When to Use an Attenuator After an LNA?

[MDScanFan]

When I updated my railband and airband setups earlier in the year I applied 15 dB attenuation between my LNA and radio. I based it on trying to keep the net gain of the LNA minus cable loss less than 10 dB. I have very strong FM, pager, and NOAA signals in my area that cause intermod issues for some of my scanners.

This weekend I decided to run some attenuation experiments. I used an IC-705, BCT125AT, and TM-281 (rail only). In all cases I found the best weak signal reception with no attenuation. Dialing in several dB of attenuation generally had no impact. Dialing in 10 dB had a noticeable impact of either losing the weak signal or decreased S/N.

My airband and railband/marine setups use a similar configuration: narrowband antenna -> FM filter -> quality LNA (NF~0.5 & G~25) -> 100' RG-6 -> step attenuator -> radio. I explored using a set of 1 and 10 dB step attenuators before the radio to examine the impact of weak signal reception. I used weak signals of ATIS and NOAA transmitters for my assessment. I used two types of signals. The first type was a weak signal that was strong enough to break the squelch. The other type could only be received with an open squelch.

This made me question the use of the attenuator. Should I only consider adding attenuation if I notice an issue at the higher signal levels, such as intermods?

 

[iMONITOR]

In my 60+ years of monitoring radio I never had a need to use an attenuator even when using a preamp.

 

[merlin]

Step attenuation is really a valuable tool, I suggest-don't leave home without it.
(I'll get back to this)
Your problem sounds more like the lack of band pass filtering. (I been there)
Try to find a 150-175 Mhz bandpass filter for your VHF-railroad and another 118-130 for your aircraft band.
If your strong signal interference is in these bands, not a lot you can do. Strong enough, can be problematic for your LNA.
Back to attenuators:
My setup, a dicone on the roof. first is an FM broadcast trap then a 1.2Db pad. This goes into a 15Db LNA.
65foot of LMR600 coax(=) to the station. Next is a 1/2/4/8 Db attenuator into switchable bandpass filters.
This goes into a variable gain LNA, 0-20 Db. Another 1/2/4/8 Db attenuator.
last is another 1.2 Db pad, these have little/no effect on signals and the improvement in stage coupling makes up for loss.
One asset is 'seeing' the desired signal on a spectrum analyzer. A cheap RTL-SDR dongle and SDR# work for that.
You can then 'tweak' the attenuators for best signal to noise ratio.
Monitoring your RSSI is acceptable
You can split or switch to scanners and SDR from there.
LOTS of railroad near me and a repeater 900 yards away. Full attenuation(-30Db), 0Db gain on my bearcat still get all bars.
A site 58 miles and outside their coverage, I still get very usable signals. Weak, but good.

 

[merlin]

In my 60+ years of monitoring radio I never had a need to use an attenuator even when using a preamp.

You might be missing a lot of stuff buried in the noise floor.

 

[sonm10]

Step attenuation is really a valuable tool, I suggest-don't leave home without it.
(I'll get back to this)
Your problem sounds more like the lack of band pass filtering. (I been there)
Try to find a 150-175 Mhz bandpass filter for your VHF-railroad and another 118-130 for your aircraft band.
If your strong signal interference is in these bands, not a lot you can do. Strong enough, can be problematic for your LNA.
Back to attenuators:
My setup, a dicone on the roof. first is an FM broadcast trap then a 1.2Db pad. This goes into a 15Db LNA.
65foot of LMR600 coax(=) to the station. Next is a 1/2/4/8 Db attenuator into switchable bandpass filters.
This goes into a variable gain LNA, 0-20 Db. Another 1/2/4/8 Db attenuator.
last is another 1.2 Db pad, these have little/no effect on signals and the improvement in stage coupling makes up for loss.
One asset is 'seeing' the desired signal on a spectrum analyzer. A cheap RTL-SDR dongle and SDR# work for that.
You can then 'tweak' the attenuators for best signal to noise ratio.
Monitoring your RSSI is acceptable
You can split or switch to scanners and SDR from there.
LOTS of railroad near me and a repeater 900 yards away. Full attenuation(-30Db), 0Db gain on my bearcat still get all bars.
A site 58 miles and outside their coverage, I still get very usable signals. Weak, but good.

Gotta ask... what's a pad?

 

[900mhz]

pad = attenuator

 

[Ubbe]

You can then 'tweak' the attenuators for best signal to noise ratio.

That's the thing to look for, either by a digital bit error value or listening by ear to an analog signal to get the lowest possible background noise.

If you amplify and increase a weak signal you will usually first get lower noise then when adding more gain it will start to increase the noise level in the receiver, when it reaches the overload level. Then if the signal level are increased even more it will create intermod in the receiver, that it works as a mixer between different signal frequencies creating IF frequencies between them that pops up in the frequency band. When that happens you'll know that you have a way too strong signal entering the receiver. But long before that happens you will have a reduced sensitivity in the receiver. It's important to check the S/N relation when dialing in the total gain figure.

/Ubbe

 

[n2pqq]

Have you looked at Par electronics filters ?

PAR Electronics | Filters for the commercial 2 way market, MATV, FM broadcast, laboratory, marine industry, amateur radio, scanner and short wave listening enthusiasts

PAR Electronics Inc. Filters for the commercial 2 way market, MATV, FM broadcast, laboratory, marine industry, amateur radio, scanner and short wave

www.parelectronics.com

I had one made for my local pager system which causes intermod

I had it cut for 152.18 which is the frequency of the paging system
I also had one made up for my local NOAA weather station 162.500

I find the pager filter works best .

 

[merlin]

@ Ubbe Brings up digital, and getting the highest RSSI with lowest BER (bit error rate) you get the digital.
A pad is a minimal loss attenuator with resistors chosen to closely match say, a filter with 50 ohm impedance into a very low impedance gain device.
If you break down the front end on a Harris commercial radio, the front end blocks are like this:
Antenna In > low pass filter > pad > bandpass filter > pad > preamp MMC > pad >bandpass filter > pad > then into the first mixer.
This is not an uncommon design with most decent receivers and the chain is usually called a preselector.
Multi band receivers have switchable filtering like in many scanners, otherwise similar.
In the world of HF below 50 Mhz, this gets more critical for the noise and often more complicated.
Well engineered radios usually add upconversion so you have a tripple conversion superhet design.
73s

 

[vagrant]

My reception here is best with 10 dB or less gain. What my location mainly needed was several notch filters. I use the filters made by Dale Parfitt linked above. He will tune them to the frequency you tell him, so find your major offending pager and NOAA frequencies. I use two 152 MHz paging filters inline with the antenna at the house it is so bad (even without amplifier). I also use one in my vehicle. I also use his NOAA filters at home and in the vehicle. I also use the amateur paging filters with my radios at home and in the vehicle. (Yeah, it is really bad here)

 

[merlin]

Have you looked at Par electronics filters ?

PAR Electronics | Filters for the commercial 2 way market, MATV, FM broadcast, laboratory, marine industry, amateur radio, scanner and short wave listening enthusiasts

PAR Electronics Inc. Filters for the commercial 2 way market, MATV, FM broadcast, laboratory, marine industry, amateur radio, scanner and short wave

www.parelectronics.com

I had one made for my local pager system which causes intermod

I had it cut for 152.18 which is the frequency of the paging system
I also had one made up for my local NOAA weather station 162.500

I find the pager filter works best .

Yea, a very good source for filters. My first stop is K&L, then Par. If they don't have it OTC they will make what you need.

 

[prcguy]

I use preamps on some repeaters and there is a method to finding the best post attenuation needed to optimize reception. Since this is a receive only application I'll forgo the repeater transmit part.

You would need a signal generator or service monitor, a directional coupler with 20-30dB coupling value over the frequency range of interest, an adjustable attenuator in 1dB steps and a SINAD meter. Place the directional coupler through path in your feedline with the coupled port pointing towards the receiver and no preamp in line.

With the scanner or receiver in FM mode, generate an RF signal with 1KHz tone at 3KHz deviation into the coupled port and find the point where the receiver measures 12dB SINAD and record the level. It should be roughly XX dB higher than the receivers rated sensitivity plus the coupling value.

Next insert the preamp with variable attenuator after it between the directional coupler and receiver and set the attenuator for a little more attenuation than the preamp has gain. Generate roughly the same level that was needed to achieve 12dB SINAD before placing the preamp and attenuator in line and start removing attenuation until you reach 12dB SINAD again. Record the attenuation value. Now reduce the signal generator level a few dB to decrease the SINAD level and take out attenuation a few dB at a time to keep the SINAD meter at 12dB. Keep doing this until you get about 1dB change in SINAD for a 2dB change in reduction of attenuation.

You have now reached a good amount of attenuation to make the best use of the preamp gain at your location using your antenna and taking site (or neighborhood) noise into account. Any more gain will not benefit reception and will just contribute to overloading the receiver and possibly creating IMD. In my experience you will end up with around 5 to maybe 8dB of preamp gain and the rest of the preamps gain will be eaten up by attenuation.

Due to site noise I find you get less useable preamp gain in the VHF and UHF ranges compared to 700/800/900MHz and higher. For example I found about 5dB of preamp gain is the limit for some of my UHF repeaters and I was able to use 8dB of gain on a 900MHz repeater which actually increased sensitivity by 6dB. So if the bare repeater receiver was meeting 12dB SINAD at .20uv, its now meeting 12dB SINAD at .10uv with preamp and post attenuation and that is a very useful upgrade. The preamp used on that 900 repeater was about 18dB gain and taking out more attenuation or not using any would not improve reception at all and would only subject the repeater receiver to problems. Plus the attenuation between the preamp and receiver gives both devices a really good 50 ohm resistive match.

Knowing that less than 10dB of preamp gain would ever be justified, its Ludacris to start with a high gain preamp like 25-30dB knowing you will only need a fraction of that. And a higher gain amplifier will run itself into saturation or create IMD well before a more appropriate lower gain amplifier assuming the IP1 and IP3 specs are similar.

 

[MDScanFan]

Thank you for all of the replies. I believe I follow what is being described above. Regarding test equipment, I am limited to my ears and a spectrum scope on my radio to make my assessments.

The thing I still don't understand is that I get the best weak signal reception with no attenuation and I get a noticeable drop in weak signal reception reception when I dial in 10 dB attenuation. I would have expected the change to be insignificant given the NF is predominantly set by the LNA up front.

The LNA I am using has IP3 ~40 dBm which is not too bad. I do agree that the gain is overkill. When coupled with a PAR FM filter and a narrow band antenna I have not noticed any intermod issues. I do have a number of filters I can play with (pager band, noaa, etc). However, they are difficult for me to swap in and out for my setups and my gut feel is that are not a contributor in this situation.

 

[prcguy]

You mentioned the LNA has 25dB gain and .5dB NF with an IP3 of 40dBm. That's better than most but its also a whopping amount of gain. Are you feeding any splitters that would account for some loss between the LNA and radio(s)? For example a 4-way splitter would be similar to sticking a 6 or 7dB attenuator in line, reducing all signals to the radio(s) by that much.

I'll stick with my statement of roughly 5 to 8dB of preamp gain with a very low noise figure feeding a single receiver to be a good range and when you get up around 10dB or more gain your asking for trouble. For some locations even a little preamp gain is enough to cause IMD problems either in the radio or in the LNA if no preselectors are used.

Thank you for all of the replies. I believe I follow what is being described above. Regarding test equipment, I am limited to my ears and a spectrum scope on my radio to make my assessments.

The thing I still don't understand is that I get the best weak signal reception with no attenuation and I get a noticeable drop in weak signal reception reception when I dial in 10 dB attenuation. I would have expected the change to be insignificant given the NF is predominantly set by the LNA up front.

The LNA I am using has IP3 ~40 dBm which is not too bad. I do agree that the gain is overkill. When coupled with a PAR FM filter and a narrow band antenna I have not noticed any intermod issues. I do have a number of filters I can play with (pager band, noaa, etc). However, they are difficult for me to swap in and out for my setups and my gut feel is that are not a contributor in this situation.

 

[MDScanFan]

Good point regarding the splitter. In my tests described above I did not have a splitter in the path. However, during normal use I tend to have either a 2-way or 4-way splitter in the path. That 3 to 6 dB of loss should not have much of an impact per my tests above.

Looking back on my notes on my original setup I had around 20 dB of loss after the LNA (2 dB cable loss + 12 dB attenuator + 6 dB splitter loss) for a net gain of 5 dB. That is what I was running for a while. Per my recent testing I will take out the 12 dB atten and roll with that for a bit.

You mentioned the LNA has 25dB gain and .5dB NF with an IP3 of 40dBm. That's better than most but its also a whopping amount of gain. Are you feeding any splitters that would account for some loss between the LNA and radio(s)? For example a 4-way splitter would be similar to sticking a 6 or 7dB attenuator in line, reducing all signals to the radio(s) by that much.

I'll stick with my statement of roughly 5 to 8dB of preamp gain with a very low noise figure feeding a single receiver to be a good range and when you get up around 10dB or more gain your asking for trouble. For some locations even a little preamp gain is enough to cause IMD problems either in the radio or in the LNA if no preselectors are used.

 

[Ubbe]

I used weak signals of ATIS and NOAA transmitters for my assessment.

Keep in mind that AM signals are not suitable to use as a test subject as a receiver has an AGC circuit that reduces gain in all amplification stages in the receiver to stop it from overloading and clipping the AM signal. You can probably use a 40dB amplifier if you only use AM mode. A weak NOAA signal are perfect to use to dial in the VHF band.

The higher you go in frequency the less gain you'll have from the amplifier and the higher the loss in the coax, so you might be able to use the amplifier without any attenuation in the 800MHz band if it has just 10dB gain left in that frequency band. You have to look at its specification. You can always insert a 10pF capacitor in series with the signal after the amplifier to equalize the gain to be more flat over the whole frequency range. Either open up the amplifier and insert the capacitor there, and be careful where you do that if it's bias-T powered, or inside a splitter, or just use a small metal box.

/Ubbe

 

[dlwtrunked]

...
I'll stick with my statement of roughly 5 to 8dB of preamp gain with a very low noise figure feeding a single receiver to be a good range and when you get up around 10dB or more gain your asking for trouble. For some locations even a little preamp gain is enough to cause IMD problems either in the radio or in the LNA if no preselectors are used.

That may not be enough gain by my experience and see:

https://www.qsl.net/zl3dw/pdf/Masthead%20Mounted%20Preamplifiers%20and%20Noise%20Figure.pdf