I had to test them out so the first thing I did was splice three 50' sections of cat5e cable I had together, and after the initial test their claim seems to be true. I tested every USB product I own, the number pad, a mouse, a joystick, and the Phidgets board, they all worked perfect. The only problem I had was with my web cam , the camera worked fine on a 25' section of cable but at 150' the laptop would recognize the camera but not display any video. I think it has to do with the web cam being so old though and not with the extenders, I need to find another web cam to test it out fully.

The next thing I had to figure out was how many wires of the cat5 cable the extender actually needed. (I asked a few companies that sell similar models what they required and they all told me it uses all of the wires.) Not believing that I started to clip one wire at a time and see what happened. After a little trail and error I figured out it uses 5 of the 8 wires. (I was hoping for only 4) This means for my application I need 5 wires for control, 2 for video which only leaves 1 extra wire. That means I will probably have to run an on board battery for main Rov power (which I thought about doing anyway)

The next test was to figure out the max range it would work at. I added the 25' section to the 150' cable and the board still worked fine. That was all the cable I had so I had to barrow a 200' box of cable from work, I spliced some ends on to that box and the board still worked. I was only shooting for a 200' tether so for me it works better than I though. Just for testing purposes I then added my 150' to that to see what happened. At 350' the board worked but only 1 function at a time, if I tried to turn 2 or more functions on at the same time the board would disconnect and my program would bomb out. I only tested the Phidget board at these ranges and I didn't try anything between the 200' and 350' range.

The last thing I wanted to try was to see if more than one USB device could be extended. I plugged the number pad which has 2 extra USB ports on it into the extender but as soon as I tried to plug the board into the number pad my computer gave me a error saying I exceeded the power requirements of my USB ports so that was the end of that tests.

UPDATE - After some further tests I found the extenders are in fact only reliable out to 150' so my tether will be shortened to that length.

My original tests where done on my laptop with it plugged into the wall with a power pack. During the firsts test the board worked fine with the 200' length of cable but when I recently tested everything out with the laptop running off the battery (which it will be out in the field) the board had a tendency to lose connection if the keys where pressed to fast in succession or if to many functions where on at once. I guess it can't push as much power to the board off the battery.

I've only done a few preliminary test on these but I had them working to at least 100' (it might have been 200' I tested them awhile ago and forget) on just 2 pairs (4 wires) of the cat5 cable. My future plans are to use a commercial Rov tether that has a cat5 cable plus a few other twisted pairs so I will be able to hit the 300' mark if I wanted too. (not that I will be able to afford 300' of it)

After much research I bought the above Velleman VM140 USB Interface Card. It seemed like a good choice with a total of 33 input/outputs: including analogue / digital and + 1 PWM. It was expensive at $170 but I've been wanting to try one of these for along time now. After sucking it up and buying one it shows up and it's a lot bigger than the posted specs of PCB dimensions: 195 x 142 x 20mm (2.7" x 5.6" x 0.8") Apparently someone a Velleman got there conversions wrong because it was actually (7.7 " x 5.6" x 0.8") even though I don't have a plan for the housings yet I know don't want to use a 6" Dia. Housing so I sent this one back. They do make a VM110 that is smaller but it has less features.