|
Interesting topics being discussed.
I was surprised to read that the RL from a short or open is 12.XXXdB. That just isn't true, ever (assuming, of course, lossless feedline/ connectors). Its 0. Always will be. Energy sent down the feedline to either the short or open has "no where" to go so is 100% reflected.?
I also read loss is always negative? When calculating power out of a system sure. But when expressing it (such and such feedline has XdB of loss) its always a positive number. Negative loss implies gain.
I see this often so I can readily understand how folks can get this wrong. Professionals get it wrong. I saw one commercial (not for ham radio) amplifier maker have a graph of return loss labeled in -Db.?
All that said, I find the nanovna a very useful device. Much easier to bring to the tower base than my HP 8753. I look forward to using it more.
Sent using
|
A loss is a negative gain. So, saying a return loss is negative simply means you are expressing that loss as a formal gain. Don't overdo the besserwisserness of your insights. Den ons 21 aug. 2019 20:07Ron Spencer via Groups.Io <ron.spencer= [email protected]> skrev:
Interesting topics being discussed.
I was surprised to read that the RL from a short or open is 12.XXXdB. That
just isn't true, ever (assuming, of course, lossless feedline/ connectors).
Its 0. Always will be. Energy sent down the feedline to either the short or
open has "no where" to go so is 100% reflected.
I also read loss is always negative? When calculating power out of a
system sure. But when expressing it (such and such feedline has XdB of
loss) its always a positive number. Negative loss implies gain.
I see this often so I can readily understand how folks can get this wrong.
Professionals get it wrong. I saw one commercial (not for ham radio)
amplifier maker have a graph of return loss labeled in -Db.
All that said, I find the nanovna a very useful device. Much easier to
bring to the tower base than my HP 8753. I look forward to using it more.
Sent using
|
This is all about definitions.
The reflection coefficient, gamma, has both magnitude and phase values and is defined by greek letter RHO. The value of RHO varies between 0 to 1.
Return loss is defined in dB and is equal to -20 log RHO. Since we are addressing voltages, thus the 20 dB multiplier.
This results in a POSITIVE dB value even though RHO is less than 1. Values range between 0 dB for RHO of 1 (total reflection) to infinite dB for RHO equal to zero (a perfectly matched system).
hp measurement systems now Keysight made available some time ago a nice slide rule reflectometer calculator part number 5952-0948 that provides these calculations and more for vector measurements.
In the amateur radio world as well the "professional" design arena where high efficiency or broadband systems may be involved, a 20 dB return loss or better puts a smile on the designers face. Attempts to accomplish better (30 dB, 40 dB or more) and at some additional cost is usually frowned upon.
Alan
|
Return loss is a very old term used in the telephone industry for a very long time. There is confusion about the sign, but the negative has been used as long at the term return loss has been in use. Most of us understand the meaning and implications.The 12 dB nonsense came up in a 100% reflection measurement because of losses in the RL bridge, but it's meaningless.The reference for either an open or short is a return loss of zero. The bridge losses should only reduce the dynamic range of a measurement, not the accuracy.However, if you have any transmission loss between the RL measuring device and the 100% reflection, it will give you a better return loss than zero.If a coax has a 3dB loss for instance, it will show (in theory) a 6 dB return loss because the RF will travel twice through a 3dB loss (the coax).
Stuart K6YAZLos Angeles, USA
toggle quoted message
Show quoted text
-----Original Message----- From: Hans J Albertsson <hans.j.albertsson@...> To: nanovna-users < [email protected]> Sent: Wed, Aug 21, 2019 11:33 am Subject: Re: [nanovna-users] New to group and thoughts on Return Loss and Loss A loss is a negative gain. So, saying a return loss is negative simply means you are expressing that loss as a formal gain. Don't overdo the besserwisserness of your insights. Den ons 21 aug. 2019 20:07Ron Spencer via Groups.Io <ron.spencer= [email protected]> skrev: Interesting topics being discussed.
I was surprised to read that the RL from a short or open is 12.XXXdB. That
just isn't true, ever (assuming, of course, lossless feedline/ connectors).
Its 0. Always will be. Energy sent down the feedline to either the short or
open has "no where" to go so is 100% reflected.
I also read loss is always negative? When calculating power out of a
system sure. But when expressing it (such and such feedline has XdB of
loss) its always a positive number. Negative loss implies gain.
I see this often so I can readily understand how folks can get this wrong.
Professionals get it wrong. I saw one commercial (not for ham radio)
amplifier maker have a graph of return loss labeled in -Db.
All that said, I find the nanovna a very useful device. Much easier to
bring to the tower base than my HP 8753. I look forward to using it more.
Sent using
|
Indeed.
Do a web search for "is return loss positive or negative".
You'll turn up lots of forums in which to pontificate.
Short answer:
It's done both ways, and nobody finds this confusing.
But they sure have fun arguing about it.
My new word for the day: besserwisserness
Could be handy, now I'll know what they mean
when people point it at me!
Jerry
toggle quoted message
Show quoted text
On Wed, Aug 21, 2019 at 11:33 AM, Hans J Albertsson wrote: Don't overdo the besserwisserness of your insights.
|
On Wed, Aug 21, 2019 at 01:08 PM, alan victor wrote:
The reflection coefficient, gamma, has both magnitude and phase values and is
defined by greek letter RHO. The value of RHO varies between 0 to 1.
Alan, I have to laugh (in a good way, not maniacally), as this is another one of those things were usage varies from one person to another. The reflection coefficient, gamma, is actually defined by the greek letter GAMMA, not RHO. To quote Keysight (page 3 of: ), rho is actually the magnitude of gamma. That is, rho, plus angle theta, defines gamma. But it's no big deal, I've seen rho and gamma used interchangeably by many others (and I've been guilty of it, too). - Jeff, k6jca
|
In my early career I had the wonderful opportunity to work, in sales, for HP. Back before it was Agilent or Keysight. Back when Bill and Dave were alive and running the company.?
They spent a great deal of time and money educating us. So we might have some credibility when talking with engineers. One thing I will always remember is that return loss is ALWAYS positive. No idea why this stuck as it has.?
Indeed you will see it expressed both ways. Only one is correct. While I am the one saying it (here) I'm only repeating what the fine folks at HP (who, I think we can all agree, knew their stuff) taught me and generations of engineers.?
Simply put, its the measurement of signal returned (reflected) vs signal applied. How can that be negative? The most you can get back is 100% (short or open) or 0 (perfectly matched load). Everything else is something in between.?
Some things are gray, some black and white. This is the later.?
Sent using
---- On Wed, 21 Aug 2019 16:58:01 -0400 Jerry Gaffke via Groups.Io <jgaffke@...> wrote ----
Indeed.
Do a web search for "is return loss positive or negative".
You'll turn up lots of forums in which to pontificate.
Short answer:
It's done both ways, and nobody finds this confusing.
But they sure have fun arguing about it.
My new word for the day: besserwisserness
Could be handy, now I'll know what they mean
when people point it at me!
Jerry
toggle quoted message
Show quoted text
On Wed, Aug 21, 2019 at 11:33 AM, Hans J Albertsson wrote: Don't overdo the besserwisserness of your insights.
|
It really is all magic.
Stuart K6YAZ
toggle quoted message
Show quoted text
-----Original Message----- From: Jeff Anderson <jca1955@...> To: nanovna-users < [email protected]> Sent: Wed, Aug 21, 2019 2:13 pm Subject: Re: [nanovna-users] New to group and thoughts on Return Loss and Loss On Wed, Aug 21, 2019 at 01:08 PM, alan victor wrote:
The reflection coefficient, gamma, has both magnitude and phase values and is
defined by greek letter RHO. The value of RHO varies between 0 to 1.
Alan, I have to laugh (in a good way, not maniacally), as this is another one of those things were usage varies from one person to another. The reflection coefficient, gamma, is actually defined by the greek letter GAMMA, not RHO.? To quote Keysight (page 3 of: ), rho is actually the magnitude of gamma.? That is, rho, plus angle theta, defines gamma. But it's no big deal, I've seen rho and gamma used interchangeably by many others (and I've been guilty of it, too). - Jeff, k6jca
|
In reading my older HP catalogs, they seldom if ever referred to return loss; it was always SWR. I found that sort of odd.
Stuart K6YAZ
toggle quoted message
Show quoted text
-----Original Message----- From: Ron Spencer via Groups.Io <ron.spencer@...> To: nanovna-users < [email protected]> Sent: Wed, Aug 21, 2019 2:26 pm Subject: Re: [nanovna-users] New to group and thoughts on Return Loss and Loss In my early career I had the wonderful opportunity to work, in sales, for HP. Back before it was Agilent or Keysight. Back when Bill and Dave were alive and running the company.? They spent a great deal of time and money educating us. So we might have some credibility when talking with engineers. One thing I will always remember is that return loss is ALWAYS positive. No idea why this stuck as it has.? Indeed you will see it expressed both ways. Only one is correct. While I am the one saying it (here) I'm only repeating what the fine folks at HP (who, I think we can all agree, knew their stuff) taught me and generations of engineers.? Simply put, its the measurement of signal returned (reflected) vs signal applied. How can that be negative? The most you can get back is 100% (short or open) or 0 (perfectly matched load). Everything else is something in between.? Some things are gray, some black and white. This is the later.? Sent using ---- On Wed, 21 Aug 2019 16:58:01 -0400 Jerry Gaffke via Groups.Io <jgaffke@...> wrote ---- Indeed. Do a web search for "is return loss positive or negative". You'll turn up lots of forums in which to pontificate. Short answer: It's done both ways, and nobody finds this confusing. But they sure have fun arguing about it. My new word for the day:? besserwisserness Could be handy, now I'll know what they mean when people point it at me! Jerry On Wed, Aug 21, 2019 at 11:33 AM, Hans J Albertsson wrote: Don't overdo the besserwisserness of your insights.
|
Judging from that web search, "return loss" is very gray.
If you're talking "s11", that's black and white,
as we have now moved from casual speech to the world of mathematics.
Jerry
toggle quoted message
Show quoted text
On Wed, Aug 21, 2019 at 02:26 PM, Ron Spencer wrote: Some things are gray, some black and white. This is the later.
|
Here is an example of how confusion is created. And how this confusion causes errors.
From a web search:
For a given RF transformer, the Insertion Loss @ 10 MHz is 0.5 dB, and the Return Loss @ 10 MHz is 25 dB, with impedances of 50 Ohm. Let's say I put in a 1 V, 10 MHz sine wave, what happens?
Return loss tells you how much of the input signal is reflected. Return loss is the ratio between the reflected power and input power:
RL=PrefPinRL=PrefPin
If the input signal is 0 dBm and there is 25 dB return loss, then the component will create a reflected wave of -25 dBm back toward the generator.
In your example, I assume you mean a 1 V rms signal (as opposed to 1 V amplitude or 1 V peak-peak). This is +13 dBm. With 25 dBm return loss the reflected wave has -12 dBm power or 56 mV rms amplitude.----
Read it carefully and you'll see this author mixed dB and dBm for return loss. dBm is a power measurement. (I also think the statement that a RL of 25dB creates a reflected was of -25dBm is confusing. I'd have to do the math but don't think its correct.)?
Return loss is completely different than power. But, as seen above, people easily mix the two. It doesn't help that the author wove in insertion loss, which most people also spec as negative, which is incorrect. When you talk about loss, its a positive number unless you are talking about gain. But if you were talking about gain you'd most likely say gain not negative loss.?
Jerry, return loss is not gray at all. But you have to sort out the garbage on the internet (LOTS of that!!) from fact.?
Judging from that web search, "return loss" is very gray.
If you're talking "s11", that's black and white,
as we have now moved from casual speech to the world of mathematics.
Jerry
toggle quoted message
Show quoted text
On Wed, Aug 21, 2019 at 02:26 PM, Ron Spencer wrote: Some things are gray, some black and white. This is the later.
|
When using a spectrum analyzer/tracking generator/RF bridge to make S11 measurements, part of the setup is to normalize the results to zero with an open at the DUT port. This the the analog of the ¡°O¡± calibration of the VNA.
A reading of ¡°-12 dB¡± or, in my case, -19 dB¡¡ is simply the inherent loss of the bridge and it must be normalized out if you need a direct reading of RL or used as an offset for the uncalibrated reading.
Return losses of DUT at the far end of a coax line require calibration at that end, not at the device itself.
WA8TOD
toggle quoted message
Show quoted text
On Aug 21, 2019, at 4:23 PM, Stuart Landau via Groups.Io <stuartl73@...> wrote: Return loss is a very old term used in the telephone industry for a very long time. There is confusion about the sign, but the negative has been used as long at the term return loss has been in use. Most of us understand the meaning and implications.The 12 dB nonsense came up in a 100% reflection measurement because of losses in the RL bridge, but it's meaningless.The reference for either an open or short is a return loss of zero. The bridge losses should only reduce the dynamic range of a measurement, not the accuracy.However, if you have any transmission loss between the RL measuring device and the 100% reflection, it will give you a better return loss than zero.If a coax has a 3dB loss for instance, it will show (in theory) a 6 dB return loss because the RF will travel twice through a 3dB loss (the coax). Stuart K6YAZLos Angeles, USA -----Original Message----- From: Hans J Albertsson <hans.j.albertsson@...> To: nanovna-users < [email protected]> Sent: Wed, Aug 21, 2019 11:33 am Subject: Re: [nanovna-users] New to group and thoughts on Return Loss and Loss A loss is a negative gain. So, saying a return loss is negative simply means you are expressing that loss as a formal gain. Don't overdo the besserwisserness of your insights. Den ons 21 aug. 2019 20:07Ron Spencer via Groups.Io <ron.spencer= [email protected]> skrev: Interesting topics being discussed.
I was surprised to read that the RL from a short or open is 12.XXXdB. That
just isn't true, ever (assuming, of course, lossless feedline/ connectors).
Its 0. Always will be. Energy sent down the feedline to either the short or
open has "no where" to go so is 100% reflected.
I also read loss is always negative? When calculating power out of a
system sure. But when expressing it (such and such feedline has XdB of
loss) its always a positive number. Negative loss implies gain.
I see this often so I can readily understand how folks can get this wrong.
Professionals get it wrong. I saw one commercial (not for ham radio)
amplifier maker have a graph of return loss labeled in -Db.
All that said, I find the nanovna a very useful device. Much easier to
bring to the tower base than my HP 8753. I look forward to using it more.
Sent using
|
Yes, Jeff. My typo error. Should be GAMMA not RHO!
|
On Wed, Aug 21, 2019 at 03:21 PM, Ron Spencer wrote:
Return loss tells you how much of the input signal is reflected. Return loss
is the ratio between the reflected power and input power:
Hi Ron, Be careful here about mixing power definition into return loss calculations. Strictly speaking the Gamma value or reflection coefficient is defined as a ratio of (V_reflected/V_incident). Again, the magnitude of the value of Gamma provides the value of RHO that I posted earlier. In a matched system you could certainly bring power into the definition but what about systems where impedance are mixed? I think you will avoid problems if you stick to VOLTAGE throughout your discussions of return loss and S11. Note, S11 is just GAMMA and as such like GAMMA is characterized with a magnitude value and a phase value. And the magnitude of S11 is just RHO. Gee, I hope I have no typos again... Alan
|
I am on holiday so did follow the various threads recently.?A bridge or Directional coupler is the heart of any VNA. Strictly speaking if you are serious, the bridge or coupler ought to be characterised carefully before you make any accurate measurements. A standard Return loss bridge is a three post device. in others like the nanovna it is in fact a 4 port device. In, out, return mag and input phase. So in principle you should require S11, S12... A 4x4 matrix with 16 components. Only a masochist would want to go about doing all these measurements. FOR most common purposes it is also not necessary. Nevertheless the information may be necessary depending on what you are doing.?That 12dB inherent bridge loss for open/short loads is related to the above 16 parameters. It is not nonsense. Theoretically you can work out that it's 12.014.. dB but as frequency increases it's value increases, it never decreases. That is why we zero it out, but there is a price. Assuming your detector can get down to 92dB (dynanic range) then at dc you will have a directivity of 80dB but as? frequency increases this drops. The directivity is the maximum range, the largest RL value you can measure with the instrument, provided it is properly calibrated.? In some cases if you want to do S11, S21 in line measurements this calibration is non trivial. at his is why professional talk about 12 points and 16 point calibration procedures etc.?For causal work a 3 point is quite adequate, in fact the minvna uses only 1 point calibration and I am ok with it for casual antenna work.Incidentally a RL of 70dB corresponds to 1.00063 vswr or a reflection coeff magnitude of 0.00031.? If anybody claims that the nanovna can measure these values accurately at even 10MHz, as Dr Kirby has objected,? then I would like to see some hard evidence.Hope this helps with the discussion. I now rather look forward to my Prague appointment for the Mozart: Don Giovanni opera, in the Opera House where he first debut it to the world in 1787.Sent from my Samsung Galaxy smartphone.
toggle quoted message
Show quoted text
-------- Original message --------From: Warren Allgyer <allgyer@...> Date: 22/08/2019 01:00 (GMT+01:00) To: [email protected] Subject: Re: [nanovna-users] New to group and thoughts on Return Loss and Loss When using a spectrum analyzer/tracking generator/RF bridge to make S11 measurements, part of the setup is to normalize the results to zero with an open at the DUT port. This the the analog of the ¡°O¡± calibration of the VNA.A reading of ¡°-12 dB¡± or, in my case, -19 dB¡¡ is simply the inherent loss of the bridge and it must be normalized out if you need a direct reading of RL or used as an offset for the uncalibrated reading.Return losses of DUT at the far end of a coax line require calibration at that end, not at the device itself.WA8TODOn Aug 21, 2019, at 4:23 PM, Stuart Landau via Groups.Io <stuartl73@...> wrote:Return loss is a very old term used in the telephone industry for a very long time. There is confusion about the sign, but the negative has been used as long at the term return loss has been in use. Most of us understand the meaning and implications.The 12 dB nonsense came up in a 100% reflection measurement because of losses in the RL bridge, but it's meaningless.The reference for either an open or short is a return loss of zero. The bridge losses should only reduce the dynamic range of a measurement, not the accuracy.However, if you have any transmission loss between the RL measuring device and the 100% reflection, it will give you a better return loss than zero.If a coax has a 3dB loss for instance, it will show (in theory) a 6 dB return loss because the RF will travel twice through a 3dB loss (the coax).Stuart K6YAZLos Angeles, USA-----Original Message-----From: Hans J Albertsson <hans.j.albertsson@...>To: nanovna-users < [email protected]>Sent: Wed, Aug 21, 2019 11:33 amSubject: Re: [nanovna-users] New to group and thoughts on Return Loss and LossA loss is a negative gain.So, saying a return loss is negative simply means you are expressing thatloss as a formal gain.Don't overdo the besserwisserness of your insights.Den ons 21 aug. 2019 20:07Ron Spencer via Groups.Io <ron.spencer@...> skrev:> Interesting topics being discussed.> > > > I was surprised to read that the RL from a short or open is 12.XXXdB. That> just isn't true, ever (assuming, of course, lossless feedline/ connectors).> Its 0. Always will be. Energy sent down the feedline to either the short or> open has "no where" to go so is 100% reflected.> > > > I also read loss is always negative? When calculating power out of a> system sure. But when expressing it (such and such feedline has XdB of> loss) its always a positive number. Negative loss implies gain.> > > > I see this often so I can readily understand how folks can get this wrong.> Professionals get it wrong. I saw one commercial (not for ham radio)> amplifier maker have a graph of return loss labeled in -Db.> > > > All that said, I find the nanovna a very useful device. Much easier to> bring to the tower base than my HP 8753. I look forward to using it more.> > Sent using > > > >
|
FWIW
I am of the opinion that it would be more beneficial to users of this group who come here for understanding if they could build knowledge on what they read rather than wading through long debates and choosing sides in hopes of serendipitously building upon their understanding.
Would it not be better if:
1) We first stress that the NanoVNA simply makes only one or two measurements... reflection only or reflection and transmission, also called scattering parameters that we often reference as S11 or S22 and S21 or S12 respectively?
2) Point out that Each measurement is two valued (a complex number) representing a vector as a ratio of the measured value with respect to an input or stimulus value (also a complex number)?
3) Ensure that users understand that these two parameters embody the entire definition of what a VNA measures? The quality, stability, and repeatability of the measurements are all hardware dependent. If the hardware is up to the task, it simply provides an output of high quality, stable, and repeatable complex number pairs.
4) Note that everything beyond this point is the result of computationally comparing the output complex values (the measurements) to the input values (the stimulus)? This includes the process of calibration which yields the complex values representative of the stimulus which is both normalized, offset (corrected), and scaled.
5) Refer to the complex numbers that are output by a VNA (NanoVNA) as reflection coefficients (CH0), and transmission coefficients (CH1)? These same numbers are all that is required to compute any and all of the display options offered and supported by the NanoVNA.
6) Focus this forum toward growing our understanding and utilization of the VNA as an RF measurement tool, with do and don¡¯t do along with why and why not reasoning?
Difficulty grasping the concept of loss due to the presence or absence of a sign seems to signal a lack of understanding at a more basic level. Most... possibly alll... practicing tradesmen wouldn¡¯t expect to see 100W (50 dBm) out of a 20 dB attenuator being driven by a 1 Watt (30 dBm) transmitter. Conversely; they would indeed anticipate the output of that attenuator to be 20 dB lower because of their expectation that the attenuator has 20 dB of loss.
They may even attempt to confirm this by measuring the output to confirm the power is 10 dBm as expected. If they inserted an additional 20 dB attenuator, they would confirm that the output was now reduced by another 20 dB, for a total of 40 dB of loss, by measuring the output and observing a level of -10 dBm.
The subtlety is clear to those skilled in the art, but possibly not at all clear to those who come here to learn. For their benefit we need to be clear and identify the subtleties.
Loss as well as gain is often expressed as a logarithmic ratio called decibels or dB. A positive result implies an increase or gain, and a negative result always implies a decrease or loss. When the sign is preserved, the result is always treated as a gain. By convention, the sign is often dropped, and the result expressed as either a gain (when the result is positive), or a loss (when the result is negative).
A common error is the use of the reference dBs... as in 10 dBs vs 1 dB. By definition, the unit of dB is a ratio, and thus a singular value and has no plural notation or reference. The correct use is 0.1 dB, 1 dB, 5 dB, etc.
Any extensions to dB, such as dBm, dBW, dBv, etc. are simply indicators of a fixed normalizing value that the ratio is in respect to. For example dBm is read as decibels with respect to a milliwatt. In these cases the result represents both a relative value (ratio in dB), and an absolute value; a consequence of being referenced to a fixed level or standard, (1 Watt, 1 Volt, a dipole, etc.)
Another common error is a belief that a dB is different when expressed as a voltage ratio or a power ratio. In fact they are identical, and the decibel is ALWAYS expressed as a POWER ratio. Apparent power is assumed, since input and output impedances can be vastly different, and true power can only be assumed when input and output impedances are precisely identical, or when complex voltage and current measurements can be made. The computation of dB differs when comparing voltage ratios or power ratios, in order to preserve this definition. Voltage or current measurements must be squared for their ratios to equate to a power ratio.
To summarize; the only requirement for a minus sign to ensure there is no ambiguity in the example given here, was when expressing the absolute magnitude of the power measured as a ratio with respect to 1 milliwatt (i.e. -10 dBm), and only then, because the value is below the 1 milliwatt reference level.
In the end; wouldn¡¯t we feel sufficiently rewarded by simply striving to become productive users of our NanoVNA¡¯s, and adopt a terminology convention that closely matches the industry and enables us to share our knowledge efficiently?
We really shouldn¡¯t require jargon police debates while we can be discussing how to ensure we are making measurements correctly, and how hams that have only used SWR meters in the past can interpret their NanoVNA measurement results, and increase their understanding of radio beyond simply matching their antenna.
There¡¯s some good minds here and some eager newbies also. Knowledge is power, and we have the opportunity to revive the technical prominence of our hobby.
.... or we can just let it evolve into another philosophical special interest chat room. :-)
--
73
Gary, N3GO
|
Gary N3GO I deeply thank you for your message.
Not only I have learned more from it more than the dozen messages before, I also retained one of the last paragraphs you wrote:
"We really shouldn¡¯t require jargon police debates while we can be discussing how to ensure we are making measurements correctly, and how hams that have only used SWR meters in the past can interpret their NanoVNA measurement results, and increase their understanding of radio beyond simply matching their antenna."
This is 100% me.
"There¡¯s some good minds here and some eager newbies also".
Not sure about the mind, but I am an eager newbie related to VNA and I want to learn. I have tried it with a "dummy load" matchbox, but I guess I was ignored. I still don't know on how to interpret the results... But I will reach there.
(For the newbies out there I am sharing this simple explanation about the Smith Chart. Its a little easier to look at it ;) )
73 de Lu¨ªs CT2FZI
|
Dr. David Kirkby from Kirkby Microwave Ltd
On Wed, 21 Aug 2019 at 19:33, Hans J Albertsson <hans.j.albertsson@...> wrote: A loss is a negative gain.
So, saying a return loss is negative simply means you are expressing that
loss as a formal gain.
Don't overdo the besserwisserness of your insights. I agree with the original poster. The return loss of a passive device should be a positive number. There was an article about this by the editor of one of the IEEE journals. Dave. -- Dr. David Kirkby, Kirkby Microwave Ltd, drkirkby@... Telephone 01621-680100./ +44 1621 680100 Registered in England & Wales. Company number 08914892. Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United Kingdom
|
nanoVNA Message
There has been some great discussion on nanoVNA reflection and calibration topics. Perhaps the following could be of interest to those who would like to see a drawing or two with coupling instructions for various feed line measurements and db gain/loss calculations. Promise -- no complex math involved. If complex math was taught from secondary school and on after trigonometry, it would be in the long run, much easier for people to understand electricity and RF as really exists. BTW, 'complex math' is in reality not very complex but very intuitive and well worth the time to study it. It makes cabling and RF measurements innumerably simpler to understand rather than the funky single-phase volts and current that most RF techs and Hams are taught.
Compliments of WAVETEK: (800) 851-1198,
<- get a PDF manual here or search for: Wavetek 3ST
(or)
wavetek.org
Create a nickname and a single-use password for these sites to be safe.
Here is a small snippet of information in the WaveTek 3ST RF diagnostic manual (very similar to what the nanoVNA is trying to do) on correct reflection instructions and theory. The manual has a pretty good write up on how these devices like WaveTek 3ST and nanoVNA should be used and hooked up with impedance bridge taps and directional couplers for accurate calculations. The manual is free and available for a download in PDF in its entirety.
2.3 INSTALLATION
Proper installation of the Transmitter is essential to correct operation of the system. The transmitter is typically installed in a headend rack. There are two different options for setting up the system. One entails inserting the sweep at a standard combining network input, with a directional coupler (test point) at the output of the combining network.
Alternately, a directional coupler may be used to insert the sweep at the output of the combiner. A directional coupler is used to sample system signals in the headend, and an additional directional coupler is used to inject the sweep (see figure). The sampling directional coupler (DC a) may be of any value (dB), but higher values are typically chosen to limit insertion loss. The sweep insertion directional coupler (DC b) tap loss must be taken into account when setting the insertion level of the transmitted sweep signal.
[Diagram 2-1 - unprintable here]
The transmitter output level is variable from +20 to +50(1) dBmV in 2 dB increments. The output level will be attenuated by the tap value of the directional coupler. For example, if a sweep signal level of +16 dBmV is desired on the system and the output of the 3ST Transmitter is set to +36 dBmV, a directional coupler value of 20 dB may be used, or the combination of a lower value directional coupler and an attenuator pad.
[Combining Network Diagram did not paste here] Get the manual to see an example.
An alternate method would be to insert the 3ST Transmitter output at one of the combining network inputs, and connect the (transmitter) reference input to the headend amplifier input test point. This method eliminates the need to add directional couplers at the combiner output.
For detailed setup information, see "Stealth Sweep Start-up Procedure", page 4-1.
Hope this helps for those of you struggling to understand these topics. My hat is off to all of you not sitting in front of a TV but using your hands and brains.
--
Prof. Wm. Clark (Ret.) CS, BSEE, BSEET (KE7FO)
|
Great post William. This board as stated previously needs to be moved to a forum where we can have sub topics for discussions. There is so much information flowing through here that much of it gets list in the weeds. It can be overwhelming. There are very skilled individuals here with VNA usage and RF. If we had it broken down by topics then it would be easier to source and direct to the appropriate topics based on need.
Maybe we should look at migrating this over to a free slack community?
toggle quoted message
Show quoted text
On Aug 31, 2019, at 9:45 PM, William Clark <wclark@...> wrote:
nanoVNA Message
There has been some great discussion on nanoVNA reflection and calibration topics. Perhaps the following could be of interest to those who would like to see a drawing or two with coupling instructions for various feed line measurements and db gain/loss calculations. Promise -- no complex math involved. If complex math was taught from secondary school and on after trigonometry, it would be in the long run, much easier for people to understand electricity and RF as really exists. BTW, 'complex math' is in reality not very complex but very intuitive and well worth the time to study it. It makes cabling and RF measurements innumerably simpler to understand rather than the funky single-phase volts and current that most RF techs and Hams are taught.
Compliments of WAVETEK: (800) 851-1198,
<- get a PDF manual here or search for: Wavetek 3ST
(or)
wavetek.org
Create a nickname and a single-use password for these sites to be safe.
Here is a small snippet of information in the WaveTek 3ST RF diagnostic manual (very similar to what the nanoVNA is trying to do) on correct reflection instructions and theory. The manual has a pretty good write up on how these devices like WaveTek 3ST and nanoVNA should be used and hooked up with impedance bridge taps and directional couplers for accurate calculations. The manual is free and available for a download in PDF in its entirety.
2.3 INSTALLATION
Proper installation of the Transmitter is essential to correct operation of the system. The transmitter is typically installed in a headend rack. There are two different options for setting up the system. One entails inserting the sweep at a standard combining network input, with a directional coupler (test point) at the output of the combining network.
Alternately, a directional coupler may be used to insert the sweep at the output of the combiner. A directional coupler is used to sample system signals in the headend, and an additional directional coupler is used to inject the sweep (see figure). The sampling directional coupler (DC a) may be of any value (dB), but higher values are typically chosen to limit insertion loss. The sweep insertion directional coupler (DC b) tap loss must be taken into account when setting the insertion level of the transmitted sweep signal.
[Diagram 2-1 - unprintable here]
The transmitter output level is variable from +20 to +50(1) dBmV in 2 dB increments. The output level will be attenuated by the tap value of the directional coupler. For example, if a sweep signal level of +16 dBmV is desired on the system and the output of the 3ST Transmitter is set to +36 dBmV, a directional coupler value of 20 dB may be used, or the combination of a lower value directional coupler and an attenuator pad.
[Combining Network Diagram did not paste here] Get the manual to see an example.
An alternate method would be to insert the 3ST Transmitter output at one of the combining network inputs, and connect the (transmitter) reference input to the headend amplifier input test point. This method eliminates the need to add directional couplers at the combiner output.
For detailed setup information, see "Stealth Sweep Start-up Procedure", page 4-1.
Hope this helps for those of you struggling to understand these topics. My hat is off to all of you not sitting in front of a TV but using your hands and brains.
--
Prof. Wm. Clark (Ret.) CS, BSEE, BSEET (KE7FO)
|
Warren Allgyer
F4WCV
CT2FZI
