Long time lurker, just joined and posting for the first time…

I have just been reading and learning here for a while, and already tried a SBGH once, but I am about to make a DBGH, so I figured I ought to join the discussion now.

My situation, briefly, is that I have one city’s antenna farm fairly nearby (~15 miles), and another city’s farm farther away (~45 mi), but at nearly 180 degree opposite directions (around 160-ish). My antenna has to be in my attic, and won’t/can’t be outdoors (attic elevation is about 25 feet up). I had been looking for a bi-directional antenna that could get both cities well, and currently have an unreflected DIY DB4 which does a fairly decent job.

When the GH design first hit the web a while back (all the buzz from the various online write-ups), I read everything linked in the (then-8-page-long) thread (original patent, etc.). Because the rod-reflected version seemed to have pretty good “backside” pickup, I decided to make a rod-reflected model, thinking that I could aim it at the farther city, and maybe the “backside” would still pick up the other, stronger city reliably. I fashioned a SBGH (6-pairs of rods) out of what I had laying around (admittedly not ideal): all the elements were 14 ga copper from some romex house wire. I attached the rods to a 1x4 using small wire staples (the kind with the curved top and plastic sheath that come with some low-voltage indoor wiring). It worked pretty well and held the rods perfectly in place. The driven element was offset forward from the rod “plane” using three small blocks of wood attached to the 1x4 at the appropriate locations.

I have to say, for being in an attic, and being made from only 14ga copper, it worked very well – but only as a “directional” antenna. It picked up the distant city’s stations better than any previous “DBx” experiments (though no previous models were reflected/directional). When aimed at the local city, it grabbed almost everything in the high 90s or 100% signal. BUT… no matter which way I aimed it, the backside was never enough to be adequately “bi-directional” for me, and the few stray "off to the side" stations were bad, so I ended up ditching it and later dismantling it. Granted, if I had intended on it being permanent, I would have made it from more sturdy materials.

Now my situation has changed and I am going to have two separate antennas, each aimed at one of the cities. Because of my previous results, I have decided to go with a DBGH for the more distant city. I will be making it out of 10ga copper this time (much more sturdy), but will be using a screen as the reflector instead of rods. I have some 1/2”x1/2” “hardware cloth” mesh that I plan on using as the reflector screen, though I may just cover the cloth with aluminum foil. The frame will again be wood.

While I have no modeling experience with software and such, I am willing to tweak my antenna per modeled recommendations, attach pictures, and report results. I am intrigued by the idea of contoured screen (“shapely”? :) ), if anyone is interested in modeling that in a "double" rather than single bay. It won’t be ideal, since it’s in an attic and most here seem to be outdoor types, but it’s something. So feel free to suggest any dimensional tweaks for my screen, and, if feasible, I’ll try it.

Sorry this is so long.

Disclaimer: Being an American, I prefer inches over millimeters. ;)

pj*4,

First let me welcome you to our forum. I was going to ask you for your location and the TVfool data, but I found that in your post #1282 on the Lumenlab DIY forum.

So as I understand your situation, you have two DB-4 clones (7" bow-ties and 7" spacing) without reflectors, in your attic and you are having dropouts on WKRC Ch31 with a Rx (dBm) of -41.1 and you have never been able to get a lock on WBDT Ch18 with a Rx (dBm) of -112.1.

You tried the SBGH, hoping for reception on all channels with a single 6 pair SBGH, with the Cincy stations coming in the back-end, but apparently that did not work for you and you dismantled the SBGH.

Now you are considering having two antennas in your attic, one aimed at Cincy and the other at Dayton, and you are trying to decide which antenna to build? A SBGH, or a DBGH, or a DB-8 with a reflector screen ?

So regrading your two problems, the occasional dropouts on Ch31 would indicate that most likely you need a bit more gain. The signal strength is quite high at RX(dBm) -41.1. Did you get clean reception on Ch31 when you tried the SBGH?

The reception of Ch18 is a more difficult problem, as the signal strength is low at Rx(dBm) -112.1, and typically the gain of the DB-4 and DB-8 at Ch 18 is not that great, as reported by hdtvprimer.com, and then on top of that, you have the additional losses due to your indoor location. (Predicting losses in your attic is difficult, but you can find some info via a google search)

Are you using a preamp to "go-after" Ch 18?
Is it at all possible to make tests outdoors, even though you can't have a permanent outdoor installation? If you can't lock with an outdoor test, then the attic is "der-funked"

The net gain of the commercial DB-4 and DB-8 antennas are available here...
http://www.hdtvprimer.com/ANTENNAS/comparing.html

....and you can compare that data, to the net gain of the current v3.0x shown in post #1 of this thread.
(Mid point Freq for Ch 18 is 497Mhz)

To sum up: for fixing the dropouts on Ch31, your DB-4 with a reflector, or the SBGH 6 pair GPLv3 should cure your problem.

For Ch18: you may, or may not get it from your indoor location. You will likely need a preamp and the highest gain antenna for Ch 18, that you can build.

You may want to give v3.0x a "spin", once I have finished the modeling optimization, since the gain performance at the low end of the UHF band is much improved.

So as I understand your situation, you have two DB-4 clones (7" bow-ties and 7" spacing) without reflectors, in your attic and you are having dropouts on WKRC Ch31 with a Rx (dBm) of -41.1 and you have never been able to get a lock on WBDT Ch18 with a Rx (dBm) of -112.1.

You tried the SBGH, hoping for reception on all channels with a single 6 pair SBGH, with the Cincy stations coming in the back-end, but apparently that did not work for you and you dismantled the SBGH.

Now you are considering having two antennas in your attic, one aimed at Cincy and the other at Dayton, and you are trying to decide which antenna to build? A SBGH, or a DBGH, or a DB-8 with a reflector screen ?
That’s pretty much it (good thing I used the same screen name here, saved me a lot of re-typing!), except that only one of my current antennas has dropouts on the “nearby” channel 31 (I don’t know if you caught that). The other antenna gets it solidly in the high 90s with little fluctuation and no dropouts. And the dropouts on the “worse” antenna are not even consistent – they seem to be affected by time of day, weather, etc. When I was running up and down the ladder to the attic, fiddling around with location and aim and then checking reception numbers (using the HD Homerun diagnostic software), I settled on locations where both of my antennas were getting all the Cincinnati stations fairly consistently and at least one was getting Dayton stations pretty well. Unfortunately, that meant “settling” for the fact that the antenna that got Dayton slightly better also (strangely, I thought) had a little problem with nearby channel 31. If I went up in the attic and played around with that antenna, I have no doubt I could get it to consistently receive 31 in the 90s, but then I would probably lose the Dayton stations, or at least make them more inconsistent. I believe it’s a matter of multipath, not lack of gain - and perhaps directly related to nothing more than its location with respect to the other antenna up there.

The reasons that I settled on my current arrangement were that, at the time, I was unclear about the tuner-choosing priority of my HTPC software, I did not know about how to create a separated EPG setup on the software (some tuners on one city, some tuners on another), and, at the time, I was less concerned about the Dayton stations. But since then, I have not only figured out considerably more about my HTPC software, but my wife has started regularly recording some shows from a Dayton PBS station (58) that she sometimes has to delete because they have too much pixellation and many dropouts (depending on time of day and weather when the shows recorded).

So, armed with the knowledge of how to manage it within my HTPC software, I am now wanting to turn my two antennas into a directional, more powerful “Dayton antenna”, and a slightly-more-powerful “Cincinnati antenna”. As you may have seen in my tvfool plot, the Cincinnati stations have a slight spread of direction (hence I would sacrifice some strength in exchange for a broader “aim”), but all the Dayton stations come from one direction.

So regrading your two problems, the occasional dropouts on Ch31 would indicate that most likely you need a bit more gain. The signal strength is quite high at RX(dBm) -41.1. Did you get clean reception on Ch31 when you tried the SBGH?
While most of this is addressed above, yes, I did get channel 31 very well with the SBGH, when it was facing Cincinnati (not from the backside when facing Dayton). But as I mentioned in my Lumenlab explanation, the SBGH was so narrowly-aimed that I had problems with even the slight spread of the Cincinnati station directions… if I aimed at the downtown towers, those few other stations that are a few degrees off were a bit inconsistent. That’s why I will make my new “Cincinnati antenna” less directional, possibly even keeping the unreflected DB4 (the better of the two) that I have now.

The reception of Ch18 is a more difficult problem, as the signal strength is low at Rx(dBm) -112.1, and typically the gain of the DB-4 and DB-8 at Ch 18 is not that great, as reported by hdtvprimer.com, and then on top of that, you have the additional losses due to your indoor location. (Predicting losses in your attic is difficult, but you can find some info via a google search)

Are you using a preamp to "go-after" Ch 18?
Is it at all possible to make tests outdoors, even though you can't have a permanent outdoor installation? If you can't lock with an outdoor test, then the attic is "der-funked"
I know my chances at channel 18 are slim, and basically zilch with my current unreflected DB4. That’s why I am hoping my new, more powerful, DBGH, aimed straight at Dayton, might have a shot. My gain should be considerably more. However, as I said over on Lumenlab… if I don’t get it, no big deal. I don’t really even watch anything on The CW… I mostly want it “because it’s there”. :)

....and you can compare that data, to the net gain of the current v3.0x shown in post #1 of this thread.
Any chance you could repost the gain graph from above, but add the curve for the screen-reflected DBGH that appeared to be the “best design” of the version 1 thread (from which this thread spawned)? It’s easier to compare when they are all on the same graph.

My apologies for the length of both of my posts. Hopefully my situation is now explained and I can talk "nuts and bolts" of antennas now.

pj^4
Assuming Autofils got the right info from Lumenlab, I have to say the WKRC issue sounds like multipath - dropouts on a strong station - while WBDT reception is an antenna design issue; a 7" bowtie array is optimized for Ch 80, not 18!

While a reflector may eliminate the multipath, it may not give enough boost to get Ch 18. An SBGH will address the design frequency issue, but you'll need the reflector for multipath at least, so it's not a complete solution, either.

If a reflector added to your current clone stablizes WKRC, an SBGH aimed at Dayton will get you WBDT. You lose 3-4 dB if you combine their signals on the mast, but that may not be as big a hit as it seems, and you can always run dual downleads to an RF switch if combining losses make a difference. Pre-amps would also eliminate the losses.

This all assumes good quality, functioning RG-6 coax. If there's any twin-lead anywhere, replace it with coax and appropriate impedence matching baluns. I've talked to folks with multipath issues who got equivalent reception disconnecting the antenna and shorting the twin-lead - it makes a great antenna, but an unwanted one when used as a downlead. When they replaced it with coax, lots more stations would lock.

Frank

Frank,

Looks like I need to make some further clarification here, since most of this is posted in the Lumenlab forum and not here…

-My understanding was that the 7” DB4 version was optimized for best “overall” coverage of the UHF TV spectrum (which is why commercial versions are around this size).

-I have two separate antennas, but each feeds a separate “HD Homerun” (HDHR) tuner. The two antennas are not joined together in any way. The HDHR is a dual tuner unit, and comes with a splitter and two short pieces of RG6, so each antenna feeds two tuners on my HTPC system. I currently have two “bi-directional” (unreflected) DB4s, which get stations from Dayton and from Cincinnati… so that means I have four tuners all set up to receive stations from both cities. But because I have figured out my HTPC software’s recording priorities, I am willing to sacrifice having that many tuners available for both cities, in order to have better and more consistent reception from each city. So instead of two bi-directional antennas (wider reception but less gain), my intent is to have two directional (reflected) antennas, one pointing north at Dayton and the other pointing south at Cincinnati.

-Since Dayton is considerably farther and also in one single direction, I am planning on making a bigger, fully-reflected, narrowly-aimed antenna for that one. I decided on a DBGH per the version 1 thread here, but then found this thread. Of course, this one isn’t modeled and “optimized” yet, so maybe when it gets there, I will dismantle and rework the DBGH.

-Since Cincinnati is much closer and the signals are much stronger, I may not bother with a reflector. Tests with reflectors in the past have given me too narrow of a reception pattern, and I end up having issues with some of the channels that come from sides. The current locations in the attic are for maximized reception of both Cincinnati and Dayton; if I only need Cincinnati on this antenna, I can probably find a location that gets me those stations reliably. My other antenna (not the one with problems on channel 31) gets all of the Cincinnati stations in the high 90s, and has channel 35 (local NBC) pretty much locked at 100% constantly.

-No twin-lead anywhere. Both antennas have baluns and good-quality RG6 to the tuners. Each antenna has about 10’ of coax total before the tuner (I mounted the HDHRs on a shelf in a second-floor hall closet so they are about a foot below the attic).

I think, at this point, I ought to be over in the other thread, since I am making a DBGH. I only posted here because I was willing to be a guinea pig for the new model, but if it's not ready yet....

Frank,
.... My other antenna (not the one with problems on channel 31) gets all of the Cincinnati stations in the high 90s, and has channel 35 (local NBC) pretty much locked at 100% constantly.
....

I think, at this point, I ought to be over in the other thread, since I am making a DBGH. I only posted here because I was willing to be a guinea pig for the new model, but if it's not ready yet....


pj*4,

Just a couple of points.
If I understood your explanation, you made two identical DB-4 clones (no reflector) and each has its own down-lead to separate tuners.

Is there a difference in Ch31 performance, using either antenna? If so, then you should double-check your down-leads and the 10 connections for the DB-4 clone. A bad connection on any of the bowties, or the mid-point balun connection could give strange results. If you get the same dropout problem on Ch 31 with either DB-4 clone, then as fbov mentioned, its most likely a multi-path problem, which you can easily solve.


The current work on v3.0x is in the "optimization phase" and it will take some time before I get that sorted out... lots of variables to fiddle with.

In the meantime, you could give the DBGH GPLv3 design a spin. For your attic, I'd just go for a simple build with romex wire or equivalent and a tin-foil reflector. That way you can keep the costs down, yet still get a reasonable hi-gain antenna to try out for Ch 18.

You'll find the net gain model results for the DBGH GPL v3 at this web page. Looks like Ch 18 is about 13.5 dBi per the model results.

http://www.digitalhome.ca/ota/superantenna/performance.htm

Is there a difference in Ch31 performance, using either antenna? If so, then you should double-check your down-leads and the 10 connections for the DB-4 clone. A bad connection on any of the bowties, or the mid-point balun connection could give strange results. If you get the same dropout problem on Ch 31 with either DB-4 clone, then as fbov mentioned, its most likely a multi-path problem, which you can easily solve.

The only issue I have seen is location in the attic... which is why I suspect that it's almost definitely multipath. The same ("worse") antenna has been in different locations in the attic and worked great. It was just a matter of trying to find a location that would get both Dayton and Cincinnati well. The current location is probably about the best a DB4 can do up there, and it just happens to either have multipath, or perhaps is on the edge of a null, for channel 31. No big deal - it will be remedied soon enough!

In the meantime, you could give the DBGH GPLv3 design a spin. For your attic, I'd just go for a simple build with romex wire or equivalent and a tin-foil reflector. That way you can keep the costs down, yet still get a reasonable hi-gain antenna to try out for Ch 18.
On the way. The wire part is done. I just need to figure out if I have enough "hardware cloth" (1/2"x1/2" metal mesh) to make the full screen, or else decide whether I want to do aluminum foil (really cheapo) or possibly go buy more mesh. It's supposed to be chilly and rainy tomorrow, and may be my last chance to spend any "quality time" :D in the attic before it gets unbearably hot, so hopefully tomorrow will be installation day.

Thanks for the suggestions!

Update: in searching around Sunday night for what to use as a reflector (without having to go buy something), I ended up settling on grid pieces from some old “Yaffa blocks” (remember those?). Each one was 15”x15”, with wire grid at 1.5” intervals. So two, side-by side, gave me a 30” wide screen, and I used 5 pairs to make a screen of 30”x75”. By the time I got it attached, the contraption was quite heavy. I managed to get it up in the attic through the access hole (had to bend the screen a bit, but it sprung back into shape easily)… only to discover that it would only stand up in the dead center of the attic, directly under the peak, which was far from the optimal location based on my reception direction and house configuration. Drat! I hooked it up and tried it anyway, but the results were disappointing.

So I took it back to the garage, and took a saw to it. A little wire reconfiguring and cutting and I had a SBGH (screen now 30”x45”). I took that back up and started playing around with it to find the best location.

The good news is that I now get several stations from the north (Dayton) much better than I was getting with my old “bi-directional” unreflected antenna. No surprise there. But unfortunately, I am still not getting enough of a lock on that one channel (18) that I wanted to try to add. It comes and goes, but never locks enough to keep it activated in my channel guide. Oh, well.

The other antenna (discussed in some previous posts here) remains where it was and gets all the south (Cincinnati) stations in the high 90s. That one never had dropouts on channel 31 and it still does not.

However, now that I have “gone back” to a SBGH, I would again like to offer my antenna as a "guinea pig" tester for the modifications that this thread will be about. In fact, next time I bring the antenna down to do work, I will likely replace the yaffa-block “screen” with some much finer “hardware cloth” (my original intent, but I did not have enough for the DBGH). The hardware cloth is stiff enough, and the openings small enough, that I can cut it with wire cutters into whatever shape is needed for very detailed “contouring” (although, once it’s cut small, I can’t put it back, so I’ll wait until we get this thing fairly optimized!)

Update:


The other antenna (discussed in some previous posts here) remains where it was and gets all the south (Cincinnati) stations in the high 90s. That one never had dropouts on channel 31 and it still does not.


So I guess your other DB-4 clone with the dropout problem, did have some fabrication problems, since I assume both DB-4's were made the same and you are now using the down-lead of the "problem" DB-4 on the SBGH?
(A bad connections on any one, or more of the 12 connections of the DB-4 clone would give strange reception results).

Would be nice if someone would model these DB-4 ;CM-4221 clones. I've seen some speculative arrangements using up to 8 bow-ties. Modeling these without a screen reflector, is quite easy to do and would provide a lot better guidance, than pure speculation. Any one that wants to try it, just contact me and I could help you.



However, now that I have “gone back” to a SBGH, I would again like to offer my antenna as a "guinea pig" tester for the modifications ...

Thanks for the offer, and once I get the optimized dimensions released, I would look forward to your results.

As I mentioned before, losses in the attic are difficult to predict, and it appears that you have found the "hot-spots" in your attic for the max signal strength.

I assume your SBGH attempts on Ch 18, did not use a preamp at the SBGH's antenna feed-point ? I would expect a good low noise preamp would give you a much better shot at locking onto Ch 18.

The SBGH GPLv3 model results predicts a gain of around 13.3 dBi at Ch18; the v3.0x version has an objective of +1.5 dBi... up to 14.8 dBi, but the results are not yet in to see how close I come to that objective. But since you are in an attic, I'd expect that the v3.0x version would also need a preamp, so you might want to take a serious look into a good low noise UHF preamp, if Ch 18 is a "must" channel.

So I guess your other DB-4 clone with the dropout problem, did have some fabrication problems, since I assume both DB-4's were made the same and you are now using the down-lead of the "problem" DB-4 on the SBGH?
I still think it was a multipath problem that the directional SBGH does not see (both because it has the screen and also because, well, I'm not even trying to get channel 31 with it :) ). Regardless, not a worry anymore.

As I mentioned before, losses in the attic are difficult to predict, and it appears that you have found the "hot-spots" in your attic for the max signal strength.
Yeah, there's pretty much a clear-cut spot for north (Dayton) reception, I had my wife on the computer reception meter at the bottom of the attic ladder, giving me reports as I moved the antenna around. I tried a bunch of different configurations, but ended up back where it was in the first place.

As far as my future results with your contouring, we'll just label it "results from an attic" or some such thing... and perhaps they can be compared to someone else's results outdoors. Of course, outside of computer modeling, everyone's "real world" results will vary according to everything from weather to planes flying by to leaves on trees... but it's always good to hear how various antenna models perform regardless.

I assume your SBGH attempts on Ch 18, did not use a preamp at the SBGH's antenna feed-point ? I would expect a good low noise preamp would give you a much better shot at locking onto Ch 18.

The SBGH GPLv3 model results predicts a gain of around 13.3 dBi at Ch18; the v3.0x version has an objective of +1.5 dBi... up to 14.8 dBi, but the results are not yet in to see how close I come to that objective. But since you are in an attic, I'd expect that the v3.0x version would also need a preamp, so you might want to take a serious look into a good low noise UHF preamp, if Ch 18 is a "must" channel.
Channel 18 is far from a "must have". As I said before, it was just something I wanted "because it is there". It wouldn't be worth buying a preamp for.

But on an "educate me" note, would that really help with only about 10' of coax from the antenna to the tuner? I thought an amp in that situation would only boost the noise (vs. signal).

......
But on an "educate me" note, would that really help with only about 10' of coax from the antenna to the tuner? I thought an amp in that situation would only boost the noise (vs. signal).

pj*4

I am not an expert on pre-amps, but the fact that you get a signal (not lockable) on Ch 18, indicates to me that a pre-amp would likely improve your reception. How much improvement depends on the noise-level of your preamp and your TV tuner. You can't do anything to improve your tuner, but you would improve your reception chances, by selecting a preamp with a low-noise figure.

If you can "borrow" a Channel Master 7777 (a decent low-noise UHF preamp), it would be worth a try, but since it's not a "must" channel no need to shell out for one.

There are likely some forum members that are experts on preamps, that can comment and provide you with a more "educated" view.

What is the boom length for this antenna? And what is the spacing between the driven element and the reflector?

If you are talking about the SBGH2/DBGH2, then these specifications have not been revealed yet (I believe because it's still in the optimization process). If you are talking about the first generation, I suggest you read the thread Generation I Gray-Hoverman Antenna (SBGH & DBGH) (http://www.digitalhome.ca/forum/showthread.php?t=81982). The first post lead to the main page that covers history, performance and detailed plans for the antenna.