What makes a TV Coax lead 75-Ohms?

[Mikey405]

I'm probably going to regret asking, and it may seem like a very silly question, but... What makes a TV coax downlead 75-ohms? Is it 75-ohms only at a certain frequency? Is it simply the reactance at a certain frequency, and if so, isn't it going to be different at Bands 1/3/4/5 and also depending on length?

I guess that this thread could go anywhere but as I was really thinking of TV coax cables, I thought this would be the place to put it.

Thanks everyone.

Kind regards.

From Mike.

[ppppenguin]

Except for secondary effects at very low and high frequencies impedance depends only on mechanical dimensions and dielectric permitivity. For all normal cables, impedance will be sensibly constant over all of bands I to V

http://en.wikipedia.org/wiki/Coaxial_cable

[Mikey405]

Hi Jeffrey.

I was really wondering where "75-ohms" comes from though. What makes it 75-ohms as opposed to 50-ohms or 100-ohms for instance. I've looked at that interesting Wikipedia article and it tells you about working out what the impedance is etc., but I was wondering what physically is "75-ohms" about it?

Will the transmitting end "see" 75-ohms when sending (say) 50MHz into a non-terminated 75-ohm cable?

Kind regards.

From Mike.

[Station X]

I think what you're asking is:-

What is characteristic impedance?

[ENGLISH VICTOR]

75 Ohms was chosen as a defacto standard for T.V. because of it's very close impedance match to a single half wave centre fed dipole. A single dipole has a characteristic impedance Z of between 78 and 72 Ohms when spaced a few wavelengths above the ground and away from surrounding objects. 50 Ohms is a good match to a quarter wave ground plane antenna although I doubt that is the reason why 50 Ohm cable is so common in proffesional R.F. coms.but it is possible. Back in 1948 my parents had a ch1 antenna fed with Belling Lee 75 Ohm TWIN feeder.
Victor.

[Mikey405]

Originally Posted by Station X

I think what you're asking is:-

What is characteristic impedance?

I guess I am Graham, but really I am wondering what it is that "sees" 75-ohms. On an unterminated line, will the sending end "see" 75-ohms from 50MHz to 1GHz for instance? If it "sees" a higher or lower impedance at any given frequency, what defines that the cable must be 75-ohms.

[Mikey405]

Just to add to the confusion... If I had a signal generator with a fixed impedance output and I started to sweep up the band from 1KHz all the way up to 1GHz, where would the cable start to be "seen" as 75-ohms. Obviously at DC and at low frequencies the cable will be seen as infinity (i.e. no load), but as the frequency goes up, is there a point at which the cable starts to be seen as something resembling 75-ohms and then it just stays there? Or does the impedance keep going down and down? If the impedance just keeps going down, at what point do they say "This is a 75-ohm cable".

[John Robson]

One school thought (Kraus) has it that 75 ohms is mainly used for reception because
the theoretical impedance for minimum attentuation of a coax line is 77 ohms, which
rounded down to 75 ohms. Hence the reason for using that type of coax in domestic
TV receivers.

While for maximum power transfer the impedance of 30 ohms is best. So as the half
way point between 77 ohms and 30 ohms is 53.5 ohms (Rounded down to 50 ohms)
was chosen as the value for impedance coax cable used for both transmission and
reception.

[Station X]

If you took a length of 75 ohm cable, terminated it in a resistive load and fed a signal to it you would find that maximum power was transferred when the load was 75 ohms.

That's not the whole story though. The loss will depend on the quality of the cable. For instance a cable with large conductors will have less loss, but it's still 75 ohm cable.

In some cases it might be better to use a low loss cable of the wrong characteristic impedance, since the mismatch is more than made up for by it having lower loss.

For instance my 80 metre dipole has a theoretical impedance of 75 ohms. However I feed it with ladder line feeder with a characteristic impedance of about 400 ohms. Its low loss when compared to 75 ohm coax makes up for the mismatch.

[Mikey405]

Originally Posted by Station X

If you took a length of 75 ohm cable, terminated it in a resistive load and fed a signal to it you would find that maximum power was transferred when the load was 75 ohms.

I can see how that would be so Graham, but isn't that simply the load at the end being 75-ohms?

[ENGLISH VICTOR]

Only a correctly terminated line presents it's characteristic impedance at the sending end at all sensible frequencies (say 100KHz. to 1GHz.) Jassic et al cover the subject in detail as do some of the R.S.G.B. publications. If you were to measure the current flowing into the coaxial line at any sensible frequency the current would be the same as would flow through a non inductive resistor of the same value as the cable's characteristic impedance.
The impedance does not change with frequency because the determination of impedance is a function of the cables inductance and capacitance per unit length. As the inductance offers a rising reactance with frequency the capacitive component offers a falling reactance and vice versa. In simplistic terms the centre conductor looks like a long inductor and the capacitance to the braid looks like a whole lot of small capacitors tapped along the inductor. A lenght of 75 Ohm often has a capacitance of 22pF per foot.
Victor.

[Mikey405]

Ah, okay, that makes sense Victor. So the total impedance of a correctly terminated circuit will always be 25-ohms? (Give or take.) 75-ohms at the sending end, 75-ohms feeder and 75-ohms termination(?)

Presumably at very low frequencies the cable is going to be "seen" as infinity though? At 1Hz, presumably the cable will have practically no effect. Is there a point at which the cable starts to be "seen" as 75-ohms and then the curve will flatten out?

[Station X]

I can see how that would be so Graham, but isn't that simply the load at the end being 75-ohms?

Yes. There's nothing magic about the figure of 75 ohms though. You can get 50 ohm cable which needs to be terminated in 50 ohms for maximum power transfer.There's no reason why cable with a characteristic impedance of 267.5 ohns shoudn't be made. It would need to be terminated in 367.5 ohms to get maximum power transfer.

[julie_m]

Originally Posted by Mikey405

I guess I am Graham, but really I am wondering what it is that "sees" 75-ohms. On an unterminated line, will the sending end "see" 75-ohms from 50MHz to 1GHz for instance? If it "sees" a higher or lower impedance at any given frequency, what defines that the cable must be 75-ohms.

I'm afraid you're going to have to bite the bullet and study some transmission line theory (stop groaning).

Here is a gross oversimplification. Bear in mind two things: (1) Nothing can travel faster than light, not even information; and (2) Energy is never created nor destroyed, it only changes from one form to another.

If you had a very long string, stretched taut, with a bell on the end of it, then in theory you could send a message faster than light because when you pulled your end of the string, the bell would ring straight away -- meaning you just transmitted information faster than light! But because of (1), that's not what would happen. The string would stretch just where it was pulled, and the stretched region would propagate along the string slower than light until it reached the bell. (You can demonstrate this with a Slinky spring stretched along a corridor.)

It's the same with electrical signals. A signal takes some finite time to propagate along a transmission line. The source doesn't know what impedance is present at the far end; so although it knows how many volts it is pushing with, it doesn't know how many amps it should send. What actually happens is that the transmission line presents a certain impedance to the source which may not have anything to do with the impedance of the load. When the signal reaches the far end, if and only if the current was right for the load, all is well and good. But if the current is wrong for the load, then there will either be a surplus or a shortfall (which is just a negative surplus). Any surplus energy is returned to the source as a reflection, which then propagates its way back along the transmission line. But the load that is sending the reflection doesn't know the impedance of the source, and there may well be a counter-reflection, and so forth, until it eventually gets swallowed up by the resistive component of the impedance.

The closer the impedance of the transmission line matches the impedances of source and load, the weaker the reflections at each stage and hence the better the power transfer.

In the case of analogue TV, a signal that has been reflected from the set to the aerial and back to the set will be able to pass through the tuner along with the "main" signal -- but delayed in time, and consequently will show up on the screen as a dim "ghost" image. This is undesirable, so proper impedance matching is very important.

[ppppenguin]

Originally Posted by Mikey405

I was really wondering where "75-ohms" comes from though. What makes it 75-ohms as opposed to 50-ohms or 100-ohms for instance. I've looked at that interesting Wikipedia article and it tells you about working out what the impedance is etc., but I was wondering what physically is "75-ohms" about it?

See the section "derived electrical parameters" in the wikipedia article. The impedance derives directly from the capacitance and inductance per unit length and is an inherent property of anything (not just coax) where these values are sensibly constant. If you terminate 75R coax in 75R all the power transmitted along it will be absorbed at the termination, none will be reflected. For other terminations, some power will be reflected. This is expressed as VSWR (for the RF folks) or return loss (for video types like me)
http://en.wikipedia.org/wiki/Return_loss
http://en.wikipedia.org/wiki/Standing_wave_ratio

What's more, the impedance looking into the end of the cable away from the 75R termination will also be 75R, no matter how long the cable. (Ignoring the resistance of the conductors).

For practical coax, the theory breaks down at low frequencies (depends on cable type but maybe sub 1 kHz for good quality cable, maybe 10s of kHz for old fashioned "low loss" types with poor braid coverage. At high frequencies skin effect and other phenomena can be serious. Almost any coax will be sensibly constant impedance from 100kHz to 1GHz though dielectric losses will be a problem at higher frequencies unless the correct cable is used.

[Mikey405]

Originally Posted by PPPPenguin

For practical coax, the theory breaks down at low frequencies (depends on cable type but maybe sub 1 kHz for good quality cable, maybe 10s of kHz for old fashioned "low loss" types

Ah, okay. That sounds sensible Jeffrey, and that does go a long way to answering my questions.

Originally Posted by ajs_derby

But the load that is sending the reflection doesn't know the impedance of the source, and there may well be a counter-reflection, and so forth, until it eventually gets swallowed up by the resistive component of the impedance.

That's a good explanation too AJS and it certainly gives me something to read up about tonight.

Thanks everyone for your answers.

Kind regards.

From Mike.

[ENGLISH VICTOR]

No Mikey the circuit impedance would be considered 75 Ohms. The cable if low loss should appear transparent and should be seen as an extension of the source neither adding or subtracting from it's impedance. It is very difficult to visualise. It is very rewarding to follow up with some study of the subject, then the visualisation will become clear. I can understand where yo get 25 Ohms from but it is not the case.
An infinately long cable will look like 75 Ohms at an infinately low frequency. Both are impossible of course but might help you visualise the point where the cable starts to take on it's characteristic impedance.
Victor

[Peter.N.]

Never was very good on remembering formulea even though I have been a TV engineer for 50+ years and have a ham radio license! I cant even remember Ohms law, only that it takes one volt to push one amp through one ohm, and thats how I do most of my calculations.
I would simplify impedance matching as being the optimum value to transmit the particular voltage/current ratio with minimum loss - at least, that's how I understand it.

[kalee20]

The characteristic impedance of a cable is the impedance you'd measure at the end of an infinitely long length. Ideally, this impedance will be resistive and independent of frequency, and you could measure it with an Avo.

For practical cables, as has been said above, impedance tends to change at higher frequencies, due to effects like dielectric loss in the insulation, skin effect in the conductors, etc. But if these are absent, you could indeed measure the characteristic impedance with an Avo. The difficulty is getting a long enough length.

A length, say enough to wrap round the earth a dozen times, should in principle be Avo measurable. The needle would be steady at 75 ohms for a couple of seconds, and then move upwards.

As a thought experiment, consider a really long length of cable. As this will have a massive total capacitance, measuring with an Avo would at first thought give a reading close to zero, rising very slowly (as with connecting your Avo to a capacitor). However, most of the massive capacitance is at a considerable distance so there's the wire inductance between it and the Avo, and this slows down the charging. It turns out that the two effects, for an infinitely long length, have a net result of a 75 ohm reading.

Once you've grasped that idea, you can chop off your length at 50 yards and replace the remaining infinite length by a 75 ohm resistor, aerial, etc and at your local end you'll never know the difference. The cable with remote termination will look just like a pure 75 ohm resistor, and all the power you feed into it will be absorbed.

[Mikey405]

Originally Posted by ENGLISH VICTOR

No Mikey the circuit impedance would be considered 75 Ohms. The cable if low loss should appear transparent and should be seen as an extension of the source neither adding or subtracting from it's impedance.

Ah, okay, thanks Victor - I'll have a look at my RSGB or ARRL book when I get home if I can find either of them. I think I can see how this works now, and you are right, it does sound like a very interesting subject.