Re: [time-nuts] WTB: GPSDO
So it seems the threat of replacement and a bunch of plugging and
replugging into different computers has spurred my recalcitrant GPSDO
back into life and it is now producing output again... This leaves me
with the original problem of low sensitivity but at least I can solve
that problem by using the "replacement GPSDO" money towards connecting
a better antenna.
I have a Symmetricom 58532A GPS antenna which has a N female connector
but my 3 current GPS receivers all have SMA female connectors. If I
want to provide capacity for at least 4 receivers fed from the same
antenna, I was wondering what the best option for a splitter and where
to do the N-to-SMA conversion. I currently have a Mini-Circuits
ZAP3PD-2 power splitter which does SMA input to 3 SMA outputs. This
seems to work but the connected devices all complain of an Antenna
short, which doesn't seem good. The other popular option seems to be
the Symmetricom 58536A 1x4 splitter which would then require 3-4
N-to-SMA cables - it looks like although this has gain, it seems to be
more of "eliminating loss" than straight gain so would presumably not
overpower the receivers' frontends. Or maybe there is another more
suitable SMA splitter in the Mini Circuits confusingly extensive
catalog ?
Regarding hand-correcting oscillators, that's not really what I am
personally in this for. I am interested in the data acquisition and
analysis parts (must be the astronomy background) and so I am wanting
to use the time-nuts part to learn about how to interface
microcontrollers and ADCs and DACs and would like to measure things at
the multi-day taus. (I'd also like to be able to eventually build a
programmable delay line either to do the sawtooth correction or
perform clock steering to generate and compare my own UTC(TAL), along
with some means of displaying multiple timescales, which may involve
another construction project as available ones either seem to have NTP
servers built in or want IRIG-B or cost a fortune (unless I'm just
googling for the wrong thing)
Thanks,
Tim
Hi Tim,
On Tue, Mar 21, 2017 at 06:36:51PM -0700, Tim Lister wrote:
[...]
I have a Symmetricom 58532A GPS antenna which has a N female connector
but my 3 current GPS receivers all have SMA female connectors. If I
want to provide capacity for at least 4 receivers fed from the same
antenna, I was wondering what the best option for a splitter and where
to do the N-to-SMA conversion. I currently have a Mini-Circuits
ZAP3PD-2 power splitter which does SMA input to 3 SMA outputs. This
seems to work but the connected devices all complain of an Antenna
short, which doesn't seem good. The other popular option seems to be
the Symmetricom 58536A 1x4 splitter which would then require 3-4
N-to-SMA cables - it looks like although this has gain, it seems to be
more of "eliminating loss" than straight gain so would presumably not
overpower the receivers' frontends. Or maybe there is another more
suitable SMA splitter in the Mini Circuits confusingly extensive
catalog ?
Your splitter looks ok, but it has DC pass through on all ports. I would
recommend to put DC-blocks on all but one. The problem is all your
receiver want to power up the antenna and deliver around 3 to 5 V supply
Voltage on the port, your splitter is short-circuit them, which is not
good, especially if the voltage does not match. (The minicircuits
datasheet for the ZA3PD-2+, I guess that is the one you meant, says
RF+DC on all ports) The short-circuit might damage your receiver.
If you take a splitter without DC pass through, you will need an
additional bias-T to feed some supply voltage again.
The special GNSS splitters forward DC only from one port and provide
some 200 Ohm DC termination together with the DC-block on the other
ports to keep the receiver happy, i.e., it seems current consumption and
therefore believes the antenna is ok. Some have an amplifier to
compensate for the loss of the splitter, but it is not necessary if the
antenna delivers enough.
So the GNSS splitter is electrically, a splitter like you have, a
separate LNA in front and all but one port with a bias-T (in reverse,
this provides the DC-block and ensures the signal is not 200 Ohm
terminated) where the DC connector has 200 Ohm termination.
I personally use a NARDA 4372A-4 (10 Euros on Ebay) with three DC-blocks
from minicircuits to split the GNSS signals and a good outdoor antenna.
Best,
Thomas, DK6KD/SA6CID
On Mar 22, 2017, at 12:40 AM, Thomas Petig thomas@petig.eu wrote:
The special GNSS splitters forward DC only from one port and provide
some 200 Ohm DC termination together with the DC-block on the other
ports to keep the receiver happy, i.e., it seems current consumption and
therefore believes the antenna is ok. Some have an amplifier to
compensate for the loss of the splitter, but it is not necessary if the
antenna delivers enough.
Just a little side query… For those using one of these sorts of splitters with a Thunderbolt, have you seen anything odd?
My thunderbolt insists on being on the DC Pass port. If you put it on a DC Block port (yes, something else is on the DC pass port and supplying DC for the antenna and splitter at the time - other connected devices work just fine) it’s deaf. I’ve worked around it by giving the pass port to the TBolt and running everything else on a block port, so it’s not a deal-breaker or anything, but it’s a bit awkward and otherwise unexplainable.
On Wed, March 22, 2017 3:52 pm, Nick Sayer via time-nuts wrote:
My thunderbolt insists on being on the DC Pass port. If you put it on a
DC Block port (yes, something else is on the DC pass port and supplying
DC for the antenna and splitter at the time - other connected devices work
just fine) it's deaf.
You need a DC load so that the GPS receiver thinks there is an active
antenna attached. I think that is just a quirk of the Thunderbolts, that
rather than just flagging an alarm and continuing to run, it gives up and
won't even try to run.
Chris Caudle
Hi,
On Wed, Mar 22, 2017 at 05:20:31PM -0500, Chris Caudle wrote:
On Wed, March 22, 2017 3:52 pm, Nick Sayer via time-nuts wrote:
My thunderbolt insists on being on the DC Pass port. If you put it on a
DC Block port (yes, something else is on the DC pass port and supplying
DC for the antenna and splitter at the time - other connected devices work
just fine) it's deaf.
You need a DC load so that the GPS receiver thinks there is an active
antenna attached. I think that is just a quirk of the Thunderbolts, that
rather than just flagging an alarm and continuing to run, it gives up and
won't even try to run.
They try to check if an active antenna is connected by measuring the DC
load. You can (not tested by me) simulate this with a Bias T and a
resistor to ground on the DC port. I think there are more receiver that
might require this. Many dedicated GPS splitters have such a DC load
'simulation' with around 200 Ohm to ground to keep the receiver happy.
Best,
Thomas DK6KD/SA6CID
--
Chris Caudle
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.
The splitters I’m using do have a 200Ω load on them. I know this because PA6H modules recognize an external antenna and use it.
On Mar 22, 2017, at 3:20 PM, Chris Caudle chris@chriscaudle.org wrote:
On Wed, March 22, 2017 3:52 pm, Nick Sayer via time-nuts wrote:
My thunderbolt insists on being on the DC Pass port. If you put it on a
DC Block port (yes, something else is on the DC pass port and supplying
DC for the antenna and splitter at the time - other connected devices work
just fine) it's deaf.
You need a DC load so that the GPS receiver thinks there is an active
antenna attached. I think that is just a quirk of the Thunderbolts, that
rather than just flagging an alarm and continuing to run, it gives up and
won't even try to run.
Chris Caudle
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.
On Thu, March 23, 2017 9:36 am, Nick Sayer via time-nuts wrote:
The splitters Iâm using do have a 200Ω load on them. I know this
because PA6H modules recognize an external antenna and use it.
I may have been thinking of a different receiver, according to the
Thunderbolt manual I just checked, the antenna alarm should not "impair
the operation" of the thunderbolt:
Bit 1: When 1, indicates that the antenna input connection is open. More
precisely, this bit indicates that the antenna input is not drawing
sufficient current. Normally, the ThunderBolt provides power to the
antennas LNA (Low Noise Amplifier) through the center conductor of the
antenna cable. On-board circuitry senses this current draw, and if low,
this condition will be indicated. However, when the antenna is powered
elsewhere (e.g. when using a splitter) then an antenna open condition is
expected and does not imply a fault nor does it impair the operation of
the ThunderBolt.
--
Chris Caudle
At 03:52 PM 3/22/2017, Nick Sayer via time-nuts wrote:
Just a little side query
For those using one of
these sorts of splitters with a Thunderbolt, have you seen anything odd?
My thunderbolt insists on being on the DC Pass port. If you put it on a DC
Mine doesn't. I just swapped ports around so that
the tbolt isn't powering the antenna, and it's
still tracking, but it is now complaining about an open antenna.
My splitter is a 4 way DirecTV with DC pass on one port. F connectors FTW.
(Before I had an outdoor timing antenna, I used
the TAPR Motorola ANT1A. It must not pull enough
current, because my tbolt would go in and out of antenna open alarm.)
--
newell N5TNL