some atheros comparison

[mp3turbo]

x
x

[lonnie]

I suspect that at some future time truth will be required for specs, but for now the best we can do is compare in identical situations and see what they do. The best to date are the CM-9 units, by almost 10 dB on total link budget (tx and rx gains).

I am sorry that the big companies play with numbers, but in the long run it will hurt them. People do not forget being mistreated and the change in chipsets with no warning is another thing that will come back to haunt these guys.

[georgew]

Re: soldered VS crimped...

Ideally a well crimped connection will be hard to tell from a properly soldered connection. There are more quality issues surounding the mating end of the connector, because it is neither crimped nor soldered. Crimps can go bad, soldered connectors can get overheated and misaligned or sloppily prepared. You can also improperly assemble some brands of N connectors in a way that the male will permanently damage any female connectors it is mated to (the only "professional" pre-made cables I ever purchased were made this way)

Perfect cables are more about careful attention to detail during assembly, and the use of good quality parts.... and proper sealing from weather.

[mp3turbo]

OMG, george, you should see how BIG PIGS we are! We used flat file to cut back rg213 (!! no lmr600, just plain old good 2.4GHz rg213 cable, equivalent to lmr400), length around 2m on each side to get it to N connector... Soldering? Do not joke!! We had 30W soldering iron, it was not able to solder tin in outer environment, we forgot it at the other roof so we didn't have anything to solder on the first roof. What we did?

WE TOOK THE CABLE WITH CONNECTORS TO KITCHEN AND WE "SOLDERED" THEM ON GAS COOKER !! NO KIDDING, PRECISE 6GHZ TECHNOLOGY GOT THIS ATTACK. AND IT IS WORKING FLAWLESSLY...

[georgew]

Ok, first of all, you are using the wrong N connectors. You MUST use n connectors that properly fit the cable. Filing down the connector is causing your cable to become an active component in the circuit, and is probably reflecting a lot of signal back to the radio... this is bad.

The reason is simple, the higher the frequency, the less of the center conductor is used by the signal. The signal is not traveling inside the conductor, it is traveling on the surface of the conductor. If you scratch the surface of the conductor, some of the radio energy will see the scratch and reflect back toward the transmitter.

A good soldering job might correct some of the conductor damage, but cutting or scratching the center conductor will make the cable lossy in both directions. The loss is due to reflected energy, in other words it will increase your SWR and decrease the efficiency of the radio.

It is the same effect that you might expect if you put a sharp bend in your coax.

At high frequencies, the radio energy is mostly outside of the center conductor, it is only using the conductor in a manner similar to how a train rides a rail. If the rail suddenly changes gage, the train derails. A train needs a smooth ransition from one rail segment to the next. Radio waves need a smooth conductor surface in the same way.

Coax connectors are designed to crefully preserve the impedance of the cable... So they must fit correctly.

I know exactly why you had to cut the center conductor. However you can't do that with LMR400 because it is aluminum, so you can't even solder to it once you have cut it down. I assume the cable you used is pure copper, so that when cut down it could stll be soldered. I have had to do this in the past, when I was sold the wrong N connectors.

I don't know how much performance you are loosing this way. It might be 1%, it might be 50%. I have heard that a kink in coax can cost as much as 50%...