Unstable data transfer rate 2°C

[Johan777]

I'm doing an experiment for a science project and decided to see what is the effect that temperature has on data transfers with an Cat5e ethernet cable. I noticed that at around 2°C the transfer rate gets unstable at about 85% of the transfer of a 100GB file. Could anyone tell me why this is happening, because this doesn't make any sense when you're thinking of the causes of resistance.

 

[kanewolf]

Do you have more information on how you are cooling the ethernet cable to 2C ? My guess is you are getting condensation somewhere. Are you using direct burial ethernet cable which would have more water protection than typical cable?

 

[Johan777]

Do you have more information on how you are cooling the ethernet cable to 2C ? My guess is you are getting condensation somewhere. Are you using direct burial ethernet cable which would have more water protection than typical cable?

I have placed the cable in a coolerbag that has air pockets in the sides to insulate the bag and placed in packs of frozen gel. I'm using a normal cable with no extra protection or insulation.

 

[kanewolf]

Here is a spec sheet from Belden on their riser cable -- http://www.belden.com/techdatas/english/1583a.pdf It is spec'ed to go to -20C. That tells me that they have tested their cables to those extremes. There has to be something about the way your experiment is being conducted that is flawed. Again, my guess is water infiltration because of the localized cold.

 

[Johan777]

Here is a spec sheet from Belden on their riser cable -- http://www.belden.com/techdatas/english/1583a.pdf It is spec'ed to go to -20C. That tells me that they have tested their cables to those extremes. There has to be something about the way your experiment is being conducted that is flawed. Again, my guess is water infiltration because of the localized cold.

It only happened at 2°C and 80°C not at room temperature(25°C), could it just be that the molecules from the wire are getting warmer energy from the air around it that it would have the same effect as if it was heated up?

 

[kanewolf]

Notice in the PDF I linked that 80C is above the rated specs, even for storage, for typical cat5 wire. Using wire out of spec says there are no guarantees of functionality.

 

[Johan777]

Notice in the PDF I linked that 80C is above the rated specs, even for storage, for typical cat5 wire. Using wire out of spec says there are no guarantees of functionality.

I know that it's above the specified working temperature, which is why it doesn't bother me and makes sense to me. But the 2°C is within the working temperature and the result of that is confusing me totally.

 

[kanewolf]

How many samples of wire have you tested with the same results? Is it a statistically significant sample size?

 

[Johan777]

How many samples of wire have you tested with the same results? Is it a statistically significant sample size?

I tested it with three wires a 1m, 5m and 10m cable and it happened with all of them.

 

[USAFRet]

Could be the shift between room temp at either end, and 2C in the middle.

 

[Johan777]

Could be the shift between room temp at either end, and 2C in the middle.

You may be right, I'll have to do more tests at other temperatures to see if the same happens there to confirm it.