[time-nuts] Building a GPSDO & trouble using Jupiter-T

Tom Curlee tcurlee at sbcglobal.net
Sun Jan 29 07:36:49 UTC 2012


Ray:

I used a similar circuit to lock my 10 GHz LO to GPS.  I used a similar Jupiter GPS board receiver (not the 'T' version).  My 10 GHz LO is one of the older Microwave Associates brick type of oscillators.  It has an internal 106.5 crystal oscillator that is in a crude oven.  The 106.5 is multiplied by 96 to get the 10.224 GHz LO frequency.  It would hold frequency fairly reasonably in my lab, but outside in the summer it would drift as much as 30 kHz.  I installed a varactor diode on the quartz oscillator to control the frequency.  I divided the 106.5 MHz down to 10 kHz and compared it with the 10 kHz out of the Jupiter in an exclusive OR gate.  I used a one op amp filter circuit that had a time constant of 5 to 10 seconds - not critical, but just needs to be slow.  If the GPS was 'warmed up' (it had been on in the last few hours), the 10 GHz LO will  lock to within a few Hz of the 10.224 GHz in less than a minute after power on.  I have some
 other frequency stability issues that I need to fix, but the GPS lock portion of the circuit works great.  

Look on eBay for GPS antennas.  I use one of the small mag mount units.  You can get one for around $6.00, shipping included.  Also look on www.halted.com and search for GPS antenna.  They have some (that I use) for $3.95, but charge $5.00 handling  for under $20 - 1 is $8.95, 2 is $12.90, etc

Let me know if I can help in any way.

73,

Tom  WB6UZZ

--- On Sat, 1/28/12, Ray Xu <rayxu123 at gmail.com> wrote:

From: Ray Xu <rayxu123 at gmail.com>
Subject: [time-nuts] Building a GPSDO & trouble using Jupiter-T
To: time-nuts at febo.com
Date: Saturday, January 28, 2012, 10:50 PM

Hi guys

I'm planning to build a GPSDO to use as a frequency ref for my GHz ventures.

I've done research in what other people have built - but I have no
experience working with long-term precision/stability products (nor do I
have the equipment to do so -- I think).

I'm using the Jupiter T (the one with the 10-pin header) GPS and its 10KHz
output in an analog PLL that controls a 10MHz VCXO.  It would be ideal for
me that if I were to multiply the 10MHz output up to 10GHz, I would get
about a few Hertz or so of inaccuracy.  It would also be ideal if I can
have a PLL lock time of a few minutes while maintaining accuracy.

However, right now as I am trying to design my project, I can't find enough
information on the web regarding the VCXO's (or the PLL oscillator, in this
case) short term accuracy effect on output frequency.  I know that its
short term accuracy depends on the "response time" of the PLL, which also
depends on the amount of jitter from the Jupiter-T's 10KHz output...  I
know that the longer the time constant for the PLL, the better accuracy,
but I do not want to wait, literally, hours for it to lock...

Also, what is the advantage of using a OCXO instead of a VCXO in terms of
short-term accuracy?  If the PLL time constant is only a few seconds, then
a crystal shouldn't deviate in frequency by too much within a few seconds,
assuming I'm using a crystal bought from a well-known manufacturer...or
could it? I am inclined towards using oscillators that do not require any
significant warm up time...

For those who have experience using the Jupiter T GPS:
I have bought this
http://www.ebay.com/itm/Navman-jupiter-T-Tu60-GPS-Kit-1pps-10khz-GPS-Module-/260790984470but
I could not get it to communicate via serial.  So far, I do not have
an
antenna available and so the antenna port is just left disconnected.  When
I turn it on, there is a 1pps and 10KHz signal and the TX line is at logic
high.  However, I cannot get it to communicate to anything else, even a
dumb terminal.  It does not respond to when I send a message ID to it.
(Because of computer difficulties, I am using the PicKit 2's UART tool).
Is my device a dud or am I doing something wrong?

Many inputs is highly appreciated.  Thanks

-- 
__________
73, Ray Xu
KF5LJO
_______________________________________________
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.


More information about the time-nuts mailing list