even bother to lock it. The advantage
with this is 1) you don't need the DAC (which just adds noise),
2) you can use good clocks like cesium or masers which don't
have EFC, and 3) you collect the phase error information for
post-processing -- which is much more accurate than trying to
steer an oscillator in real time. And also, 4) you can run multiple
receivers off the same clock if necessary.

We're waiting for some brave soul to implement an SDR-based
GPS timing receiver; we can all then experiment with the TBolt
model instead of the TIC/DAC model of GPSDO.

/tvb

There are several projects that do this.

There is one written using GNU Radio.  I forget the details but I saw
it years ago.

http://www.kamieniecki.com/krys/gps
proof of concept only, not finished

http://gps.psas.pdx.edu/
Not really SDR but
... firmware that is meant to run on the receiver board itself, giving
you direct access to the GPS chipset.

Chris Albertson
Redondo Beach, California

On 31/01/12 20:43, Attila Kinali wrote:

Yes. If you look up old papers, they already did this with Oncores,
Cesium clocks and synthesis.

If you've done it right, it lowers the receivers noise. The position
becomes less shaken by the jerk effect of the oscillator noise, while
frequency and drift also affects the solution to some degree. This
side-effect is covered in literature if you look for it.

Produceable.

No pain, no gain?

Cheers,
Magnus

On 31/01/12 22:14, Rob Kimberley wrote:

I'll have it.

The russian version is here:
http://morion.com.ru/uploaded/FE-5680_manual_rus.pdf

Browing through it, it becomes obvious that this describes the new
5680As that we got, with 60 MHz clock and all.

Steering resolution is also as Javierr reversed out.

Cheers,
Magnus

On Tue, 31 Jan 2012 21:19:50 +0000 (UTC)
cfo xnews2@luna.kyed.com wrote:

Thanks! The links worked... Dunno why using the webpage does not...
Maybe some strange interferance with my firefox version and their
javascript stuff...

The STM32F4 looks quite good. Two 32bit timers with multiple channels
(although the documentation does quite a job to hide that fact). Only
downside is that the pins for the 32bit timers are shared with the
ethernet interface, which makes most of them unusable. But i'll have
to have a better look, maybe RMII can free up enough pins for it to
work w/o an FPGA.

Thanks again!

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

On Tue, 31 Jan 2012 15:27:45 -0800
Chris Albertson albertson.chris@gmail.com wrote:

No, the system doesn't have to be small, but i'd like to have ultimate
control over every part of it. Which is kind of hard if you are running
an PC with some OS on it. Yes, i could programm the PC using the same
tools as i use for a uC, but then i'd simply waste a lot of time doing
work that others have done already... not to mention a lot of computing
power.

Beside... a uC system is easier to tinker with than a PC :)
And my main goal is to have a system to experiment with.
That requirement will probably make it more expensive than a Atom Mainboard,
but it will be much more fun and a lot more to learn :)

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

On Wed, 01 Feb 2012 01:52:16 +0100
Magnus Danielson magnus@rubidium.dyndns.org wrote:

I have not seen any such papers yet. Do you have any pointers or hints
what to search for?

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

On Tue, 31 Jan 2012 16:21:40 -0800
"Tom Van Baak" tvb@LeapSecond.com wrote:

I'm planning that... I don't think it's too difficult to do, given
all the information (papers and books) that are available on how
to build GPS receivers. But it will be a damn lot of work.
So don't hold your breath. It will probably take a couple of years
until i even get around to design a board for it, not to mention to
write all the code.

		Attila Kinali

--
The trouble with you, Shev, is you don't say anything until you've saved
up a whole truckload of damned heavy brick arguments and then you dump
them all out and never look at the bleeding body mangled beneath the heap
-- Tirin, The Dispossessed, U. Le Guin

I have opened the FTS125: the fixed OCXO 20MHz is fed using the EXT_CLK pin
7 on the CW25. Maybe it is possible to drive a CW12 with an external high
quality 20MHz but maybe a suitable firmware is then needed.

On Wed, Feb 1, 2012 at 10:28 AM, Attila Kinali attila@kinali.ch wrote:

Attila,

I don't have a link either. I would look at the usual T&F web sources:
PTTI, NIST, FCS, EFTF, ION-GPS, IEEE.

If you can't locate it let me know and I'll ask around to see who might
have done the experiments.

/tvb