Many, if not most of us live in houses with lots of modern electronics, switching PSUs in wall warts, computers and controllers on all kinds of appliances and so on, all creating wideband noise that wasn't around in ye olde days. On top of that we play "radio" with a computer and usually other electronics in the same room, with the receiver only a USB cable length away from the computer. Then some of us have neighbors with the same amount of wideband noise stuff per household, adding up to the "noise cloud" in which we live. Quite often, the coax runs for considerable lengths through this noise, sometimes even in close proximity to the noise sources.
Even when your receiver is sitting on the window sill and the coax leaves the house after a few inches, it's still running through the "noise halo" that surrounds buildings. Old sources claim that said halo can extend "5m/15ft" beyond the walls, but given the considerably higher amount of noise sources today I'd say that distance should be doubled. So we can assume that even under ideal circumstances, the coax would be running for at least 5m/15ft through noise fields.
So what does that mean? I said above that the simple shielding of RG-58 gives you an attenuation of 45dB at best. The graph above indicates that it's even worse but let's ignore that for now. Now even if your noise doesn't exceed -70dBm anywhere on the length of the coax, the best you can hope for with RG-58 is getting this down to -115dB. IIRC the internal noise floor of the RSPs is around -130dBm, so there's still 15dB of added noise and not matter how good your antenna is/could be, a whole layer of otherwise clearly discernible signals will be covered by that noise. The remaining stations will suffer from a decreased signal-to-noise ratio and fading will have more impact on intelligibility, because - after finding the best receiver and antenna your money could buy - you skimped on the coax.
Of course, unless your QTH is in the ITU "rural/remote" category, your antenna picks up considerable amounts of QRM too and maybe you're more interested in listening to rather strong broadcasting stations, so this may not matter that much for you but if you're after DX and want to get the most out of your receiver, use proper coax! H155 is not cheap, but its shield is good for 80dB of noise suppression and its 5mm diameter fits SMA connectors nicely.
Some graphics in this blog post illustrate the problem a bit:
http://newsroom.bonito.net/why-even-goo ... e/?lang=en
Reason: No reason
In most cases sheat waves on the coax shielding are more evil. The only remedy against this is common mode chokes. In the vicinity of power lines the RF noise (inside and outside of houses) has a strong magnetic component. It induces noise in the coax cable. The best shielding cannot prevent this. In my garden any(!) long coax cable (low or high quality) connected to the RX without any antenna (only 50 Ohms terminated at the far end) receives high noise in the VLF and LW section. The shielding is not the problem! Only a common mode choke with high inductance reduces the noise on the cable.
But most of the noise is picked up by the antenna itself. In our environments many strong RF noise generators pollute all the air around the houses and cause interference within a radius of several hundred meters. Nothing can help against this noise. No shielding, no burying of cables in the ground, no common mode chokes, no grounding.
Dennis from bonito complains about RF noise from homeplug powerline modems (PLC) in his neighbourhood under the URL you quoted. Then he recommends coaxial cable with good shielding in this context. But he does not prove that a better shielded cable really helped against the high broadband noise levels such PLC modems generate. The reason is simple: The noise is picked up by the antenna, not by the coaxial cable. Well, over all: He wants to sell H155 and Hyperflex cable in his shop.
Of course I prefer cables with good shielding, no doubt. But they are not the ultimate cure of noise problems. One should not expect wonders. Most of the noise comes in through the antennas.
Reason: No reason
(i live in urban environment and I know all those kind of noise...)
I use coax I find from the local TV store... "Fte maximal k290 w", in their specs they say Screening factor (30-900 MHz) > 90 dB
maybe it is a little optimistic
What about lan cable? Any experience in that? I happened to read something about twisted pair for rx antenna feed.
In a 5mm cable you can have 4 twisted pair... it could do the work of 4 coax...
thanks in advance
Reason: No reason
Well, I didn't mean to write a book here. I just felt like elaborating on why exactly RG-58 is bad m-kay, because it came up in some other thread.DanubeBCL wrote:Good shielding of the coax cable is only one small part of the story.
Actually I had a follow-up post about common mode noise saved as draft. However, dealing withDanubeBCL wrote:In most cases sheat waves on the coax shielding are more evil. The only remedy against this is common mode chokes.
common mode issues does not make much sense when your coax is quite permeable in first place. Bad shielding means direct noise pickup by the center lead (no way to cure this except using better coax), common mode currents mean noise radiation by the shield, which the antenna will pick up (can be remedied by chokes, buried feeder line, grounding... or probably not using coax at all). Even if someone's antenna situation and surroundings are excellent, both bad coax and common mode currents will spoil the results unnecessarily, but there are remedies for both.
Of course I feel sorry for your situation but I feel there's some bitter over-generalization in that statement. People in this forum live all over the world, each in a highly individual reception situation, and not all of us are in such a bad situation. Of course you are right, local "propagated" noise (as opposed by near-field noise) is hard to fight, no matter if the source is 100m or 100km away. But (hopefully) for the majority of people here, that noise floor is below the near-field noise level and once the near-field noise is properly treated, it should be true again (and a good thing too) that all of the residue noise is being picked up by the antenna in form of regular, "propagated" QRM.DanubeBCL wrote:But most of the noise is picked up by the antenna itself. In our environments many strong RF noise generators pollute all the air around the houses and cause interference within a radius of several hundred meters. Nothing can help against this noise. No shielding, no burying of cables in the ground, no common mode chokes, no grounding.
True, but nobody claimed anything like that. My point is that for us SDR users, bad coax is one ultimate source of noise problems.DanubeBCL wrote:Of course I prefer cables with good shielding, no doubt. But they are not the ultimate cure of noise problems.
Which...again...cannot be generalized that way. However, the one common issue we all really have is more or less intense noise in the shack, because even in best case we still use a computer-based receiver, usually powered by a crappy switching PSU, which gets its power from a working grid and realistically we usually have some more noise sources in the shack, and very few of us live alone in a house with no neighbors. So without good shielding, all efforts to reduce noise may be futile, that was my point about single-layer shielded coax specifically.DanubeBCL wrote:One should not expect wonders. Most of the noise comes in through the antennas.
Reason: No reason
No experience with that, sorry. Since I use the RSP on UHF too (with the same coax and often even with the same dipole), I'd fear the increased loss, besides practical considerations (I buy my coax with SMA connectors fitted ) and having to think about the impact of impedance and antenna symmetry. But symmetric feed lines are generally interesting re noise and dipoles.CaptainNemo wrote: What about lan cable? Any experience in that? I happened to read something about twisted pair for rx antenna feed.
Reason: No reason
There's other factors that influence this.
If it was ladder line or a twisted pair instead of coax it wouldn't have any "coax shielding" yet we all know it still has just fine noise immunity because the wanted signal is present as a differential between the two wires. That same differential exists between the inner and outer of coax, the "good" RF is present on the braid and on the inner as equal and opposite electric currents, and as per the ladder line and twisted pair no electric field is created by the good RF.
The bad RF, AKA "noise" exists as an RF differential against ground, it sees a run of coaxial braid as an antenna driven against ground, and it does generate an electric field. That field is radiated all along the braid and if that braid is directly connected to one half of a dipole antenna that RF also radiates from that antenna, which is not a nice thought.
I also sometimes overlook the fact that RF does not radiate "evenly" along any given length of conductor; the amount of radiation at any given point is directly proportional to the current flow at that point. Eg. the maximum RF radiated by someone transmitting on a 1/4 wave vertical is at the bottom and slowly diminishes to zero at the top of that 1/4 wave, bottom loading coils need a much lower inductance than centre loading coils and top loading coils do nothing.
One thing that struck me about the Cable RF pickup post below is that the length of the coax mentioned is 3 feet and the strong unwanted signal is 105MHz. It turns out that 3 feet of RG58 is pretty close to a half wave long at 105MHz and as we know the end of a half wave is at a very high impedance and only requires a few picofarads of capacitance to couple extremely effectively to gobs of RF on any nearby high impedance point. A good example being "through the windscreen" antennas that rely on that to function and they function very well. I'm not suggesting I know for a moment how that might be relevant to the problem Paul314 has but it's interesting.
Hence we need to be thinking about what frequency we are interested in and also realise that noise at any particular frequency will create a series of high impedance "nodes" on your coax braid that can couple so well to other conductors in their vicinity and even more importantly, vice-versa.
OK, sorry to present such a plethora of disjointed comments, it just makes me ask questions of myself that I can't answer like how much braid do we really need and what the heck is really happening in my radio shack!
Reason: No reason
I've experimented with it, if it can conduct multiple gigabit square waves over a hundred meters - and square waves are really just sine waves with a very large number of harmonics or they aren't square enough to present a decent edge to the logic at the ends - it must be good stuff!CaptainNemo wrote:What about lan cable? Any experience in that? I happened to read something about twisted pair for rx antenna feed.
In a 5mm cable you can have 4 twisted pair... it could do the work of 4 coax.
All I can say is that I was disappointed, my first experiment a couple of years ago was simply to transmit over a short length of it and measure the loss, and the loss was unacceptable. When I finally saved up for a RigExpert AA-1400 antenna analyser which is good for almost DC to 1GHz within a few dB I revisited it and it is all over the place, presenting different impedances at different frequencies.
I still was very attracted to the principle of it though and now use teflon twisted pair for my main two large 160m loop antennas. I bought the teflon on eBay, it's quite a large gauge silver plated conductor and thick insulation and I keep the number of twists down to about 1 per meter. It has an impedance of around 120 ohms.
There is one excellent use for CAT5 though, if you extract a single pair from a short length of scrap, it's extremely good for winding common mode chokes on ferrite toroids and tubes, even for TX on 100 watts, as long as it's teflon, as are the tails of CAT5E I've cut up.
Reason: No reason
I use CT100 type satellite cable for most of my listening station. It's cheap, flexible and low-loss from LF to SHF. Depending on where you get it from (cheap Chinese stuff or 'proper' stuff from respected suppliers) it's well-made and has foil and braid screening. It's 75 ohms but that shouldn't matter as most of our antennas and receivers have pretty ill-defined impedances anyway.ve3tkb wrote:How about using 75 ohm RG-6 cable instead of RG-58? It seems to work great on VHF/UHF reception on my rsp-2 pro. Would it work as well on HF?
Reason: No reason