Help need a precise frequency source

  • Thread starter Random Electron
  • Start date
R

Random Electron

Guest
Hi,

I need to calibrate my frequency counter. Being a DIYer, I thought of using
the color burst signal from a TV. However, whenever I attach a probe to the
xtal, the color blanks out. Obviously, I've pulled the phase or frequency
from normal.

Any other ideas for a precision cal frequency?

tia
 
"Random Electron" <Random@optonline.net> wrote in message
news:Xns93E98DDEDD07ERandomoptonlinenet@167.206.3.3...
Hi,

I need to calibrate my frequency counter. Being a DIYer, I thought of
using
the color burst signal from a TV. However, whenever I attach a probe to
the
xtal, the color blanks out. Obviously, I've pulled the phase or frequency
from normal.

Any other ideas for a precision cal frequency?
Click to expand...
You can build a crystal controlled oscillator (say at 10 MHz) and zero beat
it with the WWV signal using a receiver or scanner. You can buy stable
oscillators such as http://www.mmdcomp.com/pdf/OS19.pdf

You can use a resistor to keep the probe from killing the TV color burst
oscillator. Lots of ways to go. Depends on how much time and money you
want to spend and how accurate you would like your calibration to be.
 
"Charles Schuler" <charlesschuler@comcast.net> wrote in
news:SESdnWlKfv98PM6iXTWJhQ@comcast.com:

"Random Electron" <Random@optonline.net> wrote in message
news:Xns93E98DDEDD07ERandomoptonlinenet@167.206.3.3...
Hi,

I need to calibrate my frequency counter. Being a DIYer, I thought of
using
the color burst signal from a TV. However, whenever I attach a probe
to
the
xtal, the color blanks out. Obviously, I've pulled the phase or
frequency from normal.

Any other ideas for a precision cal frequency?

You can build a crystal controlled oscillator (say at 10 MHz) and zero
beat it with the WWV signal using a receiver or scanner. You can buy
stable oscillators such as http://www.mmdcomp.com/pdf/OS19.pdf

You can use a resistor to keep the probe from killing the TV color
burst oscillator. Lots of ways to go. Depends on how much time and
money you want to spend and how accurate you would like your
calibration to be.
Click to expand...
The resistor worked fine.

thanks
 
You can probably pick up the horizontal deflection signal by just putting
your probe in the vicinity of the a TV - just not too close! 15,735.... Hz.sg

--- sam | Sci.Electronics.Repair FAQ Home Page: http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Sam's Laser FAQ: http://www.repairfaq.org/sam/lasersam.htm
| Mirror Site Info: http://www.repairfaq.org/REPAIR/F_mirror.html

Important: The email address in this message header may no longer work. To
contact me, please use the Feedback Form at repairfaq.org. Thanks.
 
"Charles Schuler" (charlesschuler@comcast.net) writes:
"Random Electron" <Random@optonline.net> wrote in message
news:Xns93E98DDEDD07ERandomoptonlinenet@167.206.3.3...
Hi,

I need to calibrate my frequency counter. Being a DIYer, I thought of
using
the color burst signal from a TV. However, whenever I attach a probe to
the
xtal, the color blanks out. Obviously, I've pulled the phase or frequency
from normal.

Any other ideas for a precision cal frequency?

You can build a crystal controlled oscillator (say at 10 MHz) and zero beat
it with the WWV signal using a receiver or scanner. You can buy stable
oscillators such as http://www.mmdcomp.com/pdf/OS19.pdf

That's just going to add a degree of error. If someone is lucky,
Click to expand...
their frequency counter timebase (either the actual oscillator or
an output from a divider) is on a subharmonic of a WWV frequency,
so it can directly beat against the time station.

And unless someone does have a good frequency standard, it's likely
silly to adjust the timebase in a counter. Maybe it's off, but you don't
really know where it's going.

Michael
 
On Mon, 01 Sep 2003 20:10:37 GMT Random Electron
<Random@optonline.net> wrote:

"Charles Schuler" <charlesschuler@comcast.net> wrote in
news:SESdnWlKfv98PM6iXTWJhQ@comcast.com:

You can use a resistor to keep the probe from killing the TV color
burst oscillator. Lots of ways to go. Depends on how much time and
money you want to spend and how accurate you would like your
calibration to be.

The resistor worked fine.
Click to expand...
There was a time when the colorburst frequency was well controlled,
using cesium clocks at network headquarters. Those days are long gone
however, and each local station now generates their own colorburst
signal.

These generators aren't bad, but they really aren't a particularly
good standard. Of course you didn't mention how good a standard you
really wanted. They are probably good to one part in 10E6, but the WWV
10MHz signal is good to (I think) about 10E10. It's much better than
that when transmitted, but variations in propagation delay degrade it
for most of us (those who aren't local to NIST.)

You can do even better with WWVB at 60kHz, but that requires long term
measurement to achieve. The 60kHz signal is better only because there
is very little 24 hour variation in the propagation delay at 60kHz.

If you're interested, you can get a very informative booklet on this
subject from the Time & Frequency division at NIST.

-
-----------------------------------------------
Jim Adney jadney@vwtype3.org
Madison, WI 53711 USA
-----------------------------------------------
 
Greetings, if you need a good affordable source go here

http://www.realhamradio.com/GPS_Frequency_Standard.htm

A friend of mine has one of these and uses a distbution amplifier to
send the source to all his equipment. I recently calibrated my HP 5328A
to it :)


Random Electron wrote:

Hi,

I need to calibrate my frequency counter. Being a DIYer, I thought of using
the color burst signal from a TV. However, whenever I attach a probe to the
xtal, the color blanks out. Obviously, I've pulled the phase or frequency
from normal.

Any other ideas for a precision cal frequency?

tia
Click to expand...
 
In article <5s38lvgid9nc6ogmsp0rtd0pdvh143l65j@4ax.com>,
Jim Adney <jadney@vwtype3.org> wrote:


There was a time when the colorburst frequency was well controlled,
using cesium clocks at network headquarters. Those days are long gone
however, and each local station now generates their own colorburst
signal.

The "local" station burst/subcarrier is generally either always
Click to expand...
genlocked to the network or now uses a generator locked to GPS time/freq
reference. Broadcast sync generators don't simply derive their timing
from the power line, but do contain precision, temp-compensated
frequency determining elements, when not locked to a more precise
external source.

[...]

If you're interested, you can get a very informative booklet on this
subject from the Time & Frequency division at NIST.

NIST site at
Click to expand...
http://boulder.nist.gov/timefreq/

PDF info publication
http://boulder.nist.gov/timefreq/general/pdf/1383.pdf
--
chuck reti
detroit mi
 
mike <spamme0@juno.com> wrote in news:3F553FCF.4000503@juno.com:

Sam Goldwasser wrote:
You can probably pick up the horizontal deflection signal by just
putting your probe in the vicinity of the a TV - just not too close!
15,735.... Hz.sg

--- sam | Sci.Electronics.Repair FAQ Home Page:
http://www.repairfaq.org/
Repair | Main Table of Contents: http://www.repairfaq.org/REPAIR/
+Lasers | Sam's Laser FAQ: http://www.repairfaq.org/sam/lasersam.htm
| Mirror Site Info:
| http://www.repairfaq.org/REPAIR/F_mirror.html

Important: The email address in this message header may no longer
work. To contact me, please use the Feedback Form at repairfaq.org.
Thanks.




I'm confused.
Even if the 15,735...were dead on, how long would your gate time have
to be to get 1e6 accuracy? Most counters don't have a gate time that
long...I certainly don't have that much patience ;-)
Ditto for direct measurement of WWVB. You really need something over
10 MHz. to easily tweek a counter.
Click to expand...
No,you use a scope to compare the two signals;the reference and the counter
clock signals,adjust for minimum drift.(trigger on one channel,adjust for
min.drift on the 2nd channel.




--
Jim Yanik,NRA member
jyanik@kua.net
 
<< No,you use a scope to compare the two signals;the reference and the counter
clock signals,adjust for minimum drift.(trigger on one channel,adjust for
min.drift on the 2nd channel. >>

Jim-

This is how I adjust my HP and Fluke counters that have ovens. Another
Optoelectronics counter has a TCXO that is "pulled" when connected to an
external load such as the oscilloscope, so it must be adjusted for accurate
count.

It is often possible to pick up a signal such as a counter time base or the
master oscillator of a PLL Ham Rig, using a probe consisting of a couple turns
of stiff hookup wire, and connecting that to the vertical input of the
oscilloscope. Even though the standard and the master oscillator are not the
same frequency, the pattern will stand still if their ratio is a simple one,
such as 24 MHz compared to a 10 MHz standard.

73, Fred, K4DII
 
fmmck@aol.com (Fred McKenzie) wrote in
news:20030903133106.18133.00000377@mb-m24.aol.com:

No,you use a scope to compare the two signals;the reference and the
counter
clock signals,adjust for minimum drift.(trigger on one channel,adjust
for min.drift on the 2nd channel.

Jim-

This is how I adjust my HP and Fluke counters that have ovens.
Another Optoelectronics counter has a TCXO that is "pulled" when
connected to an external load such as the oscilloscope, so it must be
adjusted for accurate count.
Click to expand...
Have you tried a 10X probe,to reduce loading? Or found a different pick-off
point,from a buffering IC perhaps? Or maybe use the extra resistor
trick,like another 1 meg R on the probe tip?

How do you like the Optoelectronics counter? (What model is it?)
It is often possible to pick up a signal such as a counter time base
or the master oscillator of a PLL Ham Rig, using a probe consisting of
a couple turns of stiff hookup wire, and connecting that to the
vertical input of the oscilloscope. Even though the standard and the
master oscillator are not the same frequency, the pattern will stand
still if their ratio is a simple one, such as 24 MHz compared to a 10
MHz standard.

73, Fred, K4DII
Click to expand...
When measuring/setting TEK sync generators xtal clock freqs(3.579545Mhz),we
were instructed to put the Spectracom 8161 10Mhz reference into the DC509
counter's external clock input,to obtain the best possible accuracy for
setting the freq.
--
Jim Yanik,NRA member
jyanik@kua.net
 
And if you have a 10 MHZ timebase which alot of counters do , you can beat
that Xtal directly against WWV in a short wave receiver. Put a scope on the
detector output to see the really low frequency beats which can't be heard
thru the speaker.
bg
Jim Adney wrote in message <5s38lvgid9nc6ogmsp0rtd0pdvh143l65j@4ax.com>...
On Mon, 01 Sep 2003 20:10:37 GMT Random Electron
Random@optonline.net> wrote:

"Charles Schuler" <charlesschuler@comcast.net> wrote in
news:SESdnWlKfv98PM6iXTWJhQ@comcast.com:

You can use a resistor to keep the probe from killing the TV color
burst oscillator. Lots of ways to go. Depends on how much time and
money you want to spend and how accurate you would like your
calibration to be.

The resistor worked fine.

There was a time when the colorburst frequency was well controlled,
using cesium clocks at network headquarters. Those days are long gone
however, and each local station now generates their own colorburst
signal.

These generators aren't bad, but they really aren't a particularly
good standard. Of course you didn't mention how good a standard you
really wanted. They are probably good to one part in 10E6, but the WWV
10MHz signal is good to (I think) about 10E10. It's much better than
that when transmitted, but variations in propagation delay degrade it
for most of us (those who aren't local to NIST.)

You can do even better with WWVB at 60kHz, but that requires long term
measurement to achieve. The 60kHz signal is better only because there
is very little 24 hour variation in the propagation delay at 60kHz.

If you're interested, you can get a very informative booklet on this
subject from the Time & Frequency division at NIST.

-
-----------------------------------------------
Jim Adney jadney@vwtype3.org
Madison, WI 53711 USA
-----------------------------------------------
Click to expand...
 
On 02 Sep 2003 20:20:39 -0400 Sam Goldwasser <sam@saul.cis.upenn.edu>
wrote:

Jim Adney <jadney@vwtype3.org> writes:

These generators aren't bad, but they really aren't a particularly
good standard. Of course you didn't mention how good a standard you
really wanted. They are probably good to one part in 10E6, but the WWV
10MHz signal is good to (I think) about 10E10. It's much better than
that when transmitted, but variations in propagation delay degrade it
for most of us (those who aren't local to NIST.)

Sorry? :) Variations in delay may introduce phase noise/short term variations
in frequency/jitter but the long term accuracy is not affected.
Click to expand...
If you are willing to integrate over several days, and make sure you
integrate over an integral number of 24 hour periods, you'd be right.
But for shorter periods of time, as in minutes or hours, there are
diurnal changes in the propagation delay due to changes in the
atmosphere which will affect the received frequency. These changes are
small enough that most of us would never notice it, but it's all
carefully published by NIST.

I don't claim to understand all of it, but you can call NIST, as I
did, and they'll send you their booklet which is really QUITE
impressive. The copy I have is marked NIST Special Publication 432
(Revised 1990.) At that time I talked to Roger Buhler at 303-497-3281.

Since I now have the booklet in front of me, here are their figures:

The transmitted accuracy of all signals was about 1 part in 1E11 in
1990; it might well be better now.

WWV & WWVH, received accuracy = ~1E7. This would be unchanged today
becauee the variation in progagation delay is independent of
transmitted accuracy.

WWVB, received accuracy = ~1E11. I don't know if this might have
improved with improved transmitted accuracy. It's possible that with
increased transmitted accuracy they might have run into propagation
delay problems at this wavelength, too.

-
-----------------------------------------------
Jim Adney jadney@vwtype3.org
Madison, WI 53711 USA
-----------------------------------------------
 
On Wed, 3 Sep 2003 14:27:10 +0000 (UTC) Jim Yanik <jyanik@abuse.gov>
wrote:

No,you use a scope to compare the two signals;the reference and the counter
clock signals,adjust for minimum drift.(trigger on one channel,adjust for
min.drift on the 2nd channel.
Click to expand...
To do that, you need 2 signals either of the same frequency or easy
multiples of each other. This would be "hard" to do with 15...kHz and
10MHz.

-
-----------------------------------------------
Jim Adney jadney@vwtype3.org
Madison, WI 53711 USA
-----------------------------------------------
 
Jim Adney <jadney@vwtype3.org> wrote in
news:isddlv406lo0tumucn39019l2ort9vflhi@4ax.com:

On Wed, 3 Sep 2003 14:27:10 +0000 (UTC) Jim Yanik <jyanik@abuse.gov
wrote:

No,you use a scope to compare the two signals;the reference and the
counter clock signals,adjust for minimum drift.(trigger on one
channel,adjust for min.drift on the 2nd channel.

To do that, you need 2 signals either of the same frequency or easy
multiples of each other. This would be "hard" to do with 15...kHz and
10MHz.

-
-----------------------------------------------
Jim Adney jadney@vwtype3.org
Madison, WI 53711 USA
-----------------------------------------------
Click to expand...
I wouldn't use 15.734 Khz for a cal signal for a counter with a 10Mhz
clock.Of course,you could ext.reference a digital synthesized signal source
with WWVB to get a comparison signal near 15Khz,if you really wanted one.
--
Jim Yanik,NRA member
jyanik@kua.net
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top