[TowerTalk] Tin roof blues

[Guy Olinger, K2AV]

To the original poster, one of the ground-independent multiband verticals like the R8, etc, would be quite good above a tin roof.

The difficulties with ground for a vertical are

1) insufficient radial density and subsequent power loss at the ground connection used as a current sink. The R8 and its kind are a type of vertical dipole, so this loss is not present.

2) E-plane losses in the ground directly underneath the vertical, the reason a dense ground screen will help an R8 or the like. Doesn't have to actually connect to an R8, it just puts a shield over the ground.

3) vertical polarization reflection loss in the first several hundred meters. Only thing that fixes this is an obscene investment in radials, salt-water marsh or ocean shore location.

Your tin room nicely takes care of 2) above. It's about as good a dense ground screen under a vertical dipole as you will get. Plus the downward radiation will be reflected at low angles in two directions if the roof is peaked. 

When I lived in an apartment in Washington DC, I had a tar covered, barely sloped tin roof, and the vertical I had up there played like gang-busters. I just didn't know why. Now I do.

73 & good luck.

> 
> From: "Barry Kirkwood" <bjk@ihug.co.nz>
> Date: 2001/06/12 Tue PM 09:11:34 EDT
> To: "Tower Talk" <towertalk@contesting.com>
> Subject: [TowerTalk] Tin roof blues
> 
> Some suggestions on using metal roof as groundplane.
> When I lived back in the city I had a small two story town house with a
> galvanised corrugated steel roof.
> I used various quarter wave verticals with their base on a stub mast working
> against 3-4 slightly sloping quarter wave radials.
> For 80m I used a quarter wave inverted L (more like half an inverted v).
> This was alwys used with at least one quarter wave radial. These longer
> radials would run to a corner of the roof and then along the fence tops at
> the sides of the lot.
> I bonded the roof iron by spot soldering thin tinned copper wires at the
> corners of the metal sheets.
> Using a simple home made clip on rf current meter showed no current in a
> wire from the junction of the high band radials to the roof. i.e. the fan of
> radials effectively isolated the system from the metal below.
> Current flowed in this wire when I worked the 80m inverted L. I measured the
> drop in feedpoint impedance as extra 1/4 wave radials were attached to the
> 80m feedpoint.
> I came to the conclusion that the relatively small metal roof was equivalent
> to at least 5 radials so far as currents were concerned.
> I never tried the higher bands against the roof alone.
> The 80m L worked very well, better than might be reasonably expected.
> The little link wires corroded out very rapidly, so use a better bonding
> technique if going this road.
> I would certainly advise giving the metal roof a try as a groundplane.
> 73
> end
> Barry Kirkwood PhD ZL1DD
> Signal Hill Homestay
> 66 Cory Road
> Palm Beach
> Waiheke Island 1240
> NEW ZEALAND
> www.waiheke.co.nz/signal.htm
> 
> 
> 
> List Sponsor: Are you thinking about installing a tower this summer? Call us
> for information on our fabulous Trylon Titan self-supporting towers - up to
> 96-feet for less than $2000! at 888-833-3104 <A HREF="http://www.ChampionRadio.com">
> www.ChampionRadio.com</A>
> 
> -----
> FAQ on WWW:               http://www.contesting.com/FAQ/towertalk
> Submissions:              towertalk@contesting.com
> Administrative requests:  towertalk-REQUEST@contesting.com
> Problems:                 owner-towertalk@contesting.com
> 
> 



List Sponsor: Are you thinking about installing a tower this summer? Call us
for information on our fabulous Trylon Titan self-supporting towers - up to
96-feet for less than $2000! at 888-833-3104 <A HREF="http://www.ChampionRadio.com">
www.ChampionRadio.com</A>

-----
FAQ on WWW:               http://www.contesting.com/FAQ/towertalk
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com