Talking House ATU On Roof of House

' "When the ATU is elevated, the coax shield IS the ground wire."

Yes, and the Talking House/ATU combination was FCC Part 15 certified in exactly that configuration." '

Does anyone know if this true?  AFAIK there is no public information with photos or descriptions of the physical configuration used during the FCC certification of the TH system.

Good question, Rich.

Rich, I'm not sure what your question is which is referred to as a "Good Question".

The only way this can be put to rest is to obtain an answer directly from the FCC.

We do have one page of the FCC OET test data as supplied by Radio Systems, the manufacturer of the Talking House system, which describes the use of the ATU as tested.  

We do not have pictures or a description of the physical setup other than 8 meters of coax was used to connect the transmitter to the ATU.

We do have FCC certification obtained using this afformentioned information.

We can discuss ad nauseaum whether an increase in range when elevated is due to radiation from the coax or simply the mere line-of-sight gained by said elevation.

Hold a portable transceiver in your hand using either a collapsable whip or "rubber duck" antenna, an electrically short monopole compared to the frequency of operation.  There is no physical ground.  The counterpoise is perhaps your body capacitively coupled to the circuit.

At ground level check the range obtained.  Now, stand on top of a structure some distance above ground level.  There is still no physical ground connection entering the equation but the range will be markedly improved due to line-of-site i.e. no obstructions.

It would seem most agree that IF the coax has RF current on the outside it will radiate causing improved range.  Simply having the antenna elevated above obstructions would also improve range.

All of these discussions assume RF current on the outside of the coax.  Testing needs be performed to determine how much RF actually flows on the outside of the coax when the ATU is used.  

When properly terminated there is no appreciable RF on the outside of coax.  That is why a balanced dipole antenna requires a balun when fed with unbalanced coax, to prevent RF on the outside of the coax.  If you can assume RF on the outside of the coax then you are assumming the ATU is not a proper termination.

This complete discussion is moot-until the FCC addresses it because at the end of the day, our opinions and theories and physics and BS and MS have nothing to do with it because whether the coax radiates or not they can still shut you down.

"The only way this can be put to rest is to obtain an answer directly from the FCC."

Another way is possible if the holder of the documentation submitted to the FCC for the TH certification releases it to the public via a website post, or similar.

"We can discuss ad nauseaum whether an increase in range when elevated is due to radiation from the coax or simply the mere line-of-sight gained by said elevation.  Hold a portable transceiver in your hand using either a collapsable whip or "rubber duck" antenna,... Now, stand on top of a structure some distance above ground level...the range will be markedly improved due to line-of-site i.e. no obstructions."

This effect tends be true for frequencies at VHF and above, where propagation uses the space wave radiated by the antenna.   But receiver/antenna heights to 100 feet or so above the earth have little bearing on the reception of far-field ground waves in the AM broadcast band.

For example, I just took my Tecsun PL-310 to the basement of my home (which is below ground level, of course), and put it flat on the concrete floor there.  For a local AM station the signal meter read 64 dBµ.  Then I placed the Tecsun against the ceiling of the first floor, some 17 feet higher than it was before.  The meter still read 64 dBµ.

"When properly terminated there is no appreciable RF on the outside of coax.  That is why a balanced dipole antenna requires a balun when fed with unbalanced coax, to prevent RF on the outside of the coax.  If you can assume RF on the outside of the coax then you are assumming the ATU is not a proper termination."

But the TH ATU output load is not a balanced antenna, it is a monopole.  In order for that monopole to radiate it needs to be "driven against" another conductor, or some conducting path to the earth.  That is how monopoles work !

Some of that conductor length is provided by the circuit ground and chassis of the ATU.  And, as the coax shield is connected to the ATU chassis at the input of the ATU, and the ATU has no known means to prevent it, that r-f current on the chassis will flow down the outside of the coax shield -- which (of course) adds radiating length to the ~ 3-m whip considered to be "the antenna."

Other things equal, the higher field intensities produced by TH systems using elevated ATUs are due to the greater radiating length of their antenna systems, in the vertical plane.

All of these discussions consider the antenna as being end fed and having a low impedance feedpoint, i.e. quarter wave radiator or electrically short antenna with loading coil to provide a low impedance feed point.

An end fed electrical half wave antenna has a very high impedance feedpoint, i.e. greater than 3000 ohms. The result, high voltage-low current at the feedpoint.

Any return current, i.e. from ground or counterpoise, would be equally small to balance out. That being the case, any common mode current flow on the outside of a coax feed would be negligable.

How can low impedance coax be used to feed a high impedance antenna such as described? With an impedance matching network/tuner.

Coax can be link coupled to a parallel tuned L/C antenna matching network. This forms a transformer. A turns ratio of 10:1 would transform the 50 ohm coax to 5000 ohms (Zpri/Zsec = (Npri/Nsec)squared).

A document supporting this practical explanation was authored by Steve Yates - AA5TB.

This explanation could well be the key to how the Talking House ATU antenna system works with no apparent ground system. Rather, the small balancing current required is satisfied by stray capacitance which satifies the monopole return current theory.  Ground loss would be minimal as the return current is small (ground return = high impedance.)

The following link takes you to this article:   http://www.aa5tb.com/efha.html

I would suppose you choose not to read the article to imply don't confuse me with facts, my mind is made up.

Link coupling to the ATU isolates the antenna current from the feedline current.  The transformer action of the coupling provides the high voltage/low current applied to the high impedance radiator.

Current entering a node must leave the node.  I.E. the current entering the end of the half wave radiator (low due to the high impedance) must return in the same amount to balance. Power loss is I^2 R.  Ground loss would be swampped out as the ground resistance is small comparred to the feedpoint.

Understandably, a 3 meter antenna would still be a poor perfomer but per the article a physical counterpoise of .05 wavelength or the equivalent due to stray capacitance works quite well. 

Yes the feedline current will be large but the antenna return current is on the antenna side of the match (Hi-Z), not the feedline side (Low-Z).

Teach me o'master.

"Understandably, a 3 meter antenna would still be a poor perfomer but per the article a physical counterpoise of .05 wavelength or the equivalent due to stray capacitance works quite well."

Note that AA5TB's paper deals with a 1/2-wave (180-deg) end-fed monopole system.

Do you expect the same result when using a 2/100-wave (6-deg) end-fed monopole?

The clip below from his paper shows his statement about using the coax shield and transmitter in place of a 0.05-lambda counterpoise.  This is the configuration used in the TH ATU, as the cold ends of its input matching transformer windings are hard-wired to its chassis, along with the coax shield.  This configuration will cause r-f current flow on, and radiation from the outside of the coax shield.

I think AA5TB's wavelength dimension on the left side of the transformer should read <0.05 lambda (i.e. less than 0.05 lambda), not >0.05 lambda.

i'm going to end this now. i own a calibrated FIM. a nems clarke 120e. i have tested the talking house ATU in an elevated (6 meter above ground) and non elevated (1 meter above ground) installation. there is negledgable difference between the tow installs inso far as higher field stength for the elevated install versus the non elevated. the coax does not radiate that much to be of any signifigance to the FCC. granted i could not get the thing 50ft in the air but i think 20ft vs 3ft there would still be signifigant increase in F/S if the coax was radiating. when i say insignifigant increase the increas was only about 20 or so uV/m at 30m. not hundreds of uV's. there are transformers in the THII - V and the iAM and ATU which is probably to reduce common mopde currents on the coax sheild in additon to stabilizing the transmitter power output. i have done real world testing. both grounded and ungrounded. there was a bigger increase in F/S off the ground wire when ATU was grounded and elevated than there was off the coax and even then it was still only double digit increase not 100's of uV's.

can we please stop beating this officially dead horse and just let this topic die now?

No, I don't think a 1/50th wave end-fed monopole will perform as a half wave monopole.  I believe I already said that.

The example you picked to display is but one explanation of how the return current can exist.  His examples also show stray capacitance coupling to the environment.  

The total return current in this case does not rely solely on the coax shield.

The equivalent of a half wave monopole can be represented as a series R/C/L circuit just as a quarter wave monopole can be shown.  When the radiator is only 3 meters it can be made an electrical quarter wave by adding inductance. 

Question:  Can the loading coil and 3 meter radiator present the electrical equivalent of a half wave just as a loading coil and 3 meter radiator present the electrical equivalent of a quarter wave?  

The Talking House ATU loading coil is separate from the toroid coupling transformer.

That half wave equivalent would have a Hi-Z feed point which the toroid input transformer is used to match the Lo-Z coax.  The transformer action isolates antenna current from the feedline current.  The antenna return current (low side of the transformer secondary) would equal the antenna current (high side of the transformer secondary.)  With a 10:1 coupling ratio, the antenna current would be a fraction of the feedline current.

Lighten up Rev.  This is a friendly discussion and I'm learning.  I value the knowledge offered by Rich.  I just tend to not accept judgement based solely on theory.  

Scientific theories are not quite the same thing as facts, but they are often very similar; scientists usually test their theories extensively before airing them, looking for obvious problems which could cause them to be challenged.  At some point in time theories accepted as fact are challenged.  Sometimes they stand, sometimes they fall.

Once we accept the theory and discount challenge the learning ends.  Otherwise we'd still believe the Earth is flat.

As you can see, the loading coil and coupling transformer are two separate components.  The antenna current is isolated from the feedline current by transformer action.

If the 3 meter whip and loading coil can present an electrical half wave to the coupling transformer, the antenna current would be small (Hi-Z) while the feedline current could be much higher due to impedance matching to the (Lo-Z) coax.

There could be a common mode return current on the coax depending on the impedance of the secondary but the amount of current, being the same as the HI-Z antenna current, would be a fraction of the feedline current.

The best part about this is the FCC accepted this as part of the certification process.

The earth is flat here, but it may be different there.

If we terminated discussions everytime one member had enough of it, eventually part15.us would be finished.

I still think what I think no matter who's right.

If the EUT is placed on a one meter high table along with the ATU, the 8 meters of coax is wrapped in a serpentine fashion. I can assure you that the ATU was not suspended 26 feet above the test platform.

the ATU is placed on a 1 meter high table attached to a ground plane on the floor with a connecting wire with the coax along the ground in a serpentine fashion. apparently 8 meters of coax was used and i guess the talking house was placed on the table with the ATU as well. we can assume some things on testing based on what test data shows for newer transmitters that are available and what the released documents show. why Radio Systems Global just won't release all the data they have on the transmitter remains a mystery. why rangemaster wont release theirs as well is a mystery.

"If the 3 meter whip and loading coil can present an electrical half wave to the coupling transformer, the antenna current would be small (Hi-Z) while the feedline current could be much higher due to impedance matching to the (Lo-Z) coax."

The input reactance of a base-driven, 3-m monopole in the AM broadcast band can be offset to zero using a loading coil in series with its feedpoint.  But that does not change its radiation resistance -- which will still be in the vicinity of 0.1 ohm at the upper end of the band.

Radiation resistance is a function of the physical height and cross section of the monopole with respect to the applied frequency, which does not change when that monopole is externally tuned to resonance.

So that resonated, 3-m whip has a very low input impedance (about 0.1 +j 0 ohms), and is not the electrical equivalent of a monopole that is physically 1/2-wave in height, which natural input impedance at its base when using a thin conductor is on the order of 2500 -j 1200 ohms.

When that full-sized 1/2-wave monopole is externally resonated, its base feedpoint impedance is 2500 +j 0 ohms, which is far from that of the resonated 3-m whip.

Note that losses in the matching network and r-f ground system will raise the value of the real (non-j) term of these input impedances for both of these radiators.  The biggest difference in percentage will be for the 3-m monopole, for which the real term might rise to 50 ohms or more -- but still much less than the real term of the full-sized 1/2-wave monopole with those losses included.