Cable Care

Click here to go to our companion page on connector care

Click here to learn how not to trash a cal kit

New for April 2013! The cables that are used on test equipment are expensive, fragile, and you cannot tell if one is damaged just by looking at it. Even though they cost one thousand dollars or more, in one day of futzing around trying to figure out why your test equipment doesn't calibrate, a new set of cables would have paid for themselves. Take better care of the new ones! Trust us, cables don't get better just by putting them back in the drawer. Do those holes in your old socks repair themselves?

It is shocking to see how cables are treated in some laboratories. They are bundled up and tossed into a drawer, like an ordinary AC cord, they are dropped, they have things dropped on them. Then you wonder why your best technician can't calibrate your test equipment after 12 tries.

Geppetto needs to clean up his VNA bench

In laboratories there is a certain division of labor which contributes to the overall problem of "bad data". Technicians take data, and engineers examine and interpret it. Without having to interpret data, technicians don't really know when they are measuring bad data. Most people just want to work as efficiently as possible, therefore a tech can generate a lot of bad data quickly and move on to the next task. Later the engineer struggles to make sense of it all, and pounds his/her head against the wall...

Here are some hints for cable care. Write to us if you have any additions or comments!

  1. If a cable is bad, throw it out. If you must save it for lesser duties than accurate measurements, spray paint it or mark it with a sharpy. If you have any enemies, another trick is to label the cable "good" and give it to them. If they come back pissed, you can always say "it was good when I last used it" which is not really lying...
  2. Never bend a cable more than 180 degrees, unless you really, really have to.
  3. Cables should be stored in their naturally-bent condition. A flat drawer or two should be reserved for cables, with nothing else in there to contact them.
  4. When you are taking VNA data on an RF probe station, tape down the cables into fixed position before you calibrate. Then DON'T TOUCH them. Don't bump them with your arm, don't drape them with your lab coat, don't retighten them, try not to breath on them. The cables should only move the small amount that the RF probes need to move in order to calibrate and measure. If you had the job of interpreting data, you would know this is not just a lot of hot air: without good data it is impossible to optimize a design.
  5. There is no reason for you to even touch the VNA test bench when measurements are going on. Yes, you have to turn the knobs on the positioners, but no, the bench is not there for your personal arm rest, even if it is mounted on an air cushion. Don't ever set down your torque wrench on the probe station platen, if it rolls around and falls on the probes you are in big trouble.
  6. Never attach a cable at one end and let the other end dangle. The weight of the cable is more than enough to damage where it is joined to the connector. This is especially true of armor-jacketed cables, even though they look bullet proof, they weigh a lot more than an ordinary cable and thus can put a lot of strain on the connector end.
  7. Don't drop a cable, and don't drop anything onto a cable. While we're on the subject, you should have a clean setup... if the space around your VNA looks like Geppeto's workbench, move all that crap to another bench. Check out an R-rated version of Pinnochio here.
  8. Always store cables with dust shields installed on the cable ends.
  9. You are not working at Sears automotive center changing tires, you have to get in touch with your inner, feminine side. Big hands must slow down, and heavy touch must give way to feather touch. If your test station is located behind the cables, you will have to use care to reach around the cables to press the buttons. A tighter-fitting lab coat can help here, but consider reconfiguring the station so that it is away from the cables and probe station.

How do you know is a cable is damaged? First, TURN OFF the calibration on your VNA just this once and pay attention to what the test equipment is doing before you try to correct it. Put the suspect cable into an S-parameter measurement, and look at S11, S22, and S21 magnitude and angle one at a time while GENTLY wiggling the cable. Try disconnecting and reconnecting it. If you see anything but stable (+/- 0.1 dB, +/- 1 degree) performance, you have a damaged cable.

Does your lab look like it should appear on the A&E show "Hoarders"?

This could be related to your calibration problems...


Author : Unknown Editor