Re: some Nano VNA SAVER 40m antenna graphs
Excellent work, Barry. I've always contended introduction of the NANOVNAs
would be an outstanding tool for learning. You've proved that. The
universities ought to be buying these in the 100s to teach what you have
learned.
Dave - W?LEV
toggle quoted message
Show quoted text
On Thu, Aug 24, 2023 at 12:37?PM Barry K3EUI <k3euibarry@...> wrote: What I found reassuring is that the four graphs (on all three antennas)
all agree with each other!
The RETURN LOSS (dB) is sharp (yes, I plot it as a negative value) showing
the reflections are least (greatest return loss) where the SWR is lowest,
and where the PHASE graph crosses ZERO degrees, and also where the Smith
Chart is closest to the Bull¡¯s Eye.
Now ¡°matching¡± makes more sense to me.
Any one graph predicts the others.
I passed all of these ham radio tests (teenager) without really
understaning any of these graphs.
Now (I think) I understand the relationships.
the Nano VNA forced me to really learn what these concepts are all about.
TU VNA SAVER author
de k3eui barry
-- *Dave - W?LEV*
--
Dave - W?LEV
|
Donald,
I'm not sure I can get the resolution necessary from the "far" end. The cable is 100m long and the fault is about 240mm from the "near" end.
Geoff
toggle quoted message
Show quoted text
On 8/23/2023 4:07 PM, Donald S Brant Jr wrote: On Wed, Aug 23, 2023 at 04:52 PM, Geoff Peters - AB6BT wrote:
So the question is...can I trust that number?
Measure from the other end and see if the two results add up to the correct total length. If they do you can at least have better confidence. If they doi not you know there is a problem.
73, Don N2VGU
|
Re: some Nano VNA SAVER 40m antenna graphs
What I found reassuring is that the four graphs (on all three antennas) all agree with each other!
The RETURN LOSS (dB) is sharp (yes, I plot it as a negative value) showing the reflections are least (greatest return loss) where the SWR is lowest, and where the PHASE graph crosses ZERO degrees, and also where the Smith Chart is closest to the Bull¡¯s Eye.
Now ¡°matching¡± makes more sense to me.
Any one graph predicts the others.
I passed all of these ham radio tests (teenager) without really understaning any of these graphs.
Now (I think) I understand the relationships.
the Nano VNA forced me to really learn what these concepts are all about.
TU VNA SAVER author
de k3eui barry
|
Bob,
Thanks for that.
Yes, I have measured the skew, because of the twists per inch (pitch?) the tighter the twist the longer the wire physical length as well. I read that a maximum of 50nS/100 meters was the limit that the hardware was able to correct.
But for sure you can measure the difference.
Geoff
toggle quoted message
Show quoted text
On 8/23/2023 5:02 PM, Bob Ecclestone VK2ZRE wrote: Hi Geoff,
You probably already know this, but each pair in a Cat6 cable has a different "twist rate" to minimise crosstalk, so the VF for each pair will be slightly different.
Try it with a piece of scrap cable. Measure the electrical length of each pair, you will find it will be slightly different for each pair.
I was involved with a startup VDSL broadband internet provider in the early 2000s and we used custom made, multiple pair Cat5 cable for the "last mile" (10, 25, 50 & 100pr).
We ended up characterising each pair in each cable size to help with fault finding. As with coax, the VF varied slightly along a cable and across batches, but it got us in the ball park to locate a fault.
Cheers...Bob VK2ZRE
On 24/08/2023 7:14 am, Geoff Peters - AB6BT wrote:
Sigi,
Yes, I have a piece of CAT6 cable to test. I removed the jacket at 1 foot intervals for about an inch. I will test by cutting 1 wire at a time at each location and see what I measure.
Geoff
On 8/23/2023 2:02 PM, Siegfried Jackstien wrote:
take 10 inch of speaker wire.. that should show almost similar numbers.. if thats the case then yes you can trust it
dg9bfc sigi
Am 23.08.2023 um 22:52 schrieb Geoff Peters - AB6BT:
Roger,
I sent a reply but it hasn't shown up so here's take 2...
In this case, what I'm trying to determine is the minimum distance I can measure to a fault.. I used 4.8mm to simplify this discussion, I'm not sure what you mean by "only in steps of 1 so this is the limiting factor." . I understand that the actual distance/step is VF dependent.
For example, I have a CAT6 cable with an open green wire. I have calculated the VF as I know the physical length of the cable, 300 feet/91.44 meters.
Using a NanoVNA-H4 I measure the distance to the fault at about 240mm.
So the question is...can I trust that number?
Geoff
|
Hi Geoff,
You probably already know this, but each pair in a Cat6 cable has a different "twist rate" to minimise crosstalk, so the VF for each pair will be slightly different.
Try it with a piece of scrap cable. Measure the electrical length of each pair, you will find it will be slightly different for each pair.
I was involved with a startup VDSL broadband internet provider in the early 2000s and we used custom made, multiple pair Cat5 cable for the "last mile" (10, 25, 50 & 100pr).
We ended up characterising each pair in each cable size to help with fault finding. As with coax, the VF varied slightly along a cable and across batches, but it got us in the ball park to locate a fault.
Cheers...Bob VK2ZRE
toggle quoted message
Show quoted text
On 24/08/2023 7:14 am, Geoff Peters - AB6BT wrote: Sigi,
Yes, I have a piece of CAT6 cable to test. I removed the jacket at 1 foot intervals for about an inch. I will test by cutting 1 wire at a time at each location and see what I measure.
Geoff
On 8/23/2023 2:02 PM, Siegfried Jackstien wrote:
take 10 inch of speaker wire.. that should show almost similar numbers.. if thats the case then yes you can trust it
dg9bfc sigi
Am 23.08.2023 um 22:52 schrieb Geoff Peters - AB6BT:
Roger,
I sent a reply but it hasn't shown up so here's take 2...
In this case, what I'm trying to determine is the minimum distance I can measure to a fault.. I used 4.8mm to simplify this discussion, I'm not sure what you mean by "only in steps of 1 so this is the limiting factor." . I understand that the actual distance/step is VF dependent.
For example, I have a CAT6 cable with an open green wire. I have calculated the VF as I know the physical length of the cable, 300 feet/91.44 meters.
Using a NanoVNA-H4 I measure the distance to the fault at about 240mm.
So the question is...can I trust that number?
Geoff
|
Right, yes, I know the VF is independent of length.
The test has nothing to do with determining VF. I have already done that with a piece of known physical length.
The test I have set up is a piece of CAT cable. I have removed the jacket at 1 foot intervals. I will cut 1 conductor starting at the 1 foot mark and see what length the NanoVNA shows for the fault. I'll reconnect the wire I cut and repeat the process at 1 foot intervals to see how the physical length numbers compare with the measured numbers.
Geoff
toggle quoted message
Show quoted text
On 8/23/2023 3:50 PM, Roger Need via groups.io wrote: On Wed, Aug 23, 2023 at 03:10 PM, Geoff Peters - AB6BT wrote:
OK, I see what you mean, and I suspect the the VF can vary by 1% over the
length of the cable.
I will test with a piece of CAT6 by cutting one wire and reconnecting at one
foot intervals.
The VF for the TDR test will be the same no matter what length you cut the CAT6 cable. VF for a specific type of cable will vary by manufacturer and even different between production runs by the same manufacturer. But for what you are measuring this should not be a concern. Take a measured length and then adjust VF to get the same calculated length using the TDR. Then you have the VF to use for with that specific type of cable.
Roger
|
On Wed, Aug 23, 2023 at 04:52 PM, Geoff Peters - AB6BT wrote:
So the question is...can I trust that number?
Measure from the other end and see if the two results add up to the correct total length. If they do you can at least have better confidence. If they doi not you know there is a problem. 73, Don N2VGU
|
On Wed, Aug 23, 2023 at 03:10 PM, Geoff Peters - AB6BT wrote:
OK, I see what you mean, and I suspect the the VF can vary by 1% over the
length of the cable.
I will test with a piece of CAT6 by cutting one wire and reconnecting at one
foot intervals.
The VF for the TDR test will be the same no matter what length you cut the CAT6 cable. VF for a specific type of cable will vary by manufacturer and even different between production runs by the same manufacturer. But for what you are measuring this should not be a concern. Take a measured length and then adjust VF to get the same calculated length using the TDR. Then you have the VF to use for with that specific type of cable. Roger
|
For some reason, my messages are not making it to the email chain...they are in the groups.io online...
toggle quoted message
Show quoted text
On 8/23/2023 3:01 PM, Roger Need via groups.io wrote: On Wed, Aug 23, 2023 at 01:52 PM, Geoff Peters - AB6BT wrote:
In this case, what I'm trying to determine is the minimum distance I can
measure to a fault.. I used 4.8mm to simplify this discussion, I'm not sure
what you mean by "only in steps of 1 so this is the limiting factor." . I
understand that the actual distance/step is VF dependent.
The resolution is 4.8 mm but the accuracy will be worse because you cannot enter the VF to more than two digits. So if the VF is 69.5 % and you have to enter 69 or 70% that will affect the calculation because the VF used in the NanoVNA is off by 0.5 %. But for practical purposes it will be close enough.
BTW - if you select the Format as "Real" in the menu the line will move to the center of the display and shorts and open can be determined by seeing if the pulse is above or below the line just like with a TDR test box.
Roger
|
OK, I see what you mean, and I suspect the the VF can vary by 1% over the length of the cable.
I will test with a piece of CAT6 by cutting one wire and reconnecting at one foot intervals.
|
On Wed, Aug 23, 2023 at 01:52 PM, Geoff Peters - AB6BT wrote:
In this case, what I'm trying to determine is the minimum distance I can
measure to a fault.. I used 4.8mm to simplify this discussion, I'm not sure
what you mean by "only in steps of 1 so this is the limiting factor." . I
understand that the actual distance/step is VF dependent.
The resolution is 4.8 mm but the accuracy will be worse because you cannot enter the VF to more than two digits. So if the VF is 69.5 % and you have to enter 69 or 70% that will affect the calculation because the VF used in the NanoVNA is off by 0.5 %. But for practical purposes it will be close enough. BTW - if you select the Format as "Real" in the menu the line will move to the center of the display and shorts and open can be determined by seeing if the pulse is above or below the line just like with a TDR test box. Roger
|
Sigi,
Yes, I have a piece of CAT6 cable to test. I removed the jacket at 1 foot intervals for about an inch. I will test by cutting 1 wire at a time at each location and see what I measure.
Geoff
toggle quoted message
Show quoted text
On 8/23/2023 2:02 PM, Siegfried Jackstien wrote: take 10 inch of speaker wire.. that should show almost similar numbers.. if thats the case then yes you can trust it
dg9bfc sigi
Am 23.08.2023 um 22:52 schrieb Geoff Peters - AB6BT:
Roger,
I sent a reply but it hasn't shown up so here's take 2...
In this case, what I'm trying to determine is the minimum distance I can measure to a fault.. I used 4.8mm to simplify this discussion, I'm not sure what you mean by "only in steps of 1 so this is the limiting factor." . I understand that the actual distance/step is VF dependent.
For example, I have a CAT6 cable with an open green wire. I have calculated the VF as I know the physical length of the cable, 300 feet/91.44 meters.
Using a NanoVNA-H4 I measure the distance to the fault at about 240mm.
So the question is...can I trust that number?
Geoff
|
take 10 inch of speaker wire.. that should show almost similar numbers.. if thats the case then yes you can trust it
dg9bfc sigi
Am 23.08.2023 um 22:52 schrieb Geoff Peters - AB6BT:
toggle quoted message
Show quoted text
Roger,
I sent a reply but it hasn't shown up so here's take 2...
In this case, what I'm trying to determine is the minimum distance I can measure to a fault.. I used 4.8mm to simplify this discussion, I'm not sure what you mean by "only in steps of 1 so this is the limiting factor." . I understand that the actual distance/step is VF dependent.
For example, I have a CAT6 cable with an open green wire. I have calculated the VF as I know the physical length of the cable, 300 feet/91.44 meters.
Using a NanoVNA-H4 I measure the distance to the fault at about 240mm.
So the question is...can I trust that number?
Geoff
|
Roger,
I sent a reply but it hasn't shown up so here's take 2...
In this case, what I'm trying to determine is the minimum distance I can measure to a fault.. I used 4.8mm to simplify this discussion, I'm not sure what you mean by "only in steps of 1 so this is the limiting factor." . I understand that the actual distance/step is VF dependent.
For example, I have a CAT6 cable with an open green wire. I have calculated the VF as I know the physical length of the cable, 300 feet/91.44 meters.
Using a NanoVNA-H4 I measure the distance to the fault at about 240mm.
So the question is...can I trust that number?
Geoff
|
Roger,
Mainly I'm trying to determine the minimum distance to a fault that I can measure.
I understand that the 4.8mm number is dependent on the VF, I was only trying to keep things simple for this discussion.
How reasonable is it to believe I can resolve a distance to within one step?
We use CAT6 cables for interconnecting audio equipment that gets pretty rough treatment on the road. Right now I have a quad CAT6 cable with an open green wire in one of the cables. I have determined the VF as I know the physical length of the cable.
Using NanoVNA-H4, I measure the break to be at about 240mm. So the real question is can I trust that number?
Geoff
toggle quoted message
Show quoted text
On 8/23/2023 1:21 PM, Roger Need via groups.io wrote: On Wed, Aug 23, 2023 at 10:07 AM, Geoff Peters - AB6BT wrote:
How does calibration affect the Thansform measurement? You need to be calibrated over the frequency range that you select. However there is calibration interpolation used on some NanoVNA's, like the -H and -H4, so you can arrow the frequency range if you want to measure longer cables. However for best accuracy calibrate over the frequency range that accommodates your cable length. The lower frequency should be kept at 50 kHz and the upper frequency limit , number of sample points and velocity factor entered will determine the maximum length that can be measured.
If I set the stimulus start and stop to 50KHz and 1.5GHz the transform maximum
length is shown as 19.36m. If I set the sweep to 401 points what resolution
and accuracy can I expect, provided that I have the VF entered correctly.
1936mm/401 equals just over 4.8mm.
Your calculation is correct. However you won't be accurate to 4.8 mm becasue the velocity factor you enter (in percent) is only in steps of 1 so this is the limiting factor.
Roger
|
On Wed, Aug 23, 2023 at 10:07 AM, Geoff Peters - AB6BT wrote:
How does calibration affect the Thansform measurement?
You need to be calibrated over the frequency range that you select. However there is calibration interpolation used on some NanoVNA's, like the -H and -H4, so you can arrow the frequency range if you want to measure longer cables. However for best accuracy calibrate over the frequency range that accommodates your cable length. The lower frequency should be kept at 50 kHz and the upper frequency limit , number of sample points and velocity factor entered will determine the maximum length that can be measured.
If I set the stimulus start and stop to 50KHz and 1.5GHz the transform maximum
length is shown as 19.36m. If I set the sweep to 401 points what resolution
and accuracy can I expect, provided that I have the VF entered correctly.
1936mm/401 equals just over 4.8mm.
Your calculation is correct. However you won't be accurate to 4.8 mm becasue the velocity factor you enter (in percent) is only in steps of 1 so this is the limiting factor. Roger
|
Yes, that's what I do, not a problem...
toggle quoted message
Show quoted text
On 8/23/2023 12:59 PM, Clyde Spencer wrote: Make sure you enter the VF as a percentage not a decimal value. EXAMPLE 60
not .60
*Clyde K. Spencer*
On Wed, Aug 23, 2023 at 1:08?PM Geoff Peters - AB6BT <AB6BT@...>
wrote:
I have a couple questions regarding using the Transform function to
measure length of a cable.
How does calibration affect the Thansform measurement?
If I set the stimulus start and stop to 50KHz and 1.5GHz the transform
maximum length is shown as 19.36m. If I set the sweep to 401 points what
resolution and accuracy can I expect, provided that I have the VF entered
correctly.
1936mm/401 equals just over 4.8mm.
Thanks,
Geoff -- AB6BT
|
Make sure you enter the VF as a percentage not a decimal value. EXAMPLE 60 not .60 *Clyde K. Spencer* On Wed, Aug 23, 2023 at 1:08?PM Geoff Peters - AB6BT <AB6BT@...> wrote: I have a couple questions regarding using the Transform function to
measure length of a cable.
How does calibration affect the Thansform measurement?
If I set the stimulus start and stop to 50KHz and 1.5GHz the transform
maximum length is shown as 19.36m. If I set the sweep to 401 points what
resolution and accuracy can I expect, provided that I have the VF entered
correctly.
1936mm/401 equals just over 4.8mm.
Thanks,
Geoff -- AB6BT
|
I have a couple questions regarding using the Transform function to measure length of a cable.
How does calibration affect the Thansform measurement?
If I set the stimulus start and stop to 50KHz and 1.5GHz the transform maximum length is shown as 19.36m. If I set the sweep to 401 points what resolution and accuracy can I expect, provided that I have the VF entered correctly.
1936mm/401 equals just over 4.8mm.
Thanks,
Geoff -- AB6BT
|
Re: some Nano VNA SAVER 40m antenna graphs
That will be a very interesting comparison test!
I hope you publish all the details and measurements.
(Great choices of antennas, too. My full-wave horizontal loop is terrific.)
--
Doug, K8RFT
|
Barry K3EUI
Donald S Brant Jr
Roger Need
