When does coax impedance (not) matter?

[staze]

All,

So, want to run a small piece of coax between my function generator (which I'll be getting), and my scope. I have some 1M RG-58 (50?) stuff, but since the scope is going to be on top of the FG, figured a much smaller piece would work.

I also have some very short (like, 30cm or less) RG-59 (75?) stuff. For this short a distance, is the impedance really going to matter? Keep in mind, not working with anything higher than 10Mhz or so, max 20Vpp.

Regardless of answer, an explanation of why would be helpful for learning. =)

Thanks!

[SeanB]

Short lengths ( less than 1/20 wavelength ) will not be a problem.

[rf-loop]

I also have some very short (like, 30cm or less) RG-59 (75?) stuff. For this short a distance, is the impedance really going to matter?

Keep in mind, not working with anything higher than 10Mhz or so, max 20Vpp.

If this around 10MHz include also your signal highest important frequency components answer is easy.   In practice , example for 10MHz sinewave  - just go on. No matter.

But, if you look 10MHz (or 1kHz) square waves what have fast risetimes or other signal what have fast changes then there is higher frequency components  and this impedance matching is more and more important as freq rises..

[staze]

good to know. I guess once I have the FG, I'll try with some standard 50? (1M), then with a short little piece of 75?. But since 1/20th of a 10mhz signal is 1.5M, it shouldn't be a problem, seemingly. Heck, guessing I'm at more like 1/100th of a wavelength.

Thanks!

[Galaxyrise]

The 1052E has a 1M input impedance anyway; it's not like you'd have a matched pathway with just a RG58.

[SeanB]

Remember frequency of interest includes harmonics of a square wave. Thus for your 10MHz square wave you will have rounding of the 100MHz harmonic. Sine wave it will do and get you a reasonably accurate show, but with higher harmonics they will be affected somewhat.

Then again I have sent 75R video down 91R thinnet with little distortion and good colour even at the end of 100m of cable.

[staze]

Yeah, I know you could generally send video (75?) down 50? cable for short runs with little issue. I just didn't know how it faired the other way around.

Once I get an FG I'll give it a whirl and post back. But yeah, hadn't thought of the harmonics. But I should still be less than 1/10th wavelength for the 100mhz signal, so I wouldn't think I'd get MUCH rounding, but I really don't know. I just happened to have a couple of nice short 75? bnc cables. If they work well enough, I'll pick up some short 50? cables next time I do a digikey/mouser order. =)

[w2aew]

As mentioned in an earlier post - since your scope doesn't have a 50 ohm termination, the point is somewhat moot.  If you are going to be dealing with frequency components >>10MHz, then you'll want to pick up a 50ohm thru termination to properly terminate the end of the coax.

[staze]

I do have a Tee connector, and a 50R terminator. So that's not really an issue.

I'm just curious purely about the coax itself.

[vk6zgo]

Most of the time,for what you're doing, it won't make a lot of difference.

One thing,though,some of the stuff they sell as RG59 these days is nothing like the real thing.

"Real" RG59 is quite flexible,has a fairly dense copper shield,a copper inner conductor and an outer PVC sheathing which doesn't move around w.r.t the rest of the cable.

"Pretend" RG59 is stiff,has either an aluminium shield,or a fairly minimal copper one combined with some cruddy aluminium foil,a copper plated steel inner conductor,& the integrity of the sheathing is poor.

It may not fit BNC connectors made for the real stuff properly,with resultant loss of signal at embarrassing times.

RG58,by & large,is a lot closer to the original article.

[Salas]

I do have a Tee connector, and a 50R terminator. So that's not really an issue.

I'm just curious purely about the coax itself.

You need a feedthru. T will give the right level but will still create some reflections.

[staze]

huh. well, down the road I guess. feed-thrus are spendy, even over ebay. will just have to keep an eye out.

[ConKbot]

<<snip>>

RG58,by & large,is a lot closer to the original article.

Ive worked on my share of 2.4ghz 802.11x antennas with "RG-58" leads, but they have a solid core and the jacket it bigger in OD to the point I cant even fit a crimp ferrule on for replacement conenctors. (after having to break out the soldering iron to have to solder the center pin of course, not like the crimp on the center pin would stay put on solid wire)

So that plague has seemed to spread to RG-58 also. 

[larry42]

I'd be more worried about the fact that a 75 ohm coaxial connector has a different physical geometry to a 50 ohm one and usually leads to either a bad connection or damage to the equipment. If you are going to connect a 50 ohm connector to some 75 ohm cable can you also run the risk of a poor connection. ergo I would suggest to use the one meter of cable.



Sent from my HTC One S using Tapatalk 2

[alm]

Depends on which coaxial connector you use. N connectors have a different size center pin, BNC connectors don't and mate fine.