why is SMA cable so expensive?

I have a D-link ANT24-0700 2.4GHz Omni-Directional 7dBi indoor antenna. It has a 6ft. cable that I want to add an extension to so I have more area to move it for a better signal. I checked various websites for prices of SMA cables and found even a short cable isn't cheap. For example

formatting link
wants $21 for a 10ft. cable.

Also, I was told by one website's tech support that adding more cable length will cause signal loss. Why is SMA cable expensive and will more length really degrade the signal to a significant degree? Thanks.

Reply to
sillyputty
Loading thread data ...

sillyputty hath wroth:

Hmmm... There's a bit of subterfuge in the specs. The antenna gain is 7dBi. However doesn't include the 1.5meters of RG-316 coax, which has a loss of about 1.1dB/meter or -1.65dB loss. What you really have is a 5.35dBi gain antenna.

Yeah, that's about right search for: "RP-SMA extension" to find other vendors and suppliers. Be sure to check eBay.

Yep. So do the connectors. Do the math: type Insulation dB/meter loss at 2.4GHz RG174 PE 1.20 RG316 PTFE 1.10 LMR195 Foam 0.56

If you add 3 meters of extension cable, the ADDED loss would be: type loss final antenna gain dB in dBi RG174 3.6 1.75 RG316 3.3 2.05 LMR195 1.68 3.67 Now you know why coax cables on external antenna are so short.

LG316 cable in bulk costs about $0.50/ft. RF Industries RP-SMA connectors are about $5/ea. So, you have a parts cost of about $15 per pigtail. I'm not sure how much to add for labor because building pigtails is really tedious and time consuming with all the cable end preparation and crimping. Add a few dollars for sweep testing the final assembly.

Do the math. 6dB is equal to half the range. 3dB is equal to 0.7 times the range. You're starting with a 7dBi antenna. I've added the

1.65dB loss that DLink conveniently ignored, and the approximately 2dB loss that you proposed to add, for a total of of 3.7dB coax loss. That's about a 0.78 times decrease in range. Not too horrible a loss. The improved location of the antenna will also be a benefit. I would say it won't hurt to add 3 meters more coax.

Incidentally, I'm not a big fan of omni antennas for improving range and performance. Look into directional antennas for far more gain.

Reply to
Jeff Liebermann

Thanks for the info! Gonna try a DIY directional antenna. :)

-- "Your estimated wait time is 8 minutes; your actual wait time may vary." MS tech support

Reply to
sillyputty

You could add a free reflector instead of a cable. Maybe instead of the

7db antenna.

formatting link
EZ-12, printed on photo paper for thick stock, with aluminum foil glued to the sail, provides a substantial boost in signal.
formatting link
The signal with the reflector is not only 13dB stronger, it's more stable.

Reply to
dold

formatting link
with

I made a 'windsurfer' reflector and did a few initials tests (turned it in different directions), but really couldn't tell if it made a difference, though I did pick up a couple signals I hadn't before. My question is the reflector is 4.5" inches high and my antenna is 11". Can I make a longer reflector to match the length of my antenna or does it have to be a certain size?

Reply to
sillyputty

I would say that you need to cover all of the antenna. Some portion of the assembly near the base might not be an active part of the antenna, but covering less than half of your antenna is probably not good at all.

The parabolic reflectors scale in size. If you simply enlarge the template, your curve will be larger, and the focal point of the parabola will be farther out, scaled up on the drawing of the support part.

In the "original" part of the

formatting link
web site, there is a separate template for a 5/8 wave whip, instead of the common dipole. I don't know if your large antenna is still a dipole. There are some pictures, and a link to a bunch more, at the bottom of the page.

Reply to
dold

Ok, thanks. I assume changing the size and curve of the reflector moves the focal point, and the focal point should be where the antenna is? Also, some designs use metal mesh. Is there any difference between metal mesh and tin foil?

Reply to
sillyputty

As long as the distance of the wires in the mesh is less than a quarter wavelength, then no significant difference.

But, lets say you have a 36" wide by 28" tall GRID antenna that looks like a BBQ grill, it will have a 29 dB gain where a smaller 24" full metal dish will have about the same gain.

When you get into the four and six foot dishes, the mesh/grid doesn't appear to be as strong. Another reason is ice build up on a grid antenna. For good measure, I keep my four foots dishes waxed to that ice slides off, but this isn't really a concern for most people as a four foot dish will retail for about $1,500.

I really hate using Google Groups, but appears my ISP isn't downloading the alt.internet.wireless anymore.

Reply to
nevtxjustin

Just a thought, one could take a hard rubber beach ball and cut it half, then cover the inside with aluminum foil. It would be more forgiving in feed horn placement (just aim the whole dish for best signal).

Reply to
nevtxjustin

Resizing the template gives a larger curve, and the focal point, as noted on the enlarged printed template, is moved out, so it is in the correct spot for the larger curve.

Mesh and solid are the same, as long as the mesh is small enough for the frequency that you are working with. The freeantennas page suggests a 1/4" mesh. The mesh should be closer than 1/10 of a wavelength, or almost 1/2" for 802.11b.

Noted on the

formatting link
home page, at the bottom, "Undoubtedly the best implementation of the Parabolic Template Design I've seen so far."
formatting link
a fine mesh parabolic.

Reply to
dold

A parabola is a different shape than a beach ball. There is a suggestion that a common wok or salad spinner is a better shape.

formatting link

Reply to
dold

snipped-for-privacy@92.usenet.us.com hath wroth:

I've been wondering where the 1/10th wavelength rule of thumb came from. So, I just wasted the last 30 minutes Googling for references and another 30 minutes digging through my text books. It's all over the place, but never really explained. As near as I can guess(tm) it came from the radio telescope designers, that needed a magic tolerance for the smoothness of the parabolic dish (i.e. deviation from perfect parabola). Same with optical telescopes, where the reflector has to be within 1/10th of the wavelength of light to be usable. Why 1/10th? I dunno.

However, that's flatness, not grid spacing. My guess(tm) is that the flatness specification magically morphed into a grid spacing specification, with little effort.

Next, I stumbled over to the junk pile, evicted the spiders, and measured the spacing on various welded wire dishes (mostly PacWireless). 2.4cm center to center, although it did vary somewhat from dish to dish and across the surface of the larger dishes. 2.4cm / 12.5cm = 0.19 = 20% So much for the 10% rule-of-thumb.

So, I dug out the antenna muddling program (4NEC2) and proceeded to fumble around with a simple parabola with various grid spacing. I didn't see a visible change to the gain pattern or phase distribution until the grid spacing was beyond about 0.6 wavelength. Even then, it was minor until about 0.8 wavelengths.

Next, I asked myself if it was better to have a very tight mesh (small grid spacing) but with high irregularity, such as found with trying to get welded wire mesh to conform to a parabola, or if was better to have wide spacing, but a better approximation of a parabola using much fewer wires. As expected, the dense, but irregular tight mesh had lots of strange lobes in strange places, but they were minor and did not affect the main lobe in the slightest. In other words, they're the same. (I'll guess I'll have to stop complaining about wrinkle finish aluminum foil reflectors.)

Going to the sacred online book of microwave antennas at:

I find that focal length accuracy is the most critical dimension, with feed illumination angle a close second. I won't delve on the details, but I should point out that these are two items that are being literally ignored by the typical parabolic reflector that uses an omnidirectional antenna (or a USB dongle) as a feed.

So what's the problem? Well, there is none. The errors introduced by focal length inaccuracy (i.e. sloppy dish construction) are perhaps

1-2dB at worst. That's fatal for a satellite dish or high reliability terrestrial link, but perfectly acceptable for a typical Wi-Fi system.

Feed illumination is a more serious problem, as much of the RF generated by the omni antenna (or USB dongle) goes drifting off in a useless direction. My guess was about 5-6dB gain loss, but only in the transmit direction. In receive, all of the energy that hits the dish, also hits the feed antenna. So, it's asymmetrical, but can be compensated for by cranking up the tx power.

Conclusion: Use 0.2 grid spacing (about 2.5cm) for now. Don't worry too much about precision. Try to build a proper matched feed for the dish, but if that fails, the omni or USB dongle will still sorta function. Any manner of gain is better than the stock rubber ducky antenna.

Let's do the math. Pretend we have a parabolic reflect wrapped around an omni antenna feed as in the various FreeAntenna models. The illumination angle of the dish is about 120 degrees at best. That's

1/3 of a full circle, so 2/3's of the tx power goes off to who knows where. Gain loss is: 10 * log (0.3333) = -4.77dB So, the tx gain of such an antenna will about 5 dB less than the receive gain.

I've never bothered to measure this one:

and probably never will. It is closer to a hemispherical section, than a parabolic section. It sorta works, but I did it more for a joke than for anything practical or useful. Also, the aluminum is very rigid and difficult to bend into a parabolic shape.

Nice. However, the large size JPG images are missing from the web pile.

Reply to
Jeff Liebermann

I think you'll find its simply to do with the best accuracy of early testing techniques. The Ronchi test (and derivants) will get you to

1/10th wave. As far as I remember, 1/4 wave will do for optical work (the Rayleigh Limit), anything less is a bonus and focus tends to be diffraction-limited.
Reply to
Mark McIntyre

shape.http://www.usbwifi.orcon.net.nz/ Well...yeah, they are two different shapes. I didn't say they were. I said you could use a beach ball (half of a sphere) for an antenna reflector, if not better than a wok as you really have no idea what kind of curve it would be.

Reply to
nevtxjustin

Gotta love those rules of thumb. I remember it from my radar days, but that's so long ago that all I remember is thumbs. I thought spacing larger than 1/4 wave was transparent, allowing radio to pass through the walls of a rebar-reinforced building, for instance. What's the spacing for insulators in a guy wire on an active radio tower? I can't remember if the spacing was for the those radio frequencies, or if they had to accommodate our radar frequency as well.

I looked at a picture on the pacwireless web site, counted bars, and did some math. I shrugged, 'cause I didn't like that answer ;-) I presumed that it wasn't a dish, but an array of parasitic elements, like a TV antenna. (If reality doesn't fit presumptions, change the reality.)

I know that model is a wonder of modern mathematics, but I don't always believe it. As you noted somewhere that I snipped, this isn't satellite, and lots of real garbage "works" for WiFi.

I've retired the ones that you noticed, but I did recently wrap foil around a CDROM case for a USB cantenna. It was kinda wrinkly.

I thought I looked at it when I posted the link, when I was trying to see the mesh size used. Ah, some of them are there, some aren't.

formatting link
shows the tiny mesh.

Reply to
dold

The Rayleigh Limit.

Reply to
Mark McIntyre

I just look around this thread.Wow,it is so old post I think ,and I would like to talk here.Maybe someone who will see here like me.

The SMA cables with 10ft cost USD21 ,Is it including the shipping cost or not? Sorry ,I can open the link the smacables,com

Are you agree to change the antenna ,fox example 10Dbi ,12Dbi or 15Dbi if you feel the singal loss greatly.The more length of the cable ,the more loss singal of course.

The price of SMA cable with 10ft is about USD9.9 from the Rfsupplier.com (I am not spam),the shipping cost is USD5/Order.It means that whatever you buy the cables or something,then the shipping cost is USD5/order.

See you later !

------------------------------------- Happy every day ! Good luck for all the people who are kindness!Enjoy the life and Keep happy in mind!

formatting link

Reply to
Rfkiki

formatting link
:

Not all cable is made the same. Price or performance pick one.

Reply to
NotMe

Just like bottled table water and cosmetics? You have been well trained.

Reply to
Anthony R. Gold

50+ years of real world experience as an RF engineer with access to a fully equipped test lab.
Reply to
NotMe

Cabling-Design.com Forums 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.