Advice on getting a better FM signal for recording

[_pmc]

Hi all.

I installed a rooftop FM antenna and am starting to regularly record live FM music broadcasts received on a Marantz 112 tuner. I get a strong signal, but I'm also getting some sibilance ("essing") and occasional faint soft buzz-saw noise. I'm looking for advice on how I can approach improving the signal.

Here's my signal chain:

• Rooftop Stellar Labs Outdoor FM Antenna OMNIDIRECTIONAL -> 4-foot RG59/RG-6 coax -> 2-foot RG6 (through roof) to attic -> 2-way standard 5-2500MHz 75-ohm coax cable splitter to feed 1 and 2 -- each of which is a 50-100 foot in-wall RG6 coax run from the attic to different locations of the building
  • Feed 1 (computer): run from attic -> 12-foot RG-6 run to computer area -> coax/SMA adapter -> RTL/SDR Blog V4 -> USB -> Linux SDR++
  • Feed 2 (stereo): run from attic -> standard 75 to 300-ohm balen -> +/- fork-spade connectors -> 4" copper wires to FM antenna spring-pressure connectors on the Marantz 112 tuner (stereo, FM muting on)

The antenna is on a 6-foot pole, maybe 3-ish stories high. RG-6 barrel connectors through-out.

With the SDR, I measure for the station of interest at about -44dBFS strength and 42dB SNR -- which I beleive is pretty strong. I hear similar artifacts on on the SDR and can tune de-emphasis, bandwith, noise reduction, etc -- but the Marantz sounds better anyway. I record directly off the tuner's line-out to a SONY PCM-M10. To keep the audio pure/clean, I'm not going through a stereo amp.

Ideas I have:

• switch to F-connector or RG316 y-connector instead of one of those cable splitters
• add an FM antenna amp to boost the signal from the attic
• clean the tuner which is ancient...
• swap in another tuner to test (I have an NAD with digital tuning)
• Direct the line-out through a stereo amp and record the amp's output (It's a Pioneer Elite VSX-03TXH, but I have others I could try). Not sure why this would help.

I can provide audio sample if it helps. One other thing... if I wiggle the antenna wiring where they come into the tuner, I can hear signal quality change. I try to position it where it's quietest. I'm not sure if this is where I should target my energies or not.

Other ideas? Thanks.

 

[Punker X]

A couple of thoughts. Great that your invested in an antenna. The next biggest factor is alignment of your tuner. This can have the greatest affect on the performance of you tuner.

Good to remove any splitters. There's less loss in switches. The less connectors the better and a direct connection is the best. You may also want to think about an amplifier. Tunable with adjustable gain like Magnum Dynalab's work well.

I wouldn't characterize the Marantz 112 as sensitive or quiet tuners. Most Marantz sound great, but are lacking in the more advance features some other brands offered. The 2130 does have dual IF paths and some birdie filters that help with noise. There is an abundance of tuners out there that are more sensitive, quieter and sound good.

 

[_pmc]

Something else I've been chewing on...

I saw a thread on directly connecting 75ohm vs 300ohm with a balun. I'm doing the latter because the amp takes twin-leads -- but it sounds like no conversion is preferred. I'm thinking about building a short break-out (F to twin-lead) so I can eliminate that balun completely. It's also a balun I found in a box -- probably from a cable box from the late 80s.

Also, while inspecting the back panel, I realized that I've misread the labeling on the antenna connector (J001 below). It indicates that 300ohm uses the left 2 sockets, and 75ohm uses the middle 2. That's how I wired it. But I'm now realizing there's only one ground socket and the labeling is mis-leading. I rewired to use socket 1 and 3 (ground moved over). It seems better but I need to do a good listen to see.

Did I get this right?

 

[_pmc]

Great that you invested in an antenna.

It's a cheapie but seems to work well. Seems to very popular. Looks like a flying saucer hovering over the house.

think about an amplifier. Tunable with adjustable gain

I'll look into it. I didn't realize they made tunable ones. Pretty nifty.

remove any splitters. There's less loss in switches

What do you suggest? I need to run to 2 locations (ultimately 3). I know those standard cable splitters cause some signal loss -- which is why I've been thinking about fashioning some direct-connects. Never thought about a switch.

There is an abundance of tuners out there that are more sensitive, quieter and sound good.

I'll start by swapping in my NAD 4220 to see how different it performs. Maybe shopping for better / more vintage gear is warranted.

Thank you for the advice. I appreciate it.

 

[audiotemp]

Also, while inspecting the back panel, I realized that I've misread the labeling on the antenna connector (J001 below). It indicates that 300ohm uses the left 2 sockets, and 75ohm uses the middle 2. That's how I wired it. But I'm now realizing there's only one ground socket and the labeling is mis-leading. I rewired to use socket 1 and 3 (ground moved over). It seems better but I need to do a good listen to see.

The 300 Ohm connection uses the two left terminals. A 75 Ohm connection may be made to the two middle terminals as is indicated. The markings are correct!

Connecting to the 1st and 3rd terminals (counting from the left) is equivalent (same performance) to using the middle two (2nd and 3rd) as is marked unless there is some problem with the terminals themselves.

 

[_pmc]

The 300 Ohm connection uses the two left terminals. A 75 Ohm connection may be made to the two middle terminals as is indicated. The markings are correct!

Thank you. So I was right the first time. I'll switch 'em back.

The labeling still seems odd/conflicting to me. While the painted line indicators show it as you say (use 1&2 for 300-ohm, or 2&3 for 75-ohm), the labels seem strange: 75-ohm is FM&GND but 300-ohm is apparently FM&FM while the GND pin is left empty. I guess I'm showing my ignorance on how the circuit actually works.

Thank you.

 

[audiotemp]

It's one of those things that seems a little weird at first. Typically the 300 Ohm input goes to a transformer winding which has a center tap attached to chassis ground. The 75 Ohm input is between the tap and one (either) end of that coil. Since impedance of 1/2 of a coil is 1/4 that of the full coil, it works out to 300/4 = 75 Ohms.

 

[Bob@FM]

The labeling still seems odd/conflicting to me. While the painted line indicators show it as you say (use 1&2 for 300-ohm, or 2&3 for 75-ohm), the labels seem strange: 75-ohm is FM&GND but 300-ohm is apparently FM&FM while the GND pin is left empty. I guess I'm showing my ignorance on how the circuit actually works.

Way back in the 50's and 60's, most home roof antennas used 300 ohm cables to TVs and tuners. My recall was that as cable TV became more popular in the 70's, use of shielded 75 ohm cable became more popular and most home installations converted over to using that. So for a while, FM tuners were make to take both 300 and 75. Also many cable services had an FM option which was popular in cities.
Most FM tuner RF front ends are designed to use 75 ohms, unbalanced. So the most direct low loss path is to use the 75 ohm input connections. Each 75 to 300 balun has some loss. As mentioned above, the 112 tuner has an internal 300:75 balun after the 300 ohm inputs.
So when you go from 75 to 300 externally, internally it just goes from 300 back to 75 - and you just added the loss of 2 baluns for no good reason! 1:2 Splitters add loss to.
They may also be some RF noise coming out from the front end of the SDR USB tuner. I would run the test - detach it and see if it helps.

 

[_pmc]

the most direct low loss path is to use the 75 ohm input connections.

the 112 tuner has an internal 300:75 balun after the 300 ohm inputs.
So when you go from 75 to 300 externally, internally it just goes from 300 back to 75 - and you just added the loss of 2 baluns for no good reason!

Understood. I'll build a coax-to-twin-lead connector so I can connect directly at 75ohm. I used a balun for the physical conversion alone and incorrectly assumed either would be OK.

I'll try disconnecting the SDR when using the Marantz and see if that helps.

Thank you for the informative post. I had no idea cable companies offered FM like that at one time. Very interesting. We used to just use cut lengths of cheap speaker zip-wire for FM antennas back in those days -- and that was usually good enough for general listening. All these years later I'm trying to do hike it up a few notches in terms of audio quality -- and learn something through the process.

 

[Bob@FM]

In the 70's I always had an antenna for FM, but here's a discussion of FM on cable FM that time period -
FM on Cable 70s-80s
A 60" wide split speaker wire still works fine in many cases, depends on the location and signal.
For the noise problem - also try mono. If it is utterly silent in mono, the signal is strong, and noisy in stereo, try another station. Depending on the station's transmission (of HD/IBOC and level), noise with strong signals can be an HD problem that another tuner design will handle better and be silent.

 

[_pmc]

For the noise problem - also try mono.

I have noticed that oftentimes mono is clear / no noise, and enabling stereo adds the richness of channel separation, but at the expense of a field of subtle noise behind it. However, I tried it with my target station and didn't hear this effect. It sounds good both ways.

I built a short y-cable from a nice piece of RG-59 that has both foil and a steel ground braid -- and directly wired to the 75ohm inputs. I also dusted the inside of the tuner and cleaned the switches. I disconnected the SDR from the other leg running from the antenna just in case. I recorded a performance tonight as a test. So far, I can't hear an improvement from where I started -- and both the annoying sibilance and occasional interference/noise is still perceptable. Background noise may have increased as well. I'll know better when I get it into Audacity and can look at it and listen more carefully.

The next step for me is to try a different tuner -- and I think I'll also run through an amp instead of direct to the recorder. After that, I'll tackle the splitter and explore a tuner amp.

Baby steps...

 

[Bob@FM]

Your tuner may need a touch up on the alignment, either RF front end, or detector. With the detector, one little tweak can increase or decrease distortion significantly. With no test gear, you can clean the RF gangs (read how to do it properly though). Beyond that and cleaning switches or replacing caps, without experience and "know how" there's not much else one can do - if the problem is in the tuner.
You are on the right track to try another tuner or two. These units are getting to be over 50 years old, so without test gear, it's hard to say what the condition or state is.

 

[Bob King]

I have a very similar set up, same antenna, RG6 lead in that is about 25’ long. Signal strength seemed to overload the tuner, so I added a splitter to reduce the signal strength and it seemed to help. I still get the sibilance that I believe is multipath related- there is a water tower a half mile away at 90 degrees to most of the stations that I listen to. Need a more directional antenna!

 

[_pmc]

Signal strength seemed to overload the tuner

That's not something I considered. I processed last night's recording today, and I must say it sounds really good after a critical listen. I'm pretty happy. The host dialogue, however, sounded overloaded -- which is why I was worried while monitoring the recording. But the music itself did not have this characteristic. I'm now wondering if they had the gain up too high; distorting the sound on their end. That said, pre-recorded station-IDs and other DJ dialogue always seem to have that distinct essing.

Too much signal is not something I considered. I'm going to mull that one over. Thank you.

 

[KentTeffeteller]

A couple of thoughts. Great that your invested in an antenna. The next biggest factor is alignment of your tuner. This can have the greatest affect on the performance of you tuner.

Good to remove any splitters. There's less loss in switches. The less connectors the better and a direct connection is the best. You may also want to think about an amplifier. Tunable with adjustable gain like Magnum Dynalab's work well.

I wouldn't characterize the Marantz 112 as sensitive or quiet tuners. Most Marantz sound great, but are lacking in the more advance features some other brands offered. The 2130 does have dual IF paths and some birdie filters that help with noise. There is an abundance of tuners out there that are more sensitive, quieter and sound good.

I used to own a Marantz 112 for a short time. Many years ago. Was lacking in sensitivity, and tend to be noisier than I liked, and worked poorly in rural East Tennessee. I highly agree.

 

[dinger]

Even though the Steller Labs is an Omni, it still has a small amount of directionality to it and works best when the gap between the loops is towards the station of interest. You might try rotating it a bit to see if you can increase your gain or reduce mutlipath. You can also flip one of the semicircles and make it into an "S" shape to slightly increase the directionality.

 

[Bob King]

I have it oriented the opposite way, I will give that a try!

 

[_pmc]

Even though the Steller Labs is an Omni, it still has a small amount of directionality to it and works best when the gap between the loops is towards the station of interest. You might try rotating it a bit to see if you can increase your gain or reduce mutlipath. You can also flip one of the semicircles and make it into an "S" shape to slightly increase the directionality.

I had read that the connector-end should point to the target and that's what I did. That's the opposite side. I'll try a rotation. Interestingly, big city is to my south-east, but my target station is 180-degrees from there -- so I have to choose orientation. Thanks for the tip.

 

[_pmc]

fwiw, I tried a different tuner (NAD 4220) and got similar results. Listening in on the internet stream, I notice that when the host is speaking, the audio does not have that overloaded quality I hear on the tuners. My recordings are absolutely not clipped, but sometimes sound "pre-clipped" during station-ID and host dialogue (DJ).

Also, I might have noticed the occasional periodic interference buzz in the internet stream. i.e., that bit could potentially be in their source.

I'm also thinking that there could be some periodic interference from something in the house. I'm reminded how cell phone connections in the past could cause interference in my recordings if my phone was near the recorder (which I'm careful about now). It's a similar sound to that.

I'll keep tinkering with it...

Signal strength seemed to overload the tuner

This may indeed be happening.

 

[dinger]

Take a portable AM radio around the house to find sources of noise. Replace old fashion coil dimmer switches with the latest technology.