Is there any possibility of using coaxial cable as a microwave waveguide? That is, not using the inner conductor, which is grounded at both ends (as is the outer), but somehow guiding an appropriately polarised EM wave with the right wavelength through the dielectric?
So, a circular waveguide with a central conductive core and filled with coax dielectric.
Cheers
-- Syd
At very high frequencies, yes. But it will be very lossy. It will be a small circular waveguide full of lossy dielectric, with that wire inside. Analytically messy.
My first real summer job was in microwave spectroscopy. We had a long (40 feet roughly) waveguide that had a flat metal strip running down the middle, which we applied a high-voltage square wave to, to cause Stark splitting in the resonances of the gas in the waveguide. I mostly worked on the HV square wave generators.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com
How were you planning to couple to it?
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant
I'd use the best available method!
I'm just musing about using an existing long, lossy underground coax cable for data transfer. As I know nothing of waveguides, it seemed like a good place to start :-)
Cheers
-- Syd
snip)
Why do you insist on using a microwave carrier frequency? The coax is already optimized for RF.
To bad it's underground- if it weren't you could strip off the jacket and outer conductor, and use the rest as a G-line.
Mark L. Fergerson
The waveguide cutoff frequency will be really high, and the dielectric losses will be high, so it probably wouldn't go far. Electrical would be better.
-- John Larkin Highland Technology, Inc jlarkin att highlandtechnology dott com
The largest coaxes I have seen are about 0.5 m in diameter between some less than 1 MW HF transmitters and antennas.
The cut-off frequency would be in the upper VHF range anyway.
Coax cables can have waveguide modes -- they basically stop being useful as coax when you get above the cutoff frequency for the first non-"cable" mode.
That's all I remember -- E&M class was a long time ago.
-- Tim Wescott Wescott Design Services
Maybe you like to look at G-lines..
Jamie
What is the length of the cable, what cable type ?
On a continuous (no taps) coaxial cable, you should be able to use at least 256QAM on a single carrier, so 8 bits/symbol. Assuming 1 GHz bandwidth, you could transfer 8 Gbit/s, nearly as much as the 10 GbEthernet.
The DVB-C2 cable TV standard specifies up to 12 bits/symbol (4096QAM) for frequencies below 860 MHz. However, this requires COFDM and hence quite expensive equipment.
These...
...would be typical, although there is a wide variety. 10km long in a single piece. There's no insulation on the steel outer, and anything between a few metres and nearly all the cable will be wound tightly on a large drum/winch which winds in or out down a steel-lined hole.
Several hundred volts DC may be connected at one end (the surface) in order that some power gets to the bottom. Telemetry has to sit on top of this quite noisy line, and pretty much the best I've seen is 250kbps.
There is no possibility of using a different cable, what you get is what you get. There is no possibility of using repeaters or of interrupting the cable in any way.
Cheers
-- Syd
"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...
What's lossy about PE? It would be a lossy coax cable regardless if it were lossy dielectric. PVC cored coax, can you imagine? That stuff would go 'thud'. PE (and teflon, and foam) dielectrics are crisp more-or-less up to the cutoff frequency. (I don't suppose you have any coax that's big enough to see mode excitation from your 20ps step generator?)
As for coax modes, AFAIK, they are undesirable because, like all waveguide modes, they are dispersive. Along with cylindrical waveguide, there are degenerate TE/TM modes, which couple and propagate slightly differently, and mess up signals. Higher modes can alternate or spiral in various directions and polarizations, and are too closely spaced to be useful for much if any bandwidth (which occurs for any waveguide).
Tim
-- Seven Transistor Labs Electrical Engineering Consultation
Waveguides only work over significant distances if the wavelength involved is OTOO the cross sectional dimensions. If your coax dielectric is less than a few cm in diameter, your carrrier frequency is going to have to be 10 GHz or possibly much higher depending on the modes that the guide supports. Loss with any convenitional dielectric material will be prohibitive. You could burn a lot of effort on this scheme.
That's why I ask here first (ok, second after Google) thanks! I know some stuff, but there are a lot of holes.
I think I can safely rule it out as a viable method.
Cheers
-- Syd
Unfortunately, the manufacture did not give any specific attenuation figures, but at least the diameter is similar to RG-58, so it is going to have significant losses at higher frequencies. Running below 10 or
100 MHz the cable might be usable at the 10 km distance.Thus, I would expects that the 10 km cable could handle at most a few Mbits/s using some nnnQAM single carrier modulation.
That sounds a bit low even for that kind of (electrically) unspecified cable.
Keep in mind that ADSL will work to a few Mbit/s _or_ a few kilometers on twisted pair telephony cabling. Using real coaxial cables, you should be able to handle the 10km cable with at least some tens of Mbit/s data rates.
Unless you have severe power availability problems at remote site, you should be able to "tilt" the higher frequencies at the Tx site (within cable voltage limits), so that they appear on the receiver site at sufficient levels.
-110dB @ 100kHz. There's a lot of R and C.
Cheers
-- Syd
On a sunny day (Thu, 03 Jul 2014 13:02:13 +0100) it happened Syd Rumpo wrote in :
Is that thing going straight down? Ground penetrating radar?
I know on mars ground penetrating radar uses very low frequencies, IIRC just some MHz, and can reach very deep.
Else if horizontal just use radio, some of the free bands.
The bandwidth of a cable goes as 1/length**2, unfortunately. That 10km cable will have 0.01% of the bandwidth of a 100-metre cable.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant
At frequencies that propagate a circular waveguide mode in a reasonable-sized coax, any plastic will be very lossy. It's a cute idea, but probably not practical.
It would be a lossy coax cable regardless if it
We have a big delay line, something like 50 feet of 1/2 inch hardline, but it's hard to tell the mode limit from the regular losses at speeds like that. It's hard to propagate 20 ps edges any distance electrically. Singlemode fiber is amazing: the signal after 1 foot, or 1 kilometer, looks the same.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com
Sounds like you need a lot of transmit power and some really good adaptive equalization.
AT&T managed to distribute network TV signals all around the USA in the 1950's, largely over coax.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.