Calculation of the spped of a cable

By speed, you mean the propagation rate of wave along the cable?

I seem to remember it's a function of the permeability and permittivity of the the dielectric. Does the datasheet give those?

Sylvia.

Go here and download HP APPCAD

I hope your data sheet has the dielectric specs

Steve Roberts

"Enes Erdin"

** Co-axial cable velocity ( of propagation) depends on the type of di-electric material used.

Egs: % c

Polyethylene (PE) 66% Teflon 70% Foam 78....86%

The actual velocity with foam varies because of the ratio of gas to plastic varies from one example to another.

Similar numbers apply to twisted pair data cables, but vary with how the cable is installed.

The basic formula is v = 1 / sq.rt. mu

where "mu" is the " relative permittivity " of the material

and v = velocity compared to c.

..... Phil

ic

There is a book by Gayle MIner who is a perfessor someplace in Utah. The title of the book is: "Lines and Electromagnetic Fields for Engineers"

In the back, Appendix B, he has laid out all the info that you need. RG/U 8 up to RG/U 325.

Jacket material, type of connector, OD ID etc

HTH, Dave

For low loss cable, there is a rule that uses the Z0 and the mechanical dimensions. It assumes that the materials are nonmagnetic so that the inductance value is very close to what it would be for vacuum between the conductors. I know what the book looks like that has it in it but I can't find it.

From v=1/(sqrt(e) sqrt(mu0)) and Z0=sqrt(mu0) ln(D/d)/(2 pi sqrt(e)), I get v=2 pi Z0 /(mu0 ln(D/d)).

D is the inner screen diameter and d is the diameter of the centre conductor.

I hope I didn't mess this up.

Jeroen Belleman

Can there be Teflon foam ?? I assmed that was PE foam.

I guess TFE circuit board is faster than epoxy.

greg

Oh, and I should add that 2 pi / mu0 = 5000000 exactly, by definition, so that simplifies to 5000000 Z0 / ln(D/d).

Jeroen Belleman

Sure. Google ptfe foam

John

All coax-type structures propagate at C if the dielectric is vacuum. Then it drops as sqrt(dielectric const) if you use some other insulator.

John

Late at night, by candle light, Enes Erdin penned this immortal opus:

Likely to be as fast as whatever it's attached to.

- YD.

--
File corruption detected. Select option:
1 - Call the cops
2 - Call the press
3 - Bribe it

Remove HAT if replying by mail.

Never heard of a fluke cable tester?

They can even tell you how far down the line the break/short has occurred.

It's called a reflectometer:

--
VWW

Yup.

If you don't have one handy, you can fake one up with a combination of an oscilloscope, a reasonably fast pulse generator, and a pad/attenuator which matches the impedance of your cable (e.g. 50 ohms). I used this sort of setup last weekend to TDR out a misconnected coax at our city's hamshack.

The hookup:

- Start with a pulse generator that can generate a pulse with a reasonably fast rise time (a few nanoseconds is good). Set it to generate a one-microsecond pulse, every few milliseconds.

- Connect the pulse generator to a coax which goes over to your o'scope. Connect a 10:1 (or a 2:1 or 5:1) 50-ohm attenuator pad to the coax, and to a "T" connector.

- Plug the stem of the "T" into your o'scope input. Connect the free arm of the "T" to the coax you want to measure.

- Turn on the pulse generator, set the scope to trigger on the rising edge of the pulse, and set the scale of the scope so that you can see the full height of the pulse. What you'll see, typically, is the pulse trace rising from 0 volts, up to an initial level. The trace will then go out pretty much horizontally for some distance, and will then jump upwards to roughly twice the original level (if the cable is open) or down to zero again (if it's shorted) or to an intermediate level (if it's terminated by a resistance other than its own characteristic impedance), or through some other more complex pattern (if the cable is terminated in a complex impedance). If the flat trace "goes on forever", the cable's probably terminated in its characteristic impedance.

Calculate the time duration of the initial (flat, non-zero) part of the pulse. This is the time that it took the pulse to travel down the coax to the end, be reflected back, and return to the oscilloscope.

From this time, you can calculate the velocity factor of the coax (if you know its length) or the length (if you know the VF).

Example: I used a spare length of RG-58 coax, 20' long. The duration of the pulse was 62 nanoseconds (round trip) or 31 nanoseconds each way. Assuming C = one nanosecond per foot (which is within a couple of percent) this corresponds to a velocity factor of around 0.65, which is just about right for RG-58.

This sort of simple TDR setup can be very effective. When we looked at the mis-connected coax, we could "see" a slight shift in the trace height at a distance which corresponded to around 25' from our shack bulkhead... and this was where the RG-8 coax was terminated in an N connector and then connected to a heliax cable which goes to the roof. The two cables' characteristic impedances were very slightly different (although much closer than the RG-8 was to the RG-58 I was using as a jumper from the scope).

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

I know what my lines are, so my reply is:

How's your eye?

A variation on this theme is to use two probes (with matched delays). Terminate the cable. Stuff a pulse train down one end and probe both ends. Adjust the frequency so that the pulse coming out just lines up with the next one going in. Frequency is easier to measure accurately than delay. And you can use crummier pulses.

mike