Wiring question

I am looking at changing my wiring to CAT5e to increase the reliability of the signals in my 1-wire system. I saw a great deal on wire but it came with a bag of RJ45 connectors - too large for my 1-wire connection in the AAG station. I also have a lot of inexpensive phone couplers (RJ11) that would cost a lot to replace with RJ45 ones.

How are you guys running CAT5e? Are you terminating with RJ11/12? What’s the best method?

I ended up replacing all my outside RJ45 connectors with screw block terminals which eliminated many intermittant network related problems. Even thought I had the connections enclosed in a water proof box, moisture seemed to get in and corrode the connectors. On the weather station itself I removed the connector block and soldered the wires directly to it.

Inside connections are all using RJ45 connectors. Wire is all regular Cat5 cable.

I am looking at changing my wiring to CAT5e to increase the reliability of the signals in my 1-wire system.

Do you gain much benefit in using Cat5e cable? I don’t run 1-wire, but I thought it ran at 115kbps. If so even ordinary Cat5 is way over spec (good for 100Mbps). Cat5e just enhances Cat5 for short runs of Giga-bit Ethernet, i.e. 1000Mbps. So if Cat5e cable costs more than Cat5 cable wouldn’t you be better with that.

In any case with a much lower signalling rate on the cable, I wouldn’t expect the Cat5/Cat5e specs to really apply, especially if you’re not wiring it in the approved way using the normal connectors.

I don’t know what cable is appropriate and in fact don’t know why??

I can get 1000 feet of cable (CAT5e) for $70 Cdn.

My ONLY reason to go to the trouble of replacing my existing cabling (standard phone cable) is to improve the reliability of the signal. I periodically get flat-lined temps. During these problem times, the AAG weather station seems able to transmit wind speed and direction, but the temp sensor flatlines - as well as other temp sensors.

Two questions:

  1. if I switch to CAT5 will this likely fix my problem? (i.e. is cable my likely problem?)
  2. what CAT5 do I want (CAT5 or CAT5e?)
  3. why is CAT5 better than standard phone cable?

Thanks

Answers to your questions:

1 - Very good chance, you have to try it and find out.
2 - Don’t know
3 - Telephone cables are just twisted pairs, normally this is to stop cross-talk between wire and eliminate some interference. CAT5 is twisted pairs and overall shielding and is better at rejecting external interference, this enabling a cleaner signal and higher data signal rates without errors.

Bob

Comments to your answers

2 - Don’t know

5 and 5e are basically the same cable. 5e is better made and tested to meet a higher specification. The difference is unlikely to affect this application.

3 - Telephone cables are just twisted pairs, normally this is to stop cross-talk between wire and eliminate some interference. CAT5 is twisted pairs and overall shielding and is better at rejecting external interference, this enabling a cleaner signal and higher data signal rates without errors.

Actually there is no shielding in normal Cat 5 or 5e, they are just multiple twisted pair cables in a sleeve. A lot of “phone” cable, eg what you get at Radio Shack, is just 4 wires in a sleeve they are not twisted into pairs. “Real” phone wire is Cat 3 and it should be twisted into 2 pairs but the twist is pretty loose. Cat 5 is a much better made cable and specified for higher data rates.

The twist is what gives the cable its resistance to interference. In a balanced circuit it works extremely well. I don’t think one wire is a balanced circuit in which case you are really just getting the advantage of a shielding ground wire wrapped closely round the signal wire.

The twist is what gives the cable its resistance to interference. In a balanced circuit it works extremely well. I don't think one wire is a balanced circuit in which case you are really just getting the advantage of a shielding ground wire wrapped closely round the signal wire.

I don’t think there is any real advantage to a close coupled (twisted) earth/ground wire in an unbalanced circuit. Twisted pairs work in balanced circuits because the two wires as so close to each other that they pick up an almost identical signal from any interefence source and this allows the interference to be measured and removed.

Imagine a simplified system where one wire carries a voltage (say 5v), the other earth (0v). At some point in the circuit the wires pass close to a an interference signal of 1v. Since the two wires are very close to each other, they both pick up the 1v signal, meaning that the signal wire is now at 6v and the earth wire at 1v. At the receiver, a special circuit (differential amplifier) looks at the voltage on both wires. It knows the earth wire should be at 0v, but it sees 1v. It now knows that the interfering signal must be 1v, so it subtracts this from the value on the signal wire leaving the original 5v signal.

It’s more complicated in real networks because the signals are oscillating as, usually, is the interference, but the principle is the same. Another good example is RS232 (unbalanced) versus RS422 (balanced). A RS422 circuit can go a lot further than RS232 because the balanced circuit makes it more immune to interference.

Unfortunately I don’t know enough about one-wire, but it sounds unbalanced (otherwise it would probably be called two-wire!). If you want immunity to interference for a one wire circuit you either need something like coax cable (relatively expensive, but pretty good for outdoor use) or decent quality screened audio cable (perhaps less expensive, but likely to be more susceptible to UV and brittle fractures of the sleeve when exposed to the elements.

Hope this helps explain some of the theory.

Thanks all for your help on this. From what it sounds like, I may not realize any improvement with the CAT5 since the 1-wire isn’t a ‘balanced’ system. I am going to give it a shot anyway in the hope that it will help - just due to the inherent quality of the wire.

I’ll let you know how it works out…so we can add some practice to the theory! :lol:

Changing my 1 Wire system from telephone cable to Cat5e cable fixed all my problems I had where sensors would flat-line or drop out. :smiley:

Also using [patch] cable with built in foil wrap & drain wire enabled me to extend my serial cable between WM918 & PC to 50 feet. 8O :smiley:

Trying using Cat5e wire for USB cable enabled me to extend camera cable
to 65 feet. 8O 8)

How many sensors do you have, and how long are the cable runs? Are they all wired in one line?

Here’s what I have so far:

  • an AAG station with wind speed, direction and temperature. It’s located about 100 feet away.
  • pagoda with temperature, about 30 feet.
  • indoor temperature, 6 feet.
  • loft temperature, 5 feet.
  • lake temperature, about 50 feet.

All come back into a central location that ‘connects’ them together using an RJ11 gang box, then a single connection into a DS9097.

I am just about to install the AAG permanently up on an old TV tower, so any re-design ideas are certainly timely and welcome!

(I am also working to get a rain gauge and humidity.)

Lightning detector & solar sensor [temp type] are on one line about 70 feet.
Soil temp is on seperate line about 30 feet.

Before I did this in Cat5e the 1 Wire system would run OK for a while then flat line or drop out one or more sensors

I found a couple of application notes on this subject on the maxim website.

http://www.maxim-ic.com/appnotes10.cfm/ac_pk/1/ln/en

App Note 148: Guidelines for Reliable 1-Wire Networks

App Note 159: Ultra-Reliable 1-Wire Communications - this one states that the “star” (jeffcharger) wiring configuration is the least reliable.

thanks for pointing me to this. I will try to optimize my setup based on this article. Most of it is beyond my understanding - but it does offer some simple direction that I can adopt quite easily. I’ll report my findings after I work through the setup.