I've really been itching to do a quad of DAT 75's. I have a Lindsay 2 way combiner so called up Lindsay to order a 4 way model. Lindsay UHF/VHF combiners
Guess what? They no longer manufacture them. Called fellow AVS'er MaxHD to see if he could help and although he didn't have a source for the 4WCU, he did have a couple Televes ref. 5006's on hand which are basically preamp/uhf combiners all in one. Televes ref. 5006
I figured maybe I could use them with a power passing splitter to do the quad. I ordered the DAT 75's from cpc with the usual 7:00 a.m phone call due to the time difference.
Guess what? The DAT75's now have redesigned larger reflectors (presumably for added gain in the lower uhf). Needless to say, I was a little dissappointed as I wanted 4 identical antennas. I decided to try anyway though. In the meantime, I called/emailed Lindsay again (acting more desperate this time) and an extremely helpful person there found a 4WCU for me
. Through all of this in the last month or so, I've been trying various combinations for this setup and thought I'd share my results here.
4 semi-identical Televes DAT 75's
Research Comm. 9253 HDTV LNA with ps
Televes ref. 5006 x2 with ps x2
Lindsay 2WCU and 4WCU uhf combiners
Lindsay LS2A vhf combiner
power passing wide band uhf splitters of various brands
Initially, using the ref 5006's through a power-passing splitter didn't work very well. The performance overall was less than that of just a horizontal stack with the 2WCU combiner into the RC 9253 LNA. I suppose the phasing just wasn't perfect enough through the splitter. For novices out there, phasing is critical when combining antennas for added gain. The signals need to be identical as possible and in-phase which means everything needs to be symmetrical (i.e. equal signal paths and cable lengths) up to the point of combination.
Using the 2WCU as the power-passing splitter worked better. I'd say about equal in performance to the initial horizontal stack. Incidentally, this configuration isn't actually a quad stack. It's better termed a stack-of-stacks
. No matter what the combination, it seemed that a vertical stack of horizontal stacks
outperformed a horizontal stack of vertical stacks
. Not sure why that is but it wasn't even close.
Using regular resistive splitters, whether 3 dual or one 4-way didn't work at all for me. Others seem to have some success using these, but I can't say they work at all.
Next up, the true quad (actually an H-quad) using the 4WCU into both the ref 5006 (using only a single input) and the RC 9253. The ref 5006 held it's own BTW against the 9253. Not sure if I could tell any major differences. The quad, however, was overall somewhat dissappointing. I tried various stacking distances and it seemed 36 inches square worked the best. There were some channels for which it seemed to perform excellent, but others were poor. I think it may be that it's difficult to phase a true quad properly for a wide band setup and this would work better for narrow-band or single channel setups. I also couldn't really get the antennas much closer than 36 inches square (due to the megareflector on the new DAT75) so it may be that the phasing would have improved at closer distances. I did figure out a way to create an identical 4-antenna array BTW. The reflectors are interchangeable so I tried it with one large and one small on each antenna. It worked, but overall it seemed to worka little better in the pictured configuration below. Also, you'll notice I've added screening to the reflectors. This improves F/B ratio and helps me with co-channel interference I have between an analog and digital 34. Anyhow, that's a whole other story.
So, needless to say, at this point I was quite bummed with the quad stack performance. I then tried using the 2WCU and the 4WCU (used as a 2-way) to combine the stacks through the ref 5006. For some reason, this just wouldn't work. The 2WCU and 4WCU are exactly the same size/length and the only internal difference that I could see is that the 4WCU has 4 f-sockets soldered inside the end of the combiner while the 2WCU only has 2. Obviously, there must be more difference than that as they just wouldn't phase at all. Maybe someone who knows how stripline combiners work can answer this?
So, almost in desperation (again), I decided to try to use the 2WCU and the LS2A to combine each stack. The stacks were then combined through the ref 5006's dual uhf inputs. I compensated for the added length of the 2WCU by adding corresponding length to the cable coming off the LS2A (actually, I tried multiple lengths and settled on 13 inches as the best). Bingo! It worked. Performance now surpassed that of the original horizontal stack throughout the uhf band. It's difficult to quantify exactly, but I'd say my performance now is overall similar from channels 14-38 to my Triax Unix 100 Band A stack. This is excellent performance in this range for a wide-band uhf setup. For channels 39 and up, I'd say better by a small margin over the original DAT 75 stack.
Obviously, the next logical step would be to replace the LS2A with another 2WCU. However, I'm not sure when one will turn up and my WAF doing all this is currently on empty.
I'll keep you posted.