Data collection cycle -- multiple sources to one counter?
I want to monitor all my atomic devices over the long term against GPS.
Rather than devoting one counter to each source and gathering a lot more
data than is useful, I'm going to use a GPIB-controlled coax switch to
route the 1pps from each source into the counter START input. That way
I can alternately read each source for X minutes against GPS.
My question is whether there is an optimum "X" for the time to dwell on
each source. I've been thinking that 15 minutes might make sense. That
means I'd start with device A, average it against GPS for 15 minutes of
1pps samples, record that to the log file, then switch the input to
device B, dwell there for 15 minutes and record, then to device C for 15
minutes and record, then repeat the cycle.
With three devices, that would give me averages of 900 readings and a
tau of 45 minutes. Seems reasonable, but I suspect a lot of labs do
something similar and I wonder if there's tribal knowledge of the
optimum cycle.
Any thoughts?
John
Hi John:
If you have Time RAIM turned on then using GPS as the start pulse is
good, that way if the RAIM shuts off the GPS no measurement will be
started. Using GPS as the stop pulse would result in an extra long
measurement.
I'm using 1000 seconds per reading on the SR620. I like it in that you
get new readings fairly often so if something changes you don't have to
wait forever. The sigma is almost always in the 9 ns area, which is the
saw tooth size for the M12T+.
I moved an HP 59307A Dual 4-way HP-IB coax switch here to do just what
you describe, but am not happy with the performance of the FTS4060. The
longer you measure each source the fewer relay cycles the switch needs
to do, making it last longer.
When working with the HP 4380S Balanced Network Analyzer system I found
that using flexible cables resulted in an unstable system. My first
improvement was to use 0.141" semi rigid coax (it's much more on the
rigid side) and got a mandrel to bend it. Later I discovered you can
get flexible 0.141" coax that's not "rigid" and it worked great, no
mandrel needed. The point is that you probably will need to have good
control on all the coax lines. The lines will change length with
temperature, so the longer they are the more temperature variation there
will be.
Tom or Doug probably have much more experience in recording over long
time frames.
Have Fun,
Brooke Clarke, N6GCE
w/Java http://www.PRC68.com
w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com
John Ackermann N8UR wrote:
I want to monitor all my atomic devices over the long term against GPS.
Rather than devoting one counter to each source and gathering a lot more
data than is useful, I'm going to use a GPIB-controlled coax switch to
route the 1pps from each source into the counter START input. That way
I can alternately read each source for X minutes against GPS.
My question is whether there is an optimum "X" for the time to dwell on
each source. I've been thinking that 15 minutes might make sense. That
means I'd start with device A, average it against GPS for 15 minutes of
1pps samples, record that to the log file, then switch the input to
device B, dwell there for 15 minutes and record, then to device C for 15
minutes and record, then repeat the cycle.
With three devices, that would give me averages of 900 readings and a
tau of 45 minutes. Seems reasonable, but I suspect a lot of labs do
something similar and I wonder if there's tribal knowledge of the
optimum cycle.
Any thoughts?
John
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
In message 42B4BE4F.5030802@febo.com, John Ackermann N8UR writes:
I want to monitor all my atomic devices over the long term against GPS.
Rather than devoting one counter to each source and gathering a lot more
data than is useful, I'm going to use a GPIB-controlled coax switch to
route the 1pps from each source into the counter START input. That way
I can alternately read each source for X minutes against GPS.
My question is whether there is an optimum "X" for the time to dwell on
each source.
What you really need to look at is how often you revisit each source
so that you don't get roll-over ambiguities.
Switching too fast will also wear out your relays, and that is one thing
to be wary about buying automation multiplexers on eBay: A lot of what
is available is there because relays started being unreliable.
--
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.
Poul-Henning Kamp wrote:
What you really need to look at is how often you revisit each source
so that you don't get roll-over ambiguities.
I hadn't thought of that, though with visits measured in minutes I
shouldn't have problems as long as my starting time difference is a few
microseconds. I've learned that it's very handy to keep one of my
Z3801As running about 10uS fast -- I can use it to sync the 1pps
dividers when I'm measuring against GPS or another on-time source.
Switching too fast will also wear out your relays, and that is one thing
to be wary about buying automation multiplexers on eBay: A lot of what
is available is there because relays started being unreliable.
That's a very good point.
John
Brooke Clarke wrote:
Hi John:
If you have Time RAIM turned on then using GPS as the start pulse is
good, that way if the RAIM shuts off the GPS no measurement will be
started. Using GPS as the stop pulse would result in an extra long
measurement.
Interesting idea -- though I thought that the convention was to put the
reference source on STOP and the unknown on START.
I'm using 1000 seconds per reading on the SR620. I like it in that
you get new readings fairly often so if something changes you don't
have to wait forever. The sigma is almost always in the 9 ns area,
which is the saw tooth size for the M12T+.
Makes sense; my idea of 15 minutes (900 seconds) is close to that. I
guess one consideration is what gives you a convenient tau at the output
-- with three sources, 20 minutes would give you a tau of 1 hour, and
reduce the amount of switching.
I moved an HP 59307A Dual 4-way HP-IB coax switch here to do just what
you describe, but am not happy with the performance of the FTS4060.
The longer you measure each source the fewer relay cycles the switch
needs to do, making it last longer.
That's the same box I'm using. I got a bunch of them, and some of the
matching open-frame relay boxes, from a guy a couple of years ago and
they do come in handy.
When working with the HP 4380S Balanced Network Analyzer system I
found that using flexible cables resulted in an unstable system. My
first improvement was to use 0.141" semi rigid coax (it's much more on
the rigid side) and got a mandrel to bend it. Later I discovered you
can get flexible 0.141" coax that's not "rigid" and it worked great,
no mandrel needed. The point is that you probably will need to have
good control on all the coax lines. The lines will change length with
temperature, so the longer they are the more temperature variation
there will be.
That's a really good point. I don't think it's feasible to use rigid or
semi-rigid cable in the setup I'm doing now, but that certainly would be
the best answer for maximum stability. The good news is that all the
cables are relatively short (6 feet or less).
John
Brooke Clarke wrote:
When working with the HP 4380S Balanced Network Analyzer system I found
that using flexible cables resulted in an unstable system. My first
improvement was to use 0.141" semi rigid coax (it's much more on the
rigid side) and got a mandrel to bend it. Later I discovered you can
get flexible 0.141" coax that's not "rigid" and it worked great, no
mandrel needed. The point is that you probably will need to have good
control on all the coax lines. The lines will change length with
temperature, so the longer they are the more temperature variation there
will be.
If its not moved, is semi-rigid any more stable than other coaxes, if
its not moved?
I don't know all the mechanisms for delays in coax, but the temperature
coefficient of the dielectric constant of the dielectric must be one of
them, and quite possibly the most significant. I would have thought that
dominated any change due to the expansion of the metal. This would
suggest you need to find the optimal dielectric.
I've got a bit of cable here with what is basically an air dielectric,
with the inner conductor supported on a spiral of some sort, which means
99% of the dielectric is air. However, that cable is not practical at
small sizes.
PTFE might for example be a lot better/worst than polyethylene in the
variation of dielectric constant with temperature.
I'm not sure that RG405 semi-rigid (which is what I guess you mean) is
necessarily the best choice.
I used semi rigid a lot, but prefer the bigger one, since its a lot
easier to solder, as my eyses can see it!!
--
David Kirkby,
G8WRB
Please check out http://www.g8wrb.org/
of if you live in Essex http://www.southminster-branch-line.org.uk/
I just purchased, and received, the subject Frequency Electronics
distribution amplifier. It has one input and 12 outputs. The intended
frequency of operation is 5 MHz, but I suspect that any frequency selective
devices will be easy enough to change for operation at 10 MHZ. In fact, it
may work well at 10 as it is. I would know by now had the unit been
complete. It appears that part of the unit housed a battery back-up as well
as a power supply. Well, the power supply module is missing. I rather not
reverse engineer the amp, so, I m basically, after this lengthy post, am
asking if any one on here may have some information on the unit. Any help
greatly appreciated. Talking about isolation between output ports regarding
ground loops, I found it interesting that all 12 output ports are
transformer isolated, and, the output BNCs are not grounded. Actually, the
input and output connectors have three connections. There is the typical BNC
style shield which is common to the chassis ground. However, there is
another concentric smaller shield around the center conductor, and it is
between this shield and the center conductor that the signals are intended
to go. Looks like another project, not what I had in mind. The unit is full
of 2N5109 transistors. That brings back great memories as those were one of
the best medium power transistors that I used to use in the early 70's, when
I could get them. Similar to the common 2N3866 but supposed to work to
higher frequencies and boasted linear operation as well. Was used a lot in
CATV amplifiers. Regards - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
There are quite a few coax switches you can use:
HP 59307A VHF switch
HP 34907A with HP 34905A RF multiplexer module(s)
HP 3488A with 44472A VHF switch plugin(s)
HP 54300 oscilloscope probe mux
VXI system with HP E1472A 50 Ohm Multiplexers
The dwell time will depend on the quality of your
GPS source. I mean, if you're doing really long
term collection, one data point per day is all you
need. But if you have white receiver noise or
sawtooth jitter to remove then make sure your
measurement is long enough to cover that. An
M12 needs less averaging time than a VP, for
example.
But things to watch out for:
I think 15 minutes sounds fine. If you're looking
for exactly 4 readings per hour, on the hour, you
have to take into account the second(s) it will
take to activate the switch and let the line settle.
Maybe you can activate the switch in between
1 PPS pulses and just make it in time. But in
general don't take 900 second readings and
expect you'll get exactly 96 readings over the
86400 s day.
If you are measuring 1 PPS your switch set up
will probably work fine. But if you are measuring
5 MHz RF then watch out for impedance changes.
The other 3 lines will be at no load for 45 minutes
while the one you're measuring is at 50R. Most
frequency standards are affected by changes in
load. One solution is the (more expensive, usually)
VHF switches with internal termination for all inputs
regardless of the state of the switch. The other
solution is to let the frequency settle after a load
change; but this can take minutes. Constant load,
regardless of switch position, is the ideal. Another
idea is to 50R terminate each input with a BNC tee
at the switch and make your measurements at
1M. Check this with a 'scope before you trust it.
Again, all this concern is for RF signals; not to
worry about this with GPS 1 PPS signals.
To preserve relay contact life when using switches
like the 59307A, think gray code instead of binary
code; that is; make your measurements so that
you have only one relay contact change per
measurement. Look at the schematic, or listen
with your ears. You want a switch order like
00 01 11 10 instead of the normal 00 01 10 11.
Gray order is one relay change per measurement
while binary order is 1.5 (average) relay changes
per measurement; so gray gives you 50% more
relay contact life.
Expect some of your surplus switches to have
trashed contacts due to age. I don't know if you
can detect contact wear with a micro ohmmeter
but you might want to just run them through a
self test, using the same ref signal through the
relay mux and see if the 5370 standard deviation
is 25 ps (good contacts) or something much
larger (bad relay contacts) as you repeatedly
open/close them.
Most of my 1 PPS sources are deliberately set
(or allowed to be) many microseconds advanced
or retarded, to insure the phase never drifts across
zero during a measurement run. It's not that it's
a big problem, it just adds some pain to the data
analysis. Start/stop channel choice is arbitrary
and I usually pick them so that I get all positive
numbers rather than negative or mixed sign.
At the level of your test I don't think cable types
matter but at least secure them with tie wraps to
prevent phase changes due to vibration, wiggling,
and the occasional lab visitor. BNC connectors
are not as good as TNC, SMA, or N so I think
some mechanical tie down is advised if you want
clean data for a year.
/tvb
http://www.LeapSecond.com
----- Original Message -----
From: "John Ackermann N8UR" jra@febo.com
To: time-nuts@febo.com
Sent: Saturday, June 18, 2005 17:37
Subject: [time-nuts] Data collection cycle -- multiple sources to
onecounter?
I want to monitor all my atomic devices over the long term against GPS.
Rather than devoting one counter to each source and gathering a lot more
data than is useful, I'm going to use a GPIB-controlled coax switch to
route the 1pps from each source into the counter START input. That way
I can alternately read each source for X minutes against GPS.
My question is whether there is an optimum "X" for the time to dwell on
each source. I've been thinking that 15 minutes might make sense. That
means I'd start with device A, average it against GPS for 15 minutes of
1pps samples, record that to the log file, then switch the input to
device B, dwell there for 15 minutes and record, then to device C for 15
minutes and record, then repeat the cycle.
With three devices, that would give me averages of 900 readings and a
tau of 45 minutes. Seems reasonable, but I suspect a lot of labs do
something similar and I wonder if there's tribal knowledge of the
optimum cycle.
Any thoughts?
John
Tom,
You mentioned the effect of using an unterminated switch, or
even manually switching cables to a counter.
I'd'a thought that a buffer amplifier or emitter follower
would in fact buffer the oscillator from load changes. If
the standard is affected, would an external buffer (like a
line driver) isolate the effects of load changes? Any numbers
on the size of the effect?
How about permanently terminating all cables at the switch
and running a 10" cable to a counter set for high impedance
input?
Bill Hawkins
Bill Hawkins wrote:
Tom,
You mentioned the effect of using an unterminated switch, or
even manually switching cables to a counter.
I'd'a thought that a buffer amplifier or emitter follower
would in fact buffer the oscillator from load changes. If
the standard is affected, would an external buffer (like a
line driver) isolate the effects of load changes? Any numbers
on the size of the effect?
How about permanently terminating all cables at the switch
and running a 10" cable to a counter set for high impedance
input?
Bill Hawkins
An attenuator of X dB on the output of a RF source will cause a return
loss of at least 2 X dB (worst case, assuming 100% reflection), since
any signal has to pass through the attenuator twice.
The attenuator will reduce a 1W source to 250mW (6dB attenuation).
Assuming the load on the output of the attenuator is open or short (so
100% reflection), the power will be reduced by another factor of 4 (to
62.5mW) before it reaches the source. So only 6.25% is reflected back to
the source. That's a VSWR of less than 1.7.
Whether or not you can afford the 6dB loss is another matter, but
attenuators are pretty good at reducing the effect of any change in the
load impedance.
--
David Kirkby,
G8WRB
Please check out http://www.g8wrb.org/
of if you live in Essex http://www.southminster-branch-line.org.uk/
Hi David:
The problem was that you must move the cables some in order to do the
calibration. The semi-rigid coax would spring back to the position
where is was and gave good results, but conventional coax that's
flexible would never go back to where it was. The newer crinkled
surface 141 cable was both flexible and repeatable without the need for
mandrel bending.
Regular coax if not allowed to move would only have temperature variations.
Have Fun,
Brooke
David Kirkby wrote:
Brooke Clarke wrote:
When working with the HP 4380S Balanced Network Analyzer system I
found that using flexible cables resulted in an unstable system. My
first improvement was to use 0.141" semi rigid coax (it's much more on
the rigid side) and got a mandrel to bend it. Later I discovered you
can get flexible 0.141" coax that's not "rigid" and it worked great,
no mandrel needed. The point is that you probably will need to have
good control on all the coax lines. The lines will change length with
temperature, so the longer they are the more temperature variation
there will be.
If its not moved, is semi-rigid any more stable than other coaxes, if
its not moved?
I don't know all the mechanisms for delays in coax, but the temperature
coefficient of the dielectric constant of the dielectric must be one of
them, and quite possibly the most significant. I would have thought that
dominated any change due to the expansion of the metal. This would
suggest you need to find the optimal dielectric.
I've got a bit of cable here with what is basically an air dielectric,
with the inner conductor supported on a spiral of some sort, which means
99% of the dielectric is air. However, that cable is not practical at
small sizes.
PTFE might for example be a lot better/worst than polyethylene in the
variation of dielectric constant with temperature.
I'm not sure that RG405 semi-rigid (which is what I guess you mean) is
necessarily the best choice.
I used semi rigid a lot, but prefer the bigger one, since its a lot
easier to solder, as my eyses can see it!!
In message 42B5AC9A.20305@onetel.net, David Kirkby writes:
The attenuator will reduce a 1W source to 250mW (6dB attenuation).
Assuming the load on the output of the attenuator is open or short (so
100% reflection), the power will be reduced by another factor of 4 (to
62.5mW) before it reaches the source. So only 6.25% is reflected back to
the source. That's a VSWR of less than 1.7.
Be aware that there is a big difference in impact on square and sine
signals.
Even a minor change in circumstances will push the zero-crossing of a
sine wave signal one way or the other. This is because (compared to
a square-wave) the slope at the zero-crossing is very low in terms
of V/s.
Also, if measuring squarewave signals, be sure to choose your trigger
point wisely. Many digital drivers control for the V(il)/V(ih) voltages
and neither is likely to be V(max)/2.
--
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.
John Ackermann N8UR wrote:
I want to monitor all my atomic devices over the long term against GPS.
Rather than devoting one counter to each source and gathering a lot more
data than is useful, I'm going to use a GPIB-controlled coax switch to
route the 1pps from each source into the counter START input. That way
I can alternately read each source for X minutes against GPS.
Just a thought (and probably a stupid one). Is the time delay through a PIN diode switch
http://www.minicircuits.com/npswitch.shtml
sufficiently repeatable to allow such measurements, or would any changes of delay with temperature,
time etc mess up your measurements? Has anyone evaluated PIN diodes for this purpose? They would
certainly get around any worries of wearing them out, as there are no moving parts.
You can buy a 4-way single throw switch with SMA connectors for under $110 which has a minimum of 40
dB (typcally 60dB) isolation at 10MHz.
http://www.minicircuits.com/dg03-214.pdf
--
Dr. David Kirkby PhD CEng MIEE,
Senior Research Fellow,
Department of Medical Physics,
Mallet Place Engineering Building,
Gower St,
University College London,
London WC1E 6BT.
Well, since I have not received any replies to my post, I figured I would
reply to it :). Actually I just want to share some more information and
inquire a bit. I figured out that the unit operates on 24 VDC based on the
front panel lights. So far so good. The problem is the connectors. I do not
believe that in my 50 years of being involved in electronics that I have not
seen any like these before, but, I do not believe I have. Of course my
memory is probably getting worse with the years rather than better. On an
initial look the connectors appear to be BNCs, but they are far from it.
Besides having the attributes I described in my initial post I noticed that
instead of having the two typical bayonet type slots like a BNC, these have
three. So, there are three posts on the sides of the sockets requiring three
slots in the plugs. Has any one ever come across any of these? What are they
even called? I really believe that for what I have in mind the isolation
will be of little consequence, so, more than likely I will just change out
the connectors. Again, another project. The nice thing is there is room in
the unit for a power supply and a Rubidium so it will make a nice lab unit.
Still need to check the frequency response as well. Regards - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Mike Feher
Sent: Sunday, June 19, 2005 9:42 AM
To: 'Discussion of precise time and frequency measurement'
Subject: [time-nuts] FE-70E
I just purchased, and received, the subject Frequency Electronics
distribution amplifier. It has one input and 12 outputs. The intended
frequency of operation is 5 MHz, but I suspect that any frequency selective
devices will be easy enough to change for operation at 10 MHZ. In fact, it
may work well at 10 as it is. I would know by now had the unit been
complete. It appears that part of the unit housed a battery back-up as well
as a power supply. Well, the power supply module is missing. I rather not
reverse engineer the amp, so, I m basically, after this lengthy post, am
asking if any one on here may have some information on the unit. Any help
greatly appreciated. Talking about isolation between output ports regarding
ground loops, I found it interesting that all 12 output ports are
transformer isolated, and, the output BNCs are not grounded. Actually, the
input and output connectors have three connections. There is the typical BNC
style shield which is common to the chassis ground. However, there is
another concentric smaller shield around the center conductor, and it is
between this shield and the center conductor that the signals are intended
to go. Looks like another project, not what I had in mind. The unit is full
of 2N5109 transistors. That brings back great memories as those were one of
the best medium power transistors that I used to use in the early 70's, when
I could get them. Similar to the common 2N3866 but supposed to work to
higher frequencies and boasted linear operation as well. Was used a lot in
CATV amplifiers. Regards - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
At 09:22 AM 6/20/2005 , Mike Feher wrote:
Besides having the attributes I described in my initial post I noticed that
instead of having the two typical bayonet type slots like a BNC, these have
three. So, there are three posts on the sides of the sockets requiring three
slots in the plugs. Has any one ever come across any of these? What are they
even called? I really believe that for what I have in mind the isolation
Sounds like triax. Surprised you couldn't find it in your Pasternack catalog.
How 'bout:
http://www.amphenolrf.com/products/triax.asp
--
newell N5TNL
Mike,
Those are a triaxial variation on the BNC connector. They are fairly
common in the measurement world. A couple of examples that come to
mind are the HP3438A, and the Keithley 616 Electrometer.
They are quite expensive, and tricky to wire up.
-Chuck
Mike Feher wrote:
Well, since I have not received any replies to my post, I figured I would
reply to it :). Actually I just want to share some more information and
inquire a bit. I figured out that the unit operates on 24 VDC based on the
front panel lights. So far so good. The problem is the connectors. I do not
believe that in my 50 years of being involved in electronics that I have not
seen any like these before, but, I do not believe I have. Of course my
memory is probably getting worse with the years rather than better. On an
initial look the connectors appear to be BNCs, but they are far from it.
Besides having the attributes I described in my initial post I noticed that
instead of having the two typical bayonet type slots like a BNC, these have
three. So, there are three posts on the sides of the sockets requiring three
slots in the plugs. Has any one ever come across any of these? What are they
even called? I really believe that for what I have in mind the isolation
will be of little consequence, so, more than likely I will just change out
the connectors. Again, another project. The nice thing is there is room in
the unit for a power supply and a Rubidium so it will make a nice lab unit.
Still need to check the frequency response as well. Regards - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
I thought of "triax" but most of those I have seen were different. I'll look
at the URL you sent along. I get so many Pasternak catalogs that I always
throw them out. Now I wish I had one handy. Regardless, even if I would find
it in there they would probably want a fortune for them. I just looked at
the URL and those could well be it. It does show a BNC with 3 pins 120
degrees apart. Thanks for the lead. Now, does any one have about 13 of them
in their junk box? - Thanks - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Scott Newell
Sent: Monday, June 20, 2005 11:04 AM
To: Discussion of precise time and frequency measurement
Subject: RE: [time-nuts] FE-70E
At 09:22 AM 6/20/2005 , Mike Feher wrote:
Besides having the attributes I described in my initial post I noticed that
instead of having the two typical bayonet type slots like a BNC, these have
three. So, there are three posts on the sides of the sockets requiring
three
slots in the plugs. Has any one ever come across any of these? What are
they
even called? I really believe that for what I have in mind the isolation
Sounds like triax. Surprised you couldn't find it in your Pasternack
catalog.
How 'bout:
http://www.amphenolrf.com/products/triax.asp
--
newell N5TNL
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
At 10:18 AM 6/20/2005 , Mike Feher wrote:
I thought of "triax" but most of those I have seen were different. I'll look
at the URL you sent along. I get so many Pasternak catalogs that I always
throw them out. Now I wish I had one handy. Regardless, even if I would find
Give it a week, and I'm sure you'll get another in the mail. ;-)
it in there they would probably want a fortune for them. I just looked at
the URL and those could well be it. It does show a BNC with 3 pins 120
degrees apart. Thanks for the lead. Now, does any one have about 13 of them
in their junk box? - Thanks - Mike
Might be less trouble and expense to convert the box to BNC.
--
newell N5TNL
I agree, I think I am just going to change out the connectors. Thanks again
& 73 - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Scott Newell
Sent: Monday, June 20, 2005 11:43 AM
To: Discussion of precise time and frequency measurement
Subject: RE: [time-nuts] FE-70E
At 10:18 AM 6/20/2005 , Mike Feher wrote:
I thought of "triax" but most of those I have seen were different. I'll
look
at the URL you sent along. I get so many Pasternak catalogs that I always
throw them out. Now I wish I had one handy. Regardless, even if I would
find
Give it a week, and I'm sure you'll get another in the mail. ;-)
it in there they would probably want a fortune for them. I just looked at
the URL and those could well be it. It does show a BNC with 3 pins 120
degrees apart. Thanks for the lead. Now, does any one have about 13 of them
in their junk box? - Thanks - Mike
Might be less trouble and expense to convert the box to BNC.
--
newell N5TNL
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Telco equipment has a BNC shell over a two pin connector
for shielded twisted pair. The connector is keyed by
different heights for the two pins. I haven't found it
because I don't know what to call it.
Pasternak?
Bill Hawkins
At 11:45 AM 6/20/2005 , Bill Hawkins wrote:
Telco equipment has a BNC shell over a two pin connector
for shielded twisted pair. The connector is keyed by
different heights for the two pins. I haven't found it
because I don't know what to call it.
Twinax?
--
newell N5TNL
Bill -
Yes, I have seen and am familiar with the connectors you mentioned. Maybe
after seeing the photos I posted earlier you will agree that these are
pretty unique. Pasternak is a stateside company dealing in connectors and
cable assemblies and for some reason feel the need to inundate us with their
small, but thick, catalog on a fairly regular basis. As suggested by Scott,
chances are good that they would have the connectors I need, but, I doubt if
the price would be attractive or affordable, unless I had no other
alternative. Regards - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Bill Hawkins
Sent: Monday, June 20, 2005 12:45 PM
To: Discussion of precise time and frequency measurement
Subject: [time-nuts] Strange connectors
Telco equipment has a BNC shell over a two pin connector
for shielded twisted pair. The connector is keyed by
different heights for the two pins. I haven't found it
because I don't know what to call it.
Pasternak?
Bill Hawkins
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
In message: 00f001c575b7$6cec0800$0500a8c0@darius.domain.actdsltmp
"Bill Hawkins" bill@iaxs.net writes:
: Telco equipment has a BNC shell over a two pin connector
: for shielded twisted pair. The connector is keyed by
: different heights for the two pins. I haven't found it
: because I don't know what to call it.
:
: Pasternak?
Twinax?
Warner
Twin-ax BNC. Trompeter carries them, last I heard.
Dave
----- Original Message -----
From: "Bill Hawkins" bill@iaxs.net
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Sent: Monday, 20 June, 2005 9:45 AM
Subject: [time-nuts] Strange connectors
Telco equipment has a BNC shell over a two pin connector
for shielded twisted pair. The connector is keyed by
different heights for the two pins. I haven't found it
because I don't know what to call it.
Pasternak?
Bill Hawkins
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Turns out that twinax is the IBM token ring cable and connector
that screws in.
But the hint led to twin BNC as Amphenol 31-224 and Pomona.
The Pomona description is a PDF at the link below:
http://www.tequipment.net/pdf/Pomona/4957.pdf
Many thanks for your help.
Bill Hawkins
In message 00f501c575d0$77ce8e00$0500a8c0@darius.domain.actdsltmp, "Bill Hawk
ins" writes:
Turns out that twinax is the IBM token ring cable and connector
that screws in.
Actually, that's not the case.
"Twinax" was the shielded twisted pair 5250 terminal bus for Series/3
(System 34, 36 & 38) "MiniMainFrames" from IBM.
The original IBM connector was more or less N-style.
Shielded twisted pair as a concept has been known for a lot longer
however the oldest use I know off was for microphones in the pre
WWII time frame.
--
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.
I don't think that that is what you want. The twin
BNC has a end view that looks sort of like this:
((O|o))
Where the "O" is a female pin, and the "o" is a male pin.
The teflon center conductor is stepped, so that it encloses
the female pin, but leaves the male pin bare.
The other identifying feature is it has a two pin bayonnette.
-Chuck
Bill Hawkins wrote:
Turns out that twinax is the IBM token ring cable and connector
that screws in.
But the hint led to twin BNC as Amphenol 31-224 and Pomona.
The Pomona description is a PDF at the link below:
http://www.tequipment.net/pdf/Pomona/4957.pdf
Many thanks for your help.
Bill Hawkins
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Ah, yes, incompetence explains it. I'd never seen twinax
connectors in use, but there's a twinax antenna connector
on the back of my 1955 R-390 receiver that fits an IBM
connector.
Twin BNC has one each male and female pin with a split
level insulator that exposes the male pin (hope this
gets through your spam checker).
Bill Hawkins
-----Original Message-----
From: phk@critter.freebsd.dk [mailto:phk@critter.freebsd.dk]On Behalf Of
Poul-Henning Kamp
Sent: Monday, June 20, 2005 2:59 PM
To: bill@iaxs.net; Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Strange connectors
In message 00f501c575d0$77ce8e00$0500a8c0@darius.domain.actdsltmp, "Bill
Hawk
ins" writes:
Turns out that twinax is the IBM token ring cable and connector
that screws in.
Actually, that's not the case.
"Twinax" was the shielded twisted pair 5250 terminal bus for Series/3
(System 34, 36 & 38) "MiniMainFrames" from IBM.
The original IBM connector was more or less N-style.
Shielded twisted pair as a concept has been known for a lot longer
however the oldest use I know off was for microphones in the pre
WWII time frame.
--
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.
In message 000201c575d8$0ee1ec40$0500a8c0@darius.domain.actdsltmp, "Bill Hawk
ins" writes:
Twin BNC has one each male and female pin with a split
level insulator that exposes the male pin (hope this
gets through your spam checker).
Right, that's because twin BNC is polirized, Twinax wasn't.
I even managed to find a picture of a twinax on google:
http://www.provantage.com/FullSize/HW_43231.JPG
--
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.
From: "Dr. David Kirkby" drkirkby@medphys.ucl.ac.uk
Subject: Re: [time-nuts] Data collection cycle -- multiple sources to one counter?
Date: Mon, 20 Jun 2005 13:31:35 +0100
Message-ID: 42B6B727.7000207@medphys.ucl.ac.uk
John Ackermann N8UR wrote:
I want to monitor all my atomic devices over the long term against GPS.
Rather than devoting one counter to each source and gathering a lot more
data than is useful, I'm going to use a GPIB-controlled coax switch to
route the 1pps from each source into the counter START input. That way
I can alternately read each source for X minutes against GPS.
Just a thought (and probably a stupid one). Is the time delay through a PIN
diode switch
http://www.minicircuits.com/npswitch.shtml
sufficiently repeatable to allow such measurements, or would any changes of
delay with temperature, time etc mess up your measurements? Has anyone
evaluated PIN diodes for this purpose? They would certainly get around any
worries of wearing them out, as there are no moving parts.
It should be possible to use some form of solid state switching, but
dependencies to environmental aspects needs to be understood.
One of the things you do want is good isolation from the other input signals,
so the 5 MHz sine from one clock doesn't leak into the 5 MHz sine of another
clock and as these beat against each other (as they tend to do) the beating
appears as a component on the two clocks measurments. The signal-to-noise needs
to be low enought.
Dependencies on junction temperature, rail voltage (control current) and
signal level all comes into play. It would be fund to investigate such a
thing. Hmm... a bunch of 1N4148 should be interesting to test (just for fun).
Should be a good task for the network analyser to chew on. ;O)
Please note that the Mini-Circuits ones have DC-blockers and is listed from
10 MHz and upwards to whatever limit each device is listed for. If you look for
phase-stability you would want the -3 dB corners to be as far away from your
signal as possible as a rule of thumb, we know the phase changes alot at the
corners, but can also change alot in the pass-band. For 1 PPS signals DC levels
should be passed too.
You can buy a 4-way single throw switch with SMA connectors for under $110
which has a minimum of 40 dB (typcally 60dB) isolation at 10MHz.
Yes, but can we do better (and cheaper)?
Cheers,
Magnus
At 04:44 PM 6/20/2005, Poul-Henning Kamp wrote...
Right, that's because twin BNC is polirized, Twinax wasn't.
Twinax is polarized by a keyway in the shell.
Instead of minicircuits I would be looking for video-cross-bar chips.
Maxim as a bunch:
http://para.maxim-ic.com/compare.asp?Fam=Vid_Mux_Buff&Tree=SwitchMux&HP=SwitchMux.cfm&ln=
A MAX458 8x4 switch would be able to drive two counters which could
measure any two of eight sources against each other.
http://pdfserv.maxim-ic.com/en/ds/MAX458-MAX459.pdf
There is even an eval board which does most of the job already:
http://pdfserv.maxim-ic.com/en/ds/MAX459EVKIT.pdf
As far as I can tell you could drive it from a parallel port
with bit-banging.
--
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.
In message 75587.1119301272@critter.freebsd.dk, "Poul-Henning Kamp" writes:
There is even an eval board which does most of the job already:
http://pdfserv.maxim-ic.com/en/ds/MAX459EVKIT.pdf
As far as I can tell you could drive it from a parallel port
with bit-banging.
Forgot to add: $72 list price directly from maxim. (Expect some
shipping cost too.)
--
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.
Poul-Henning Kamp wrote:
Instead of minicircuits I would be looking for video-cross-bar chips.
Any reason to prefer them over the pin diode or GaAs switches from
Minicircuits? I must admit to having no idea how they would compare here.
Maxim as a bunch:
http://para.maxim-ic.com/compare.asp?Fam=Vid_Mux_Buff&Tree=SwitchMux&HP=SwitchMux.cfm&ln=
A MAX458 8x4 switch would be able to drive two counters which could
measure any two of eight sources against each other.
http://pdfserv.maxim-ic.com/en/ds/MAX458-MAX459.pdf
There is even an eval board which does most of the job already:
http://pdfserv.maxim-ic.com/en/ds/MAX459EVKIT.pdf
As far as I can tell you could drive it from a parallel port
with bit-banging.
One thing to note is that Maxim have a very nice policy on samples,
which effectively means for experimental purposes, the chips are free.
Just register on their web site, tick what chips you want (limit of
about 6 or 8, no more than two of the same type) and they courier them
to the next day. If you need more than a couple of one type, pick a
standard and an extended temperature range model, and perhaps a mil spec
one too. Or order some more the next day.
I don't think you can get evaluation boards for nothing, but I have not
actually tried.
In contrast, I've always had to pay for Mincircuits bits, but do happen
to have one of their PIN diode switches sitting around.
--
David Kirkby,
G8WRB
Please check out http://www.g8wrb.org/
of if you live in Essex http://www.southminster-branch-line.org.uk/
HI David:
A PIN diode is great for switching RF frequencies above the diode's
lifetime. But not so good at switching DC coupled signals.
The FET type switches, like the MAX458 are designed for switching DC
coupled signals.
Have Fun,
Brooke Clarke, N6GCE
w/Java http://www.PRC68.com
w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com
David Kirkby wrote:
Poul-Henning Kamp wrote:
Instead of minicircuits I would be looking for video-cross-bar chips.
Any reason to prefer them over the pin diode or GaAs switches from
Minicircuits? I must admit to having no idea how they would compare here.
Maxim as a bunch:
http://para.maxim-ic.com/compare.asp?Fam=Vid_Mux_Buff&Tree=SwitchMux&HP=SwitchMux.cfm&ln=
A MAX458 8x4 switch would be able to drive two counters which could
measure any two of eight sources against each other.
http://pdfserv.maxim-ic.com/en/ds/MAX458-MAX459.pdf
There is even an eval board which does most of the job already:
http://pdfserv.maxim-ic.com/en/ds/MAX459EVKIT.pdf
As far as I can tell you could drive it from a parallel port
with bit-banging.
One thing to note is that Maxim have a very nice policy on samples,
which effectively means for experimental purposes, the chips are free.
Just register on their web site, tick what chips you want (limit of
about 6 or 8, no more than two of the same type) and they courier them
to the next day. If you need more than a couple of one type, pick a
standard and an extended temperature range model, and perhaps a mil spec
one too. Or order some more the next day.
I don't think you can get evaluation boards for nothing, but I have not
actually tried.
In contrast, I've always had to pay for Mincircuits bits, but do happen
to have one of their PIN diode switches sitting around.
MEMS switches are another possibility. They are still maturing but a few
companies are actually selling devices and evaluation units. About a year or
so ago I was looking into the feasibility of a 256 by 256 IF switching
matrix, and MEMS was one of the possibilities. It became pretty obvious
early on that it would just not really work out as I needed to switch up to
2 GHz with fairly high isolation. MEMS is however slowly is evolving into a
technology that may be mature enough and cost effective enough in the near
future (another 10 years). - Mike
Mike B. Feher, N4FS
89 Arnold Blvd.
Howell, NJ, 07731
732-886-5960
-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Brooke Clarke
Sent: Monday, June 20, 2005 6:49 PM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Data collection cycle -- multiple sources to
onecounter?
HI David:
A PIN diode is great for switching RF frequencies above the diode's
lifetime. But not so good at switching DC coupled signals.
The FET type switches, like the MAX458 are designed for switching DC
coupled signals.
Have Fun,
Brooke Clarke, N6GCE
w/Java http://www.PRC68.com
w/o Java http://www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com
David Kirkby wrote:
Poul-Henning Kamp wrote:
Instead of minicircuits I would be looking for video-cross-bar chips.
Any reason to prefer them over the pin diode or GaAs switches from
Minicircuits? I must admit to having no idea how they would compare here.
Maxim as a bunch:
A MAX458 8x4 switch would be able to drive two counters which could
measure any two of eight sources against each other.
http://pdfserv.maxim-ic.com/en/ds/MAX458-MAX459.pdf
There is even an eval board which does most of the job already:
http://pdfserv.maxim-ic.com/en/ds/MAX459EVKIT.pdf
As far as I can tell you could drive it from a parallel port
with bit-banging.
One thing to note is that Maxim have a very nice policy on samples,
which effectively means for experimental purposes, the chips are free.
Just register on their web site, tick what chips you want (limit of
about 6 or 8, no more than two of the same type) and they courier them
to the next day. If you need more than a couple of one type, pick a
standard and an extended temperature range model, and perhaps a mil spec
one too. Or order some more the next day.
I don't think you can get evaluation boards for nothing, but I have not
actually tried.
In contrast, I've always had to pay for Mincircuits bits, but do happen
to have one of their PIN diode switches sitting around.
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
Brooke Clarke wrote:
HI David:
A PIN diode is great for switching RF frequencies above the diode's
lifetime. But not so good at switching DC coupled signals.
The FET type switches, like the MAX458 are designed for switching DC
coupled signals.
Have Fun,
Brooke Clarke, N6GCE
Minicicurits have GaAs switches too, which are DC coupled. Some of them have an minimum isolation of
80 dB (100dB typical) in the range DC-10MHz, but work to 5GHz. But they only have SPDT. The SP4T ones
are DC-3 GHz and have a somewhat lower isolation. To get more than 4 positions, you would need to
combine them.
I would have thought something that works to a high frequency (3~5GHz), must have a small area, and
so the delays repeatable. The MAX458 is speced to only 100MHz, and has internal buffers, so my guess
is that it would be less repeatable than a GaAs switch that works from DC to 3GHz. I'm only basing
this on eductated gueses, not practical measurements.
Many of the High Electron Mobility Transistors (or whatever you want to call them, as different
manufacturers call them different things), have maximum frequencies of operation well in excess of
20GHz. These are tiny so its hard to see how you could get much variation of time delay through the
channel. However, since they have high gain from DC to light, they can be difficult to keep stable in
low frequency applications.
PS,
I don't have shares in Minicircuits.
--
Dr. David Kirkby PhD CEng MIEE,
Senior Research Fellow,
Department of Medical Physics,
Mallet Place Engineering Building,
Gower St,
University College London,
London WC1E 6BT.
Brooke Clarke wrote:
A PIN diode is great for switching RF frequencies above the diode's
lifetime. But not so good at switching DC coupled signals.
The FET type switches, like the MAX458 are designed for switching DC
coupled signals.
Does anybody know what the HP 59307A uses ? It can switch from DC to 500
MHz and the measured DC resistance is on the order of 0.2 ohm. Looking
into the unit reveals that the switching device is shaped like a fat IC,
thermally coupled with the back panel.
73 Alberto I2PHD
Alberto di Bene wrote:
Does anybody know what the HP 59307A uses ? It can switch from DC to 500
MHz and the measured DC resistance is on the order of 0.2 ohm. Looking
into the unit reveals that the switching device is shaped like a fat IC,
thermally coupled with the back panel.
73 Alberto I2PHD
Sounds like a dip packaged reed relay.
-Chuck
Chuck Harris wrote:
Alberto di Bene wrote:
Does anybody know what the HP 59307A uses ? It can switch from DC to 500
MHz and the measured DC resistance is on the order of 0.2 ohm. Looking
into the unit reveals that the switching device is shaped like a fat IC,
thermally coupled with the back panel.
73 Alberto I2PHD
Sounds like a dip packaged reed relay.
-Chuck
It could be, but does it exist a reed relay with 4 positions ? The two
switches in the 59307A are single pole, 4-way each.
73 Alberto I2PHD