Some calculations about GAIN, LNA, ATTENUATORS, COAX...


Ruben Navarro Huedo (EA5BZ)
 
Edited

Hello friends:
My knowledge of RF concepts are medium to low.
I am learning a lot reading this group.
Here are some thinkings about my setup at this moment:

Using an Airspy R2.
Main use are frequencies from 130 to 150 Mhz (Mainly 137 SAT Frequencies).
I am using:
Antenna --> Around 6 meters CO22 Coax Cable--> LNA with HPF and BPF included for 137.5Mhz ( https://store.uputronics.com/index.php?route=product/product&path=59&product_id=94 ) --> 30 meters Ecoflex 10 Coax cable --> Airspy R2

Ecoflex 10 has an attenuation of around 5db in 100m (144mhz). Then i will have 1.5db of attenuation in 30 meters.
Lets estimate a loss in connectors and converters of around 1 db more ¿? ----> Then i have around 2.5db of total attenuation.

With this calculation i think my LNA has too much gain. (M
anufacturer talks about 23db gain).
Is this correct?
I will have best results with a 15db attenuator after the LNA.

I am having the best results using GAIN 10 and Linearity.
What do you think about all this?

Thank's a lot.



Eric Oyen <eric.oyen@...>
 

with that low a loss figure, an LNA is completely unnecessary. It will overload the receiver and introduce a lot of extra noise.

now, if you had 300 meters of coax between you and the antenna, then the LNA would be appropriate. sometimes, too much of a thing can be bad.

DE n7zzt Eric


On Nov 17, 2017, at 8:16 AM, Ruben Navarro Huedo (EA5BZ) wrote:

Hello friends:
My knowledge of RF concepts are medium to low.
I am learning a lot reading this group.
Here are some thinkings about my setup at this moment:

Using an Airspy R2.
Main use are frequencies from 130 to 150 Mhz (Mainly 137 SAT Frequencies).
I am using:
Antenna --> Around 6 meters CO22 Coax Cable--> LNA with HPF and BPF included for 137.5Mhz ( https://store.uputronics.com/index.php?route=product/product&path=59&product_id=94 ) --> 30 meters Ecoflex 10 Coax cable --> Airspy R2

Ecoflex 10 has an attenuation of around 5db in 100m. Then i will have 1.5db of attenuation in 30 meters.
Lets estimate a loss in connectors and converters of around 1 db more ¿? ----> Then i have around 2.5db of total attenuation.

With this calculation i think my LNA has too much gain. (M
anufacturer talks about 23db gain).
Is this correct?
I will have best results with a 15db attenuator after the LNA.

I am having the best results using GAIN 10 and Linearity.
What do you think about all this?

Thank's a lot.




Ruben Navarro Huedo (EA5BZ)
 

Tnx for your opinion Eric


prog
 

On Fri, Nov 17, 2017 at 07:16 am, Ruben Navarro Huedo (EA5BZ) wrote:
Hello friends:
My knowledge of RF concepts are medium to low.
I am learning a lot reading this group.
Here are some thinkings about my setup at this moment:

Using an Airspy R2.
Main use are frequencies from 130 to 150 Mhz (Mainly 137 SAT Frequencies).
I am using:
Antenna --> Around 6 meters CO22 Coax Cable--> LNA with HPF and BPF included for 137.5Mhz ( https://store.uputronics.com/index.php?route=product/product&path=59&product_id=94 ) --> 30 meters Ecoflex 10 Coax cable --> Airspy R2

Ecoflex 10 has an attenuation of around 5db in 100m (144mhz). Then i will have 1.5db of attenuation in 30 meters.
Lets estimate a loss in connectors and converters of around 1 db more ¿? ----> Then i have around 2.5db of total attenuation.

With this calculation i think my LNA has too much gain. (M
anufacturer talks about 23db gain).
Is this correct?
I will have best results with a 15db attenuator after the LNA.

I am having the best results using GAIN 10 and Linearity.
What do you think about all this?

Thank's a lot.
It depends on your antenna and your RF environment. Note that this particular LNA has a HPF at the input and a BPF at the output but it could still overload if the antenna has too much gain and the FM BC signals too strong.
Try comparing the SNR of a stable signal with and without the LNA. If it helps, then you can keep it.


Ruben Navarro Huedo (EA5BZ)
 
Edited

My antenna is a commercial 137mhz turnstile antenna with no gain. 2 phased dipoles with 8 reflector radials.

I think FMBC is not overloading the receiver.

i will try a SNR test using my VNA to generate an ON-AIR constant signal.

Thank's a lot for your help.


Ruben Navarro Huedo (EA5BZ)
 

Spectrum Spy


prog
 

Looks good. No signs of overloading and the SAW filter is doing its job.


Ruben Navarro Huedo (EA5BZ)
 

Thank's for your help.

I will continue using R2 for VHF.
HF+ will be used in a very remote (an very RF QUIET) location with a RX LOOP (Using LZ1AQ amplifier).
I will use, of course, spyserver and a raspberry pi.
If Network bandwidth is good we will make it public.


jdow
 

10 dB would be adequate to excessive depending on the noise figure of the AirSpy R2's tuner. If it's about 4 dB total noise figure at the antenna would be about 6 dB. Presume the preamp is 1 dB noise figure.

So 10 dB gain would reduce noise figure to 2 dB in a reasonable compromise with what it would do to IMD.

6 dB gain would have a noise figure at the antenna of about 3 dB with better IMD performance.

12 dB gain would give 1.6 dB.

15 dB gain would give 1.3 dB.

The full 20 dB would give 1.11 dB noise figure and "pretty darned poor" overload. Use this if overload is not a problem and you must have a low noise figure. 10 dB is a reasonable compromise and buys you a "decent" 4 dB noise figure improvement. 6 dB, if you have some high level signals, at least gets you a 3 dB improvement which is not to be sneered. 3 dB or more is often worth fighting for. 1 dB is worth fighting for when dealing with radio astronomy and deep space communications. Improving the antenna even slightly is usually a better way to buy 1 dB Eb/N0 improvement. Do it only when Eb/N0 is marginal to your needs.

Hopefully that will help you figure out your needs.

{^_^} Joanne

On 2017-11-17 07:16, Ruben Navarro Huedo (EA5BZ) wrote:
Hello friends:
My knowledge of RF concepts are medium to low.
I am learning a lot reading this group.
Here are some thinkings about my setup at this moment:
Using an Airspy R2.
Main use are frequencies from 130 to 150 Mhz (Mainly 137 SAT Frequencies).
I am using:
Antenna --> Around 6 meters CO22 Coax Cable--> LNA with HPF and BPF included for 137.5Mhz ( https://store.uputronics.com/index.php?route=product/product&path=59&product_id=94 ) --> 30 meters Ecoflex 10 Coax cable --> Airspy R2
Ecoflex 10 has an attenuation of around 5db in 100m. Then i will have 1.5db of attenuation in 30 meters.
Lets estimate a loss in connectors and converters of around 1 db more ¿? ----> Then i have around 2.5db of total attenuation.
With this calculation i think my LNA has too much gain. (Manufacturer talks about 23db gain).
Is this correct?
I will have best results with a 15db attenuator after the LNA.
I am having the best results using GAIN 10 and Linearity.
What do you think about all this?
Thank's a lot.


Eric Oyen <eric.oyen@...>
 

YW. :)

having been a ham for a number of years (25 currently) and an avid CB radio enthusiast and SWL before that, I have had to learn a lot of stuff about coax cables, connectors and proper matching to the line. receive matching isn't nearly as critical unless you want maximum transfer from the antenna to the line. Connectors do add some loss (depending on type). your typical UHF type connector (pl-259/so-259) add about .5 Db per connector. throw a barrel connector into that and you will have 1.5 Db loss at that point. Barring any other issues (such as line age, water infiltration, etc), a good system will get most of the signal from the antenna to the receiver with little noticeable degradation in signal. 

SOme exceptions do need to be noted to the above. An antenna that is substantially out of resonance will appear to give larger losses at the feed point. In the case of rg-58/u cable, that loss on HF can be rather large (to the tune of 12 Db or more). this is usually only for transmitted power from the line to the antenna and may have minimal effect on receive. I would recommend a little friendly reading at the ARRL website (more specifically the publications "ARRL antenna handbook" and "Antenna compendium". both have extensive information on coax types, VF figures, losses, etc.

DE n7zzt Eric

On Nov 17, 2017, at 9:08 AM, Ruben Navarro Huedo (EA5BZ) wrote:

Tnx for your opinion Eric


jdow
 

0.5 dB per connector?
1) Quit using UHF connectors. They suck dead bunnies through garden hoses, especially at, cough, UHF.
2) Install your nice type N, BNC, or TNC connectors CAREFULLY and precisely. (For higher frequencies than GPS consider SMA - or waveguide.

Good RF connectors installed properly should lose maybe 1% or 2% of your signal. A half dB sounds like somebody installed a UHF connector with pigtails.

{o.o} (Nearly 60 years in ham radio.)

On 2017-11-17 11:14, Eric Oyen wrote:
YW. :)
having been a ham for a number of years (25 currently) and an avid CB radio enthusiast and SWL before that, I have had to learn a lot of stuff about coax cables, connectors and proper matching to the line. receive matching isn't nearly as critical unless you want maximum transfer from the antenna to the line. Connectors do add some loss (depending on type). your typical UHF type connector (pl-259/so-259) add about .5 Db per connector. throw a barrel connector into that and you will have 1.5 Db loss at that point. Barring any other issues (such as line age, water infiltration, etc), a good system will get most of the signal from the antenna to the receiver with little noticeable degradation in signal.
SOme exceptions do need to be noted to the above. An antenna that is substantially out of resonance will appear to give larger losses at the feed point. In the case of rg-58/u cable, that loss on HF can be rather large (to the tune of 12 Db or more). this is usually only for transmitted power from the line to the antenna and may have minimal effect on receive. I would recommend a little friendly reading at the ARRL website (more specifically the publications "ARRL antenna handbook" and "Antenna compendium". both have extensive information on coax types, VF figures, losses, etc.
DE n7zzt Eric
On Nov 17, 2017, at 9:08 AM, Ruben Navarro Huedo (EA5BZ) wrote:

Tnx for your opinion Eric


doug
 

On 11/17/2017 02:14 PM, Eric Oyen wrote:
YW. :)

having been a ham for a number of years (25 currently) and an avid CB radio enthusiast and SWL before that, I have had to learn a lot of stuff about coax cables, connectors and proper matching to the line. receive matching isn't nearly as critical unless you want maximum transfer from the antenna to the line. Connectors do add some loss (depending on type). your typical UHF type connector (pl-259/so-259) add about .5 Db per connector. throw a barrel connector into that and you will have 1.5 Db loss at that point. Barring any other issues (such as line age, water infiltration, etc), a good system will get most of the signal from the antenna to the receiver with little noticeable degradation in signal.

SOme exceptions do need to be noted to the above. An antenna that is substantially out of resonance will appear to give larger losses at the feed point. In the case of rg-58/u cable, that loss on HF can be rather large (to the tune of 12 Db or more). this is usually only for transmitted power from the line to the antenna and may have minimal effect on receive. I would recommend a little friendly reading at the ARRL website (more specifically the publications "ARRL antenna handbook" and "Antenna compendium". both have extensive information on coax types, VF figures, losses, etc.

DE n7zzt Eric

On Nov 17, 2017, at 9:08 AM, Ruben Navarro Huedo (EA5BZ) wrote:

Tnx for your opinion Eric
Eric is far too pessimistic, and in a certain case, wrong. To wit:

At HF (i.e., below 30 MHz) the attenuation of UHF connectors is probably < 0.2dB. It might go to 0.5 dB at 144MHz. I personally HATE UHF connectors,
primarily because they are hard to mate properly (and some don't mate at all--some Chinese ones are made wrong) and it also bugs me that they are
NOT 50-Ohm matched. The point is that at HF and low VHF the mismatch does not cost much. K2RIW has a big paper on that--look him up!

Now as to cable loss: Cable loss does not depend on whether you transmit or receive, unless you transmit so much power that the cable dielectric
gets hot and melts, in which case, the loss will become infinite shortly. Think of the cable as an attenuator. It doesn't care which end is which, or
which direction the signal is moving. However, it will contribute a higher loss if connected to a mismatched load, so be sure your antenna has the
lowest SWR possible. The reason is that the resistive loss of the cable conductors will increase with a mismatch because the conductors will see
peaks in current at the maximum current points of the standing wave. However, a 12dB loss could only result in a very long cable with a high loss
at the frequency of interest combined with a very high VSWR at the antenna. This would be because of bad design right from the start! No-one
would ever design such a system! Note that the SWR mismatch will affect the total loss in EITHER direction. The cable is passive!
As to receive and transmit, assume the SWR is 1:1. Again, the cable is just an attenuator. As such, its loss at the frequency of interest will increase the
system noise figure by the effective attenuation of the cable at the frequency of interest. Suppose you have a 100 foot length of LMR-240 cable, and you want
to communicate on 145 MHz. Assume Type N connectors, which are virtually lossless at 145 MHz. The matched loss of this cable at 145 MHz is 3.5dB/100 feet.
If the noise figure of your 2-meter receiver is 2dB, then your system noise figure is 5.5dB. If you are transmitting 100 Watts, then your antenna will only
have about 45 W into it. Obviously if the antenna is mismatched, the system loss will be greater, in BOTH directions, transmit and receive!

--Doug, WA2SAY, retired RF engineer.


Ruben Navarro Huedo (EA5BZ)
 

Thank's for all replies !
Continue learning :-)


David Eckhardt
 

You are correct.  23 to 25 dB of gain is far too much for the AirSpy products.  Instead of using that preamp with filters, I'd suggest using a single J-310 'narrow band' preamp.  It will give  you about 12 to 13 dB of gain at the WXsat frequencies.  Thats plenty to overcome the cable loss and drop the NF of the AirSpy in your installation to around 2 to 2.5 dB.  The AirSpy products I've had and one I still have exhibit a NF around 4 to 5 dB.  That's plenty good for HF and likely even the WX sats.  However, a single J-310 preamp narrow banded design will do you far more good than a broadband 23 dB gain preamp.  I'd reference a www search on N6CA's U-310 preamp.  He presents necessary data for 6, 2, and 1.25 meters.  I believe the design is also covered in a number of the ARRL publications addressing amateur satellite communications.

Dave - WØLEV

On Fri, Nov 17, 2017 at 1:16 PM, Ruben Navarro Huedo (EA5BZ) <runahue@...> wrote:
Hello friends:
My knowledge of RF concepts are medium to low.
I am learning a lot reading this group.
Here are some thinkings about my setup at this moment:

Using an Airspy R2.
Main use are frequencies from 130 to 150 Mhz (Mainly 137 SAT Frequencies).
I am using:
Antenna --> Around 6 meters CO22 Coax Cable--> LNA with HPF and BPF included for 137.5Mhz ( https://store.uputronics.com/index.php?route=product/product&path=59&product_id=94 ) --> 30 meters Ecoflex 10 Coax cable --> Airspy R2

Ecoflex 10 has an attenuation of around 5db in 100m. Then i will have 1.5db of attenuation in 30 meters.
Lets estimate a loss in connectors and converters of around 1 db more ¿? ----> Then i have around 2.5db of total attenuation.

With this calculation i think my LNA has too much gain. (M
anufacturer talks about 23db gain).
Is this correct?
I will have best results with a 15db attenuator after the LNA.

I am having the best results using GAIN 10 and Linearity.
What do you think about all this?

Thank's a lot.





--
Dave - WØLEV
Just Let Darwin Work


jdow
 

U310? For quite some time that was my goto device. Then it became an invisible product when Siliconix went away and I had moved to GPS work from fast (7 kilohops/sec) frequency hoppers. (Gawd these SDRs have interesting potential in THAT field!) Siliconix had packages with matched pairs and quads. With them you could get amazing dynamic range with low noise figure and low power consumption. (The 2Nxxxx counterpart didn't seem to be as good. Siliconix seemed to have some nice secret sauce involved or something.)

{^_^}

On 2017-11-17 15:02, David Eckhardt wrote:
You are correct.  23 to 25 dB of gain is far too much for the AirSpy products. Instead of using that preamp with filters, I'd suggest using a single J-310 'narrow band' preamp.  It will give  you about 12 to 13 dB of gain at the WXsat frequencies.  Thats plenty to overcome the cable loss and drop the NF of the AirSpy in your installation to around 2 to 2.5 dB.  The AirSpy products I've had and one I still have exhibit a NF around 4 to 5 dB.  That's plenty good for HF and likely even the WX sats.  However, a single J-310 preamp narrow banded design will do you far more good than a broadband 23 dB gain preamp.  I'd reference a www search on N6CA's U-310 preamp.  He presents necessary data for 6, 2, and 1.25 meters.  I believe the design is also covered in a number of the ARRL publications addressing amateur satellite communications.
Dave - WØLEV
On Fri, Nov 17, 2017 at 1:16 PM, Ruben Navarro Huedo (EA5BZ) <runahue@gmail.com <mailto:runahue@gmail.com>> wrote:
Hello friends:
My knowledge of RF concepts are medium to low.
I am learning a lot reading this group.
Here are some thinkings about my setup at this moment:
Using an Airspy R2.
Main use are frequencies from 130 to 150 Mhz (Mainly 137 SAT Frequencies).
I am using:
Antenna --> Around 6 meters CO22 Coax Cable--> LNA with HPF and BPF included
for 137.5Mhz (
https://store.uputronics.com/index.php?route=product/product&path=59&product_id=94
<https://store.uputronics.com/index.php?route=product/product&path=59&product_id=94>
) --> 30 meters Ecoflex 10 Coax cable --> Airspy R2
Ecoflex 10 has an attenuation of around 5db in 100m. Then i will have 1.5db
of attenuation in 30 meters.
Lets estimate a loss in connectors and converters of around 1 db more ¿?
----> Then i have around 2.5db of total attenuation.
With this calculation i think my LNA has too much gain. (Manufacturer talks
about 23db gain).
Is this correct?
I will have best results with a 15db attenuator after the LNA.
I am having the best results using GAIN 10 and Linearity.
What do you think about all this?
Thank's a lot.
--
*Dave - WØLEV
*
*/Just Let Darwin Work/*


Eric Oyen <eric.oyen@...>
 

agreed on the UHF connectors. the problem is they are the most marketed connector out there.

btw, waveguide? any way you slice it, that is going to get expensive! Hardline of the 3/8ths inch variety would make a cheaper alternative (but at $3 per foot, its still insane).

now, other connectors that are lower loss start with the N connector. expensive but guaranteed up through 3 Ghz. BNC might make a more viable uhf type, so long as you don't go too far above 500 Mhz. many others are available, but not sure on specs or cost.

DE n7zzt Eric

On Nov 17, 2017, at 12:27 PM, jdow wrote:

0.5 dB per connector?
1) Quit using UHF connectors. They suck dead bunnies through garden hoses, especially at, cough, UHF.
2) Install your nice type N, BNC, or TNC connectors CAREFULLY and precisely. (For higher frequencies than GPS consider SMA - or waveguide.

Good RF connectors installed properly should lose maybe 1% or 2% of your signal. A half dB sounds like somebody installed a UHF connector with pigtails.

{o.o} (Nearly 60 years in ham radio.)

On 2017-11-17 11:14, Eric Oyen wrote:
YW. :)
having been a ham for a number of years (25 currently) and an avid CB radio enthusiast and SWL before that, I have had to learn a lot of stuff about coax cables, connectors and proper matching to the line. receive matching isn't nearly as critical unless you want maximum transfer from the antenna to the line. Connectors do add some loss (depending on type). your typical UHF type connector (pl-259/so-259) add about .5 Db per connector. throw a barrel connector into that and you will have 1.5 Db loss at that point. Barring any other issues (such as line age, water infiltration, etc), a good system will get most of the signal from the antenna to the receiver with little noticeable degradation in signal.
SOme exceptions do need to be noted to the above. An antenna that is substantially out of resonance will appear to give larger losses at the feed point. In the case of rg-58/u cable, that loss on HF can be rather large (to the tune of 12 Db or more). this is usually only for transmitted power from the line to the antenna and may have minimal effect on receive. I would recommend a little friendly reading at the ARRL website (more specifically the publications "ARRL antenna handbook" and "Antenna compendium". both have extensive information on coax types, VF figures, losses, etc.
DE n7zzt Eric
On Nov 17, 2017, at 9:08 AM, Ruben Navarro Huedo (EA5BZ) wrote:
Tnx for your opinion Eric


doug
 

On 11/17/2017 08:06 PM, Eric Oyen wrote:
agreed on the UHF connectors. the problem is they are the most marketed connector out there.

btw, waveguide? any way you slice it, that is going to get expensive! Hardline of the 3/8ths inch variety would make a cheaper alternative (but at $3 per foot, its still insane).

now, other connectors that are lower loss start with the N connector. expensive but guaranteed up through 3 Ghz. BNC might make a more viable uhf type, so long as you don't go too far above 500 Mhz. many others are available, but not sure on specs or cost.

DE n7zzt Eric

BNC is rated to 400 MHz. I have used TNC up to 1 GHZ on some MIL-STD hardware I designed. I don't remember its specified frequency rating.
I think the RF construction (as against the physical construction) of the two types (N or BNC) is not very different, at least when connecting either one to a coax like
LMR 240. For LMR 400, I would not like a conversion to BNC. If you need LMR 400, for most purposes you should stay with type N. BTW: BNC connectors definitely
leak RF at higher VHF frequencies. If you are concerned about RF leakage from a smaller coax coax connector, you might consider TNC. It's not very popular, it seems,
but it is a screw-type version of BNC, and is more RF-tight. If you have sensitive receivers and need to keep rf out, don't use BNC. This is a job for TNC or SMA.

(I used to use TNC and coax for medium voltage DC connections in my ham shack using 50 Ohm coax, back when stations were modular--say 1965.)

As a working RF engineer, in some 1 GHz RF designs, I have used TNC at around a KW peak rf power, with semi-rigid coax. Proved out and tested in Mil-Spec test routines.

Join my crusade against UHF connectors! Let's get them out of ham radio! They're over 75 years old, designed when the best RF engineers were just trying to get
their arms around COAX CABLE. I don't think the MIT papers were even out yet.

--doug, WA2SAY


jdow
 

If you play around you will notice that the mating surfaces of the type N and type BNC connectors are amazingly close to each other. BNC is specified to work to 4 GHz. SMA works much higher. But cable losses become ridiculous. So if you are fighting for that last fraction of a dB it's time to consider wave guide of one sort or another OR mount the GHz level electronics up at the antenna with conventional coax down to the rest of the electronics, much as you get with satellite TV.

{^_^}

On 2017-11-17 17:06, Eric Oyen wrote:
agreed on the UHF connectors. the problem is they are the most marketed connector out there.
btw, waveguide? any way you slice it, that is going to get expensive! Hardline of the 3/8ths inch variety would make a cheaper alternative (but at $3 per foot, its still insane).
now, other connectors that are lower loss start with the N connector. expensive but guaranteed up through 3 Ghz. BNC might make a more viable uhf type, so long as you don't go too far above 500 Mhz. many others are available, but not sure on specs or cost.
DE n7zzt Eric
On Nov 17, 2017, at 12:27 PM, jdow wrote:

0.5 dB per connector?
1) Quit using UHF connectors. They suck dead bunnies through garden hoses, especially at, cough, UHF.
2) Install your nice type N, BNC, or TNC connectors CAREFULLY and precisely. (For higher frequencies than GPS consider SMA - or waveguide.

Good RF connectors installed properly should lose maybe 1% or 2% of your signal. A half dB sounds like somebody installed a UHF connector with pigtails.

{o.o} (Nearly 60 years in ham radio.)

On 2017-11-17 11:14, Eric Oyen wrote:
YW. :)
having been a ham for a number of years (25 currently) and an avid CB radio enthusiast and SWL before that, I have had to learn a lot of stuff about coax cables, connectors and proper matching to the line. receive matching isn't nearly as critical unless you want maximum transfer from the antenna to the line. Connectors do add some loss (depending on type). your typical UHF type connector (pl-259/so-259) add about .5 Db per connector. throw a barrel connector into that and you will have 1.5 Db loss at that point. Barring any other issues (such as line age, water infiltration, etc), a good system will get most of the signal from the antenna to the receiver with little noticeable degradation in signal.
SOme exceptions do need to be noted to the above. An antenna that is substantially out of resonance will appear to give larger losses at the feed point. In the case of rg-58/u cable, that loss on HF can be rather large (to the tune of 12 Db or more). this is usually only for transmitted power from the line to the antenna and may have minimal effect on receive. I would recommend a little friendly reading at the ARRL website (more specifically the publications "ARRL antenna handbook" and "Antenna compendium". both have extensive information on coax types, VF figures, losses, etc.
DE n7zzt Eric
On Nov 17, 2017, at 9:08 AM, Ruben Navarro Huedo (EA5BZ) wrote:
Tnx for your opinion Eric


jdow
 

On 2017-11-17 17:46, doug wrote:
On 11/17/2017 08:06 PM, Eric Oyen wrote:
agreed on the UHF connectors. the problem is they are the most marketed connector out there.

btw, waveguide? any way you slice it, that is going to get expensive! Hardline of the 3/8ths inch variety would make a cheaper alternative (but at $3 per foot, its still insane).

now, other connectors that are lower loss start with the N connector. expensive but guaranteed up through 3 Ghz. BNC might make a more viable uhf type, so long as you don't go too far above 500 Mhz. many others are available, but not sure on specs or cost.

DE n7zzt Eric

BNC is rated to 400 MHz.
Slipped a decimal point there, keno sabe.

I have used TNC up to 1 GHZ on some MIL-STD hardware I designed. I don't remember its specified frequency rating.
0 to 11 GHz - chiefly due to the more rigid connection. Both N and TNC "die" about where the 3/8' and 1/4" cables become hybrid waveguide/coax assemblies.

I think the RF construction (as against the physical construction) of the two types (N or BNC) is not very different
In a severe pinch you can cut the outer shell with its bayonet construction off the male BNC and mate it to a female N. It works the other way, too, but the cutting is harder. (Kids, don't try this at home. This is for trained professionals only. And replace BOTH connectors ASAP. This is strictly a temporary setup. (No, don't ask.))

If you need LMR 400, for most purposes you should stay with type N.
Use N for power handling and physical connection strength. Use BNC if you have to made with an existing BNC connector. Use SMA if required. (annecdote warning) We discovered on the GPS prelaunch test receiver that you CANNOT assemble the versions of RG223U (double shielded 1/4") type N connectors that were available in the 70s with a low VSWR at GPS frequencies. Both above and below GPS frequencies it worked fine. We ended up switching to N to SMA bulkhead connectors with SMA to RG223U cable used inside the test set. The kludge had an adequate SWR, something like 1.1 vs 1.7 to 2 if I recall correctly.

BTW: BNC connectors definitely
leak RF at higher VHF frequencies.
Well, that varies with how straight you keep the connection. That's where TNC shines as it's more rigid when mated.

In all cases go to wave guide if the power involved exceeds the power handling capacity of the coax alternatives at the frequency of interest. (1 kW is not going to be happy in 1/4" cable at any frequency let alone 1 GHz.)

As a working RF engineer, in some 1 GHz RF designs, I have used TNC at around a KW peak rf power, with semi-rigid coax. Proved out and tested in Mil-Spec test routines.
Gutsy. For radar pulses I might do that. For continuous duty the cables would be room heaters until they melted. {^_-}

Join my crusade against UHF connectors! Let's get them out of ham radio! They're over 75 years old, designed when the best RF engineers were just trying to get
their arms around COAX CABLE. I don't think the MIT papers were even out yet.
1930s or so, I believe. (Wikipiddle more or less agrees with my memory.) Most pieces of equipment around here got the UHF connectors ripped off and N's installed. They mate just a whole lot better than the iffishness of the UHF maybe-connector. I might use UHF connectors for high voltage DC connectors with the SO239 on the power supply end of a hard wired cable on the power consumer end. Just don't spill a drink on it.

--doug, WA2SAY
{^_-} Joanne, who'd also like to modify all audio connections to differential pairs. Ban the mini-plug and mini-hack er jack.


Joe M.
 

I think you meant 4000 MHz (AKA 4 GHz).

<<https://www.amphenolrf.com/connectors/bnc.html>>

Joe M.

On 11/17/2017 8:46 PM, doug wrote:

BNC is rated to 400 MHz.

--doug, WA2SAY