|
Coming along nicely here, but here are some questions that might need answers or indicate a need for more research:
1. One of the roles i am hoping the nanoVNA will fill is that of signal generator. I want to program the device for the desired sweep frequencies, use a cable from ch0 to a female F coax adaptor to a home brew coupling capacitor DC isolation probe. The probe will supply a grounding clip. I will not have anything going to ch1. I will use the probe to inject the swept frequencies from ch0 into live circuits either to trace via other devices, or to evaluate the circuit. Will this damage the device? Should I leave ch1 open or short it?
2. If I understand what I have read, calibration establishes the scaling relationship between the input parameters and output results. If not calibrated for each set of input parameters, then the output results are unreliable. This is taking the BIG VIEW as a general statement. I am sure some parameters may be changed without re-calibrating but I am too inexperienced to know what they are.
3. If (2) is correct, then the only time you HAVE to re-calibrate is when you are interested in the results.
4. Even if (2) is correct, best to get into the habit of calibration every time I turn it on or change parameters; like socks before shoes.
I have read the Guide and watched the videos. I mainly want to validate my interpretation of all I looked at. Also, I have not found much of anything about a capacitor for DC isolation.
All comments appreciated,
Billy
<>[cid:55e85da0-003e-40c5-814b-f75d440c91ee][cid:2f624a96-5108-43a9-89db-54f54c537d27]<><>
|
ONE CAUTION: You are using a decoupling capacitor. If you are coupling to
anything like vacuum tube circuits with high voltages, the spike which
occurs when you initially connect the "test end" of the decoupling
capacitor to your circuit could produce enough voltage impulse to damage
the input of the VNA. That rapid spike will propagate RF energy from the
small arc right into the NANO. I'd Strongly recommend some reasonably high
series resistance between the "test end" of our decoupling capacitor and
the circuit under test. After all, the NANOs operate on 3.5 volts and
even small sparks or arcs potentially produce RF energy from DC to
light.......
Dave - W?LEV
toggle quoted message
Show quoted text
On Sun, Dec 19, 2021 at 12:03 AM N5SE <bwmoore@...> wrote: Coming along nicely here, but here are some questions that might need
answers or indicate a need for more research:
1. One of the roles i am hoping the nanoVNA will fill is that of signal
generator. I want to program the device for the desired sweep frequencies,
use a cable from ch0 to a female F coax adaptor to a home brew coupling
capacitor DC isolation probe. The probe will supply a grounding clip. I
will not have anything going to ch1. I will use the probe to inject the
swept frequencies from ch0 into live circuits either to trace via other
devices, or to evaluate the circuit. Will this damage the device? Should I
leave ch1 open or short it?
2. If I understand what I have read, calibration establishes the
scaling relationship between the input parameters and output results. If
not calibrated for each set of input parameters, then the output results
are unreliable. This is taking the BIG VIEW as a general statement. I am
sure some parameters may be changed without re-calibrating but I am too
inexperienced to know what they are.
3. If (2) is correct, then the only time you HAVE to re-calibrate is
when you are interested in the results.
4. Even if (2) is correct, best to get into the habit of calibration
every time I turn it on or change parameters; like socks before shoes.
I have read the Guide and watched the videos. I mainly want to validate my
interpretation of all I looked at. Also, I have not found much of anything
about a capacitor for DC isolation.
All comments appreciated,
Billy
<
[cid:55e85da0-003e-40c5-814b-f75d440c91ee][cid:2f624a96-5108-43a9-89db-54f54c537d27]< ><>
--
*Dave - W?LEV*
*Just Let Darwin Work*
|
Another reason to include a high value series resistor (high, meaning a
few thousand ohms) is to reduce the load placed on the circuit under
test by the signal injection probe. The impedance looking into either
of the ports of the VNA is about 50 ohms. Many circuits that you might
want to test will have a much higher impedance to ground. Connecting a
low-impedance probe would almost short-circuit the circuit being tested.
The injection probe's series impedance could be made high using either a
high value resistor, or a small value capacitor (high reactance).
A note on "sweeping": the VNA probably does not continuously sweep the
frequency, but most likely jumps the frequency from one value to the
next, so that, if the VNA is set to cover 3 to 4 MHz with 100 points,
the frequency will make 100 jumps (steps) of 10kHz.
Dave WA8YWQ
toggle quoted message
Show quoted text
On 2021-12-18 16:17, W0LEV wrote: ONE CAUTION: You are using a decoupling capacitor. If you are coupling to
anything like vacuum tube circuits with high voltages, the spike which
occurs when you initially connect the "test end" of the decoupling
capacitor to your circuit could produce enough voltage impulse to damage
the input of the VNA. That rapid spike will propagate RF energy from the
small arc right into the NANO. I'd Strongly recommend some reasonably high
series resistance between the "test end" of our decoupling capacitor and
the circuit under test. After all, the NANOs operate on 3.5 volts and
even small sparks or arcs potentially produce RF energy from DC to
light.......
Dave - W?LEV
On Sun, Dec 19, 2021 at 12:03 AM N5SE <bwmoore@...> wrote:
Coming along nicely here, but here are some questions that might need
answers or indicate a need for more research:
1. One of the roles i am hoping the nanoVNA will fill is that of signal
generator. I want to program the device for the desired sweep frequencies,
use a cable from ch0 to a female F coax adaptor to a home brew coupling
capacitor DC isolation probe. The probe will supply a grounding clip. I
will not have anything going to ch1. I will use the probe to inject the
swept frequencies from ch0 into live circuits either to trace via other
devices, or to evaluate the circuit. Will this damage the device? Should I
leave ch1 open or short it?
2. If I understand what I have read, calibration establishes the
scaling relationship between the input parameters and output results. If
not calibrated for each set of input parameters, then the output results
are unreliable. This is taking the BIG VIEW as a general statement. I am
sure some parameters may be changed without re-calibrating but I am too
inexperienced to know what they are.
3. If (2) is correct, then the only time you HAVE to re-calibrate is
when you are interested in the results.
4. Even if (2) is correct, best to get into the habit of calibration
every time I turn it on or change parameters; like socks before shoes.
I have read the Guide and watched the videos. I mainly want to validate my
interpretation of all I looked at. Also, I have not found much of anything
about a capacitor for DC isolation.
All comments appreciated,
Billy
< [cid:55e85da0-003e-40c5-814b-f75d440c91ee][cid:2f624a96-5108-43a9-89db-54f54c537d27]< ><>
--
*Dave - W?LEV*
*Just Let Darwin Work*
|
1. The nanovna will not work well as a signal sweep generator for at least
two reasons: a) as noted in a previous response, it's output is not a
continuous sweep, and b) it outputs a square wave, not a sine wave. You
can read more about these issues in forum posts by searching for the
terms. But bottom line is, don't do it, you won't be happy with all you
have to do to make it work. Instead, buy a tinySA - it will do that job
nicely, since its design includes that purpose.
2. Only kind-of sort-of, if you define input parameters as the cables and
test fixture setup you attach to the vna terminals, along with the
frequency range selected. A perhaps better statement about calibration is
that it creates a set of normalizing coefficients that compensate for your
test cables and fixture across the frequency scan points the vna is using.
3. Obviously, if you don't care about the results you don't have to
calibrate. More specifically, if you don't care about the exact values at
each frequency point, but only want to see what frequency has the dip in
SWR or is the resonant point, etc., you don't need to calibrate.
Calibration does not effect the frequency measurements.
4. If you calibrate for a specific test cabling setup, over a given
frequency range, and save that calibration, you never have to recalibrate
for it again, just recall the saved calibration. If a saved calibration
used, for an example, an HF range from 3-30MHz, and you wanted to zoom in
and set the range to 13-15MHz, the nano would still give reasonably good
results without recalibration, since it does a nice job of interpoating to
create a new set of coefficients in the sub-range. This works very well for
quick checks of performance at different sub-ranges of a calibration.
However, for greatest accuracy, you need to recalibrate at the new range.
toggle quoted message
Show quoted text
On Sat, Dec 18, 2021, 4:57 PM David McQuate <mcquate@...> wrote: Another reason to include a high value series resistor (high, meaning a
few thousand ohms) is to reduce the load placed on the circuit under
test by the signal injection probe. The impedance looking into either
of the ports of the VNA is about 50 ohms. Many circuits that you might
want to test will have a much higher impedance to ground. Connecting a
low-impedance probe would almost short-circuit the circuit being tested.
The injection probe's series impedance could be made high using either a
high value resistor, or a small value capacitor (high reactance).
A note on "sweeping": the VNA probably does not continuously sweep the
frequency, but most likely jumps the frequency from one value to the
next, so that, if the VNA is set to cover 3 to 4 MHz with 100 points,
the frequency will make 100 jumps (steps) of 10kHz.
Dave WA8YWQ
On 2021-12-18 16:17, W0LEV wrote:
ONE CAUTION: You are using a decoupling capacitor. If you are coupling to
anything like vacuum tube circuits with high voltages, the spike which
occurs when you initially connect the "test end" of the decoupling
capacitor to your circuit could produce enough voltage impulse to damage
the input of the VNA. That rapid spike will propagate RF energy from the
small arc right into the NANO. I'd Strongly recommend some reasonably high
series resistance between the "test end" of our decoupling capacitor and
the circuit under test. After all, the NANOs operate on 3.5 volts and
even small sparks or arcs potentially produce RF energy from DC to
light.......
Dave - W?LEV
On Sun, Dec 19, 2021 at 12:03 AM N5SE <bwmoore@...> wrote:
Coming along nicely here, but here are some questions that might need
answers or indicate a need for more research:
1. One of the roles i am hoping the nanoVNA will fill is that of signal
generator. I want to program the device for the desired sweep frequencies,
use a cable from ch0 to a female F coax adaptor to a home brew coupling
capacitor DC isolation probe. The probe will supply a grounding clip. I
will not have anything going to ch1. I will use the probe to inject the
swept frequencies from ch0 into live circuits either to trace via other
devices, or to evaluate the circuit. Will this damage the device? Should I
leave ch1 open or short it?
2. If I understand what I have read, calibration establishes the
scaling relationship between the input parameters and output results. If
not calibrated for each set of input parameters, then the output results
are unreliable. This is taking the BIG VIEW as a general statement. I am
sure some parameters may be changed without re-calibrating but I am too
inexperienced to know what they are.
3. If (2) is correct, then the only time you HAVE to re-calibrate is
when you are interested in the results.
4. Even if (2) is correct, best to get into the habit of calibration
every time I turn it on or change parameters; like socks before shoes.
I have read the Guide and watched the videos. I mainly want to validate my
interpretation of all I looked at. Also, I have not found much of anything
about a capacitor for DC isolation.
All comments appreciated,
Billy
< [cid:55e85da0-003e-40c5-814b-f75d440c91ee][cid:2f624a96-5108-43a9-89db-54f54c537d27]<
><>
--
*Dave - W?LEV*
*Just Let Darwin Work*
|
Stan: All very good comments. I would like to expand on your statement regarding calibration interpolation. How "nice" a job it does depends somewhat on the quality of the instrument and setup (cables, adapters, etc.); If the setup is well-matched (to 50¦¸) and low loss, and does not have any resonances, the software will be able to interpolate well. I do not believe that it will extrapolate outside of a calibrated range, however.
But you need to keep in mind the sweep width and number of discrete measurement points so that you do not miss an important feature by stepping past it; this would be an issue with a narrowband filter or high-Q antenna such as a tuned loop..
73, Don N2VGU
|
Stan, Yes, I had already picked up on that. Also picked up on degradation outside the calibration range. My post was sent out before I understood the full implications of calibration Thanks a lot, 30 Billy ________________________________ From: [email protected] < [email protected]> on behalf of Donald S Brant Jr via groups.io <dsbrantjr@...> Sent: Sunday, December 19, 2021 10:48 To: [email protected] < [email protected]> Subject: Re: [nanovna-users] NanoVNA as a Signal Generator Stan: All very good comments. I would like to expand on your statement regarding calibration interpolation. How "nice" a job it does depends somewhat on the quality of the instrument and setup (cables, adapters, etc.); If the setup is well-matched (to 50¦¸) and low loss, and does not have any resonances, the software will be able to interpolate well. I do not believe that it will extrapolate outside of a calibrated range, however. But you need to keep in mind the sweep width and number of discrete measurement points so that you do not miss an important feature by stepping past it; this would be an issue with a narrowband filter or high-Q antenna such as a tuned loop.. 73, Don N2VGU
|
Dave, That "spike" was the exact reason for my post. I was worried about destroying my NanoVNA even though no outside data was sent to it. Even though I would not be injection signals in circuits fed by higher than 13.6 Volts, I have decided to abandon using the NanoVNA as a signal generator. I have a number of more attractive options. Many thanks to all who commented on my post. 30 Billy ________________________________ From: [email protected] < [email protected]> on behalf of W0LEV via groups.io <davearea51a@...> Sent: Saturday, December 18, 2021 18:17 To: NANO VNA < [email protected]> Subject: Re: [nanovna-users] NanoVNA as a Signal Generator ONE CAUTION: You are using a decoupling capacitor. If you are coupling to anything like vacuum tube circuits with high voltages, the spike which occurs when you initially connect the "test end" of the decoupling capacitor to your circuit could produce enough voltage impulse to damage the input of the VNA. That rapid spike will propagate RF energy from the small arc right into the NANO. I'd Strongly recommend some reasonably high series resistance between the "test end" of our decoupling capacitor and the circuit under test. After all, the NANOs operate on 3.5 volts and even small sparks or arcs potentially produce RF energy from DC to light....... Dave - W?LEV On Sun, Dec 19, 2021 at 12:03 AM N5SE <bwmoore@...> wrote: Coming along nicely here, but here are some questions that might need
answers or indicate a need for more research:
1. One of the roles i am hoping the nanoVNA will fill is that of signal
generator. I want to program the device for the desired sweep frequencies,
use a cable from ch0 to a female F coax adaptor to a home brew coupling
capacitor DC isolation probe. The probe will supply a grounding clip. I
will not have anything going to ch1. I will use the probe to inject the
swept frequencies from ch0 into live circuits either to trace via other
devices, or to evaluate the circuit. Will this damage the device? Should I
leave ch1 open or short it?
2. If I understand what I have read, calibration establishes the
scaling relationship between the input parameters and output results. If
not calibrated for each set of input parameters, then the output results
are unreliable. This is taking the BIG VIEW as a general statement. I am
sure some parameters may be changed without re-calibrating but I am too
inexperienced to know what they are.
3. If (2) is correct, then the only time you HAVE to re-calibrate is
when you are interested in the results.
4. Even if (2) is correct, best to get into the habit of calibration
every time I turn it on or change parameters; like socks before shoes.
I have read the Guide and watched the videos. I mainly want to validate my
interpretation of all I looked at. Also, I have not found much of anything
about a capacitor for DC isolation.
All comments appreciated,
Billy
<
[cid:55e85da0-003e-40c5-814b-f75d440c91ee][cid:2f624a96-5108-43a9-89db-54f54c537d27]< ><>
-- *Dave - W?LEV* *Just Let Darwin Work*
|
David, Thanks so much for you input, proper coupling internal to the circuit is surely interfered with by the probe. Good point to consider. Due to this and a number of other reasons, I have decided to abandon using the NanoVNA as a signal generator. I have a number of more attractive options. Many thanks to all who commented on my post. 30 Billy ________________________________ From: [email protected] < [email protected]> on behalf of David McQuate via groups.io <mcquate@...> Sent: Saturday, December 18, 2021 18:57 To: [email protected] < [email protected]> Subject: Re: [nanovna-users] NanoVNA as a Signal Generator Another reason to include a high value series resistor (high, meaning a few thousand ohms) is to reduce the load placed on the circuit under test by the signal injection probe. The impedance looking into either of the ports of the VNA is about 50 ohms. Many circuits that you might want to test will have a much higher impedance to ground. Connecting a low-impedance probe would almost short-circuit the circuit being tested. The injection probe's series impedance could be made high using either a high value resistor, or a small value capacitor (high reactance). A note on "sweeping": the VNA probably does not continuously sweep the frequency, but most likely jumps the frequency from one value to the next, so that, if the VNA is set to cover 3 to 4 MHz with 100 points, the frequency will make 100 jumps (steps) of 10kHz. Dave WA8YWQ On 2021-12-18 16:17, W0LEV wrote: ONE CAUTION: You are using a decoupling capacitor. If you are coupling to
anything like vacuum tube circuits with high voltages, the spike which
occurs when you initially connect the "test end" of the decoupling
capacitor to your circuit could produce enough voltage impulse to damage
the input of the VNA. That rapid spike will propagate RF energy from the
small arc right into the NANO. I'd Strongly recommend some reasonably high
series resistance between the "test end" of our decoupling capacitor and
the circuit under test. After all, the NANOs operate on 3.5 volts and
even small sparks or arcs potentially produce RF energy from DC to
light.......
Dave - W?LEV
On Sun, Dec 19, 2021 at 12:03 AM N5SE <bwmoore@...> wrote:
Coming along nicely here, but here are some questions that might need
answers or indicate a need for more research:
1. One of the roles i am hoping the nanoVNA will fill is that of signal
generator. I want to program the device for the desired sweep frequencies,
use a cable from ch0 to a female F coax adaptor to a home brew coupling
capacitor DC isolation probe. The probe will supply a grounding clip. I
will not have anything going to ch1. I will use the probe to inject the
swept frequencies from ch0 into live circuits either to trace via other
devices, or to evaluate the circuit. Will this damage the device? Should I
leave ch1 open or short it?
2. If I understand what I have read, calibration establishes the
scaling relationship between the input parameters and output results. If
not calibrated for each set of input parameters, then the output results
are unreliable. This is taking the BIG VIEW as a general statement. I am
sure some parameters may be changed without re-calibrating but I am too
inexperienced to know what they are.
3. If (2) is correct, then the only time you HAVE to re-calibrate is
when you are interested in the results.
4. Even if (2) is correct, best to get into the habit of calibration
every time I turn it on or change parameters; like socks before shoes.
I have read the Guide and watched the videos. I mainly want to validate my
interpretation of all I looked at. Also, I have not found much of anything
about a capacitor for DC isolation.
All comments appreciated,
Billy
< [cid:55e85da0-003e-40c5-814b-f75d440c91ee][cid:2f624a96-5108-43a9-89db-54f54c537d27]< ><> -- *Dave - W?LEV* *Just Let Darwin Work*
|
So (newbie here) I've seen some versions of nanoVNA that capture 1000 data points rather than 101. What are the consequences of the larger number of points? Thanks, Rich AI7KI On Sun, Dec 19, 2021 at 9:48 AM Donald S Brant Jr <dsbrantjr@...> wrote: Stan: All very good comments. I would like to expand on your statement
regarding calibration interpolation. How "nice" a job it does depends
somewhat on the quality of the instrument and setup (cables, adapters,
etc.); If the setup is well-matched (to 50¦¸) and low loss, and does not
have any resonances, the software will be able to interpolate well. I do
not believe that it will extrapolate outside of a calibrated range,
however.
But you need to keep in mind the sweep width and number of discrete
measurement points so that you do not miss an important feature by stepping
past it; this would be an issue with a narrowband filter or high-Q antenna
such as a tuned loop..
73, Don N2VGU
|
Rich AI7KI I am probably more newbie than you and not qualified to answer the question. But I may be able to answer part of the question. I own a NanoVNA-H. The firmware supports more than ten different versions of NanoVNA; many of which have different sizes and resolution of display. On this forum, there are people that own a NanoVNA of each possible version. I am too uneducated at this time explain all the different possibilities; perhaps another forum member can jump in and elaborate. 30 Billy ________________________________ From: [email protected] < [email protected]> on behalf of Richard Groot via groups.io <richgroot@...> Sent: Sunday, December 19, 2021 11:26 To: [email protected] < [email protected]> Subject: Re: [nanovna-users] NanoVNA as a Signal Generator So (newbie here) I've seen some versions of nanoVNA that capture 1000 data points rather than 101. What are the consequences of the larger number of points? Thanks, Rich AI7KI On Sun, Dec 19, 2021 at 9:48 AM Donald S Brant Jr <dsbrantjr@...> wrote: Stan: All very good comments. I would like to expand on your statement
regarding calibration interpolation. How "nice" a job it does depends
somewhat on the quality of the instrument and setup (cables, adapters,
etc.); If the setup is well-matched (to 50¦¸) and low loss, and does not
have any resonances, the software will be able to interpolate well. I do
not believe that it will extrapolate outside of a calibrated range,
however.
But you need to keep in mind the sweep width and number of discrete
measurement points so that you do not miss an important feature by stepping
past it; this would be an issue with a narrowband filter or high-Q antenna
such as a tuned loop..
73, Don N2VGU
|
Re: Number of points, I do not know of any downside except perhaps a slower sweep. There are software applications which will take control of the VNA and allow a 10,001 or greater point sweep.
Having said that for most uses 101 or 201 points is perfectly suitable; many commercial VNAs had similar numbers of points and were perfectly useable.
73, Don N2VGU
|
On Sun, Dec 19, 2021 at 09:41 AM, Richard Groot wrote: So (newbie here) I've seen some versions of nanoVNA that capture 1000 data
points rather than 101. What are the consequences of the larger number of
points?
Disadvantage - with more points you get a slower sweep time. Advantages - - with more points the step size for a given frequency range is smaller so you get more plotting "definition". When measuring an antenna, for example, this will give a better estimate of the frequency with minimum SWR. - also you can "zoom in" (without recalibrating) and get better results if you have more data points. For example if you have 401 calibrated points over frequency range "A" and then zoom in by a factor of four to frequency range "B" 100 of the points will be close to what you get if you calibrate and measure over frequency range "B" with a 100 point sweep. Three interpolated points will be also available. Roger
|
N5SE
Donald S Brant Jr
Roger Need