Ethernet, can distance be increased by raising voltage ?

[ptricks]

I have a buried ethernet line that is about 600ft long that I can only get to sync at 100Mbit. Wifi is not possible because this connection needs to be wired for work reasons (security concerns).
The cable that is buried is shielded , gel filled and 23AWG solid copper if that matters.
I've tried different switches and routers all commercial quality stuff and that doesn't seem to matter.

I saw extenders are sold that are poe powered and can be buried in ground mid point at 300ft or so to boost the signal but I really don't want to have to dig this
line up as I am not sure exactly where it is buried, and prefer not to bury a new one due to numerous trees, roots, irrigation plumbing, etc.
Is ethernet limited in distance because the voltage drops are too high as the distance increases or because of crosstalk between the pairs that causes too many errors.

If it is voltage related , would it be possible to build a circuit that on end takes a normal ethernet signal , converts that voltage higher, and then a circuit
on the other end that drops back to normal ethernet levels ?  Or am I just asking for something that will not work  and I need buy up some fiber and do some serious digging ?

[nctnico]

I suppose the cable is rated for CAT5 which is 100Mbit maximum. What you could do is run two 100Mbit/s connections in parallel as 100Mbit uses 2 out of 4 pairs. But at 600feet is already nearly twice the maximum recommended length for 100Mbit and maybe running two connection is parallel is pushing it over the limits.

If you need more bandwidth, then installing a fiber is a much better option. Using a pipe this time might be a good choice so you can pull new wires through it when necessary. Something like 2" in diameter.

[Andy Chee]

The issue isn't voltage.  The issue is likely signal reflections.  You're already doubling the maximum specified distance for ethernet.

If the cable is buried in conduit, then you should be able to use the existing CAT6 cable to "pull through" the new fiber cable.

[madires]

The maximum segment length for TP is 100 m (328 ft). The main reasons are attenuation (also crosstalk and other cable parameters) and timing constraints in the ethernet protocol. The most reasonable solution would be to go for multimode fiber. The second best (without digging) is to get a pair of VDSL or G.Fast modems (master and slave) used as an ethernet bridge.

[ajb]

How much bandwidth do you actually need?  Is 100Mbit inadequate?  I'm actually a bit surprised it even links at 100Mb at 600ft.  Do you actually get a good link, IE, are you able to actually send data over it without a bunch of lost packets?

madires' suggestions are good.  If you need higher throughput and have line-of-sight, there are point-to-point microwave systems that might work.  I don't know if that alleviates your security concerns about wifi, but at least any potential attacker would need to be roughly in sightlines rather than just in range of an omnidirectional wifi station.

[BennoG]

also consider that if you have a lightning strike near one of the buildings then you will fry all connected devices because of the potential difference between both buildings during the strike.

Benno