Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!bambam!hjp From: hjp@bambam.bedrock.com (Howard J. Postley) Newsgroups: comp.dcom.lans Subject: Re: Cable testers Message-ID: <157@bambam.bedrock.com> Date: 14 Jun 89 16:57:15 GMT References: <2021@wasatch.utah.edu> <99@rsiatl.UUCP> Organization: Ideal Point, Inc.; Marina del Rey, CA Lines: 65 In article <99@rsiatl.UUCP>, jgd@rsiatl.UUCP (John G. De Armond) writes: > In article <2021@wasatch.utah.edu> haas@wasatch.utah.edu (Walt Haas) writes: > >I'm shopping for a good way to test coax and twisted pair cables used in > >Ethernet, Arcnet and broadband. > >... > I've observerd the Cable Scanner in use and have used it a bit. My general > impression is that it is useful for determining the length of a known good > cable and for finding catastrophic failures (cuts, shorts, etc) but for > the problems we've had, it has been of little use. The catastrophic > failures can usually be observed visually (someone ran a chair over the > cable, etc). When the failures are marginal or intermittent, these little > pocket instruments leave a lot to be desired. A crushed cable or an > intermittent connector or an improperly crimped connector will not show > up on these boxes unless the condition is bad. This is 100% contrary to what I have found. We use our for exactly what is described above, and it works great! The only problem is that it tells you that there is a problem, but not in which direction so you have to move it to another test point. Crushed cables, bad crimps, and bad connectors and Ts are very hard to find visually. The cable scanner may ignore them on a first pass but if you adjust the tollerance threshold, it will tell you if your cable doesn't meet the spec for it's type, even if it's a perfectly good cable. The cable scanner knows what the impedance is supposed to be, what the tollerances are supposed to be. It has a toggle for teflon and PVC cable to accomodate for the minor differences between them. > What you have to do is determine whether the threshold of detection is > in the same ballpark as when the tranceivers quit working. I've found > little to beat the 'O-scope based units, especially the ones like the > Tektronics scope with the strip-chart option. You can literally > detect non-intrusive bending of a cable with this unit. Connectors - even > correctly crimped ones - show up clearly. And since you can make hardcopy, > you can build a history file of each cable. This history is the > absolute best way to quickly diagnose changes. You can also quickly > detect unauthorized new taps with little trouble. The cable scanner has a way to plug it into a normal o-scope to get this kind of information. The price difference is significant, to say the least. It also has an interface to a PC (and the software to use it) which can add histrories and a good bit of other information. > If your budget can handle it, get both instruments. If you can only > afford one, get the scope. It is more than worth the effort in > handling. Our budget could afford it and we found that the duplicity was totally unnecessary. We felt that our money was much better spent on an ethernet analyzer than on more cable analysis equipment. Tek makes some great stuff but the price you pay is based on something else, especially in this case. You could easily duplicate that kind of unit by integrating several pieces for less than 20% of the cost. Maybe closer to 10%. The thing is, you could afford to buy several cable scanners and have a bunch of people checking the various subnets of a new installation at the same time; that will also save you money. //hjp -- Howard Postley internet: hjp@bambam.bedrock.com Ideal Point, Inc. uucpnet: uunet!bambam!hjp phonenet: +1 215 578 6901 uspsnet: 13428 Maxella Av M/S 236; Marina del Rey, CA 90292