Saturday, September 12, 2009

Finally Confirmed - The QDFA Blows The Beverage Away

Yes, it does. At nighttime, when strong signals from Europe dominate the band, the QDFA has proven to be totally superior. In a large number of instances where EU QRM totally obliterated the NA frequency when using the beverage, the QDFA either brought audible levels, or even readable levels. If I had audible levels with the beverage, the QDFA produced readable levels. And in every instance where the beverage had readable levels from North America, the QDFA produced better readability.

There is no going away from this. Check it out yourself. These 10-second recordings from 1510, 1520, 1540, 1550, 1560, 1570 and 1600 should provide enough evidence. The first 5 seconds are recorded off the beverage, the last 5 off the QDFA.

So what brought these dramatic changes from last weekend? I don't know for sure. Maybe the radials I put out worked. Maybe propagation didn't favour the properties of the QDFA. Maybe there is some other obscure reason. My unqualified guess is that the radials worked. Btw I added another set of radials just before it got dark; they run perpendicular to the loop.

Of course this doesn't change the fact that the QDFA sensitivity rolls off from around 1000 kHz and downwards, and the fact that when the interference is reduced by approaching daylight the beverage is always more effective. But we knew that all along. The QDFA is set up for being able to DX the difficult midnight to dawn period. Thanks Dallas Lankford for this extraordinary experience!

Update: It appears that the effect of the QDFA bleeds away when civil twilight approaches. 45 minutes before sunrise (twilight lasts much longer at northern latitudes) there is virtually no difference between the QDFA and the beverage, although EU interference is still quite heavy.

14 comments:

Dallas Lankford said...

Congratulations Bjarne!!! Great job!!! It is virtually certain that you have identified the cause of the no-null problem for the K-QDFA (poor ground, namely the rocky ground of Kongsfjord) and fixed it (radials). Yes, as you observed, the QDFA nulls are poor or unstable during sunset and sunrise “transition.” In my experience, this is a feature of all highly directional phased arrays, whether electronically null steered (two antennas phased with a phaser) or fixed phase shifts (like the QDFA), and whether here in North Louisiana, at Grayland, at Quoddy Head, or at Kongsfjord. According to what others have told me, terminated beverages also have poor or unstable nulls during sunset and sunrise transition. The QDFA sensitivity at the lower MW frequencies can probably be fixed with additional preamps inserted at certain junctions in the QDFA phaser. I believe I know how to do it, but it will take some time to implement and test my ideas. This work will probably not be completed this fall because it will require the design and development of new small inexpensive preamps. Meanwhile, enjoy your new toy, the K-QDFA. I wish I could drive a few miles from my house to a wonderful MW DX sight like yours.

Bjarne Mjelde said...

Thanks for giving us this wonderful new toy to play with, Dallas!

Guy Atkins said...

Hi Bjarne! This is exciting news, and I'm very glad to hear you are now experiencing the same amazing reception improvement that I experienced last spring with Dallas' QDFA. Your recordings are clear proof; I hope other DXers will be encouraged enough by our results to make the effort to construct a QDFA, or experiment with a variant.

What are the lengths of the radials you have attached? Do they connect to the horizontal leg of the delta loops at the primary winding, or one of the sides of the secondary winding?

Bjarne Mjelde said...

Thanks Guy. Yes, this antenna deserves a wider audience. The radials are roughly the same length as the horisontal element of the loop, and they are centered in a star formation in each loop's center. So there are four sets of four radials.

Guy Atkins said...

Thanks Bjarne. I think I understand your placement of the radials. However, are they electrically connected to the loop, or physically apart, such as a "reflector" element in a cubical quad, etc.?

Bjarne Mjelde said...

Guy, they are not electrically connected to the loop elements.

Walter said...

Once again, after listening to those examples, I'm thrilled to know that I'll soon have a QDFA as well. I'm concerned a bit about the comments about the dawn period. This is my favourite time to listen. In Masset on the Queen Charlottes (fairly high latitudes), the Beverage is a fantastic performer at LSR. I admit, I've never had very good luck hearing much overnight there, partly due, I suppose, to so much domestic splatter. The QDFA should help this for certain. The only other array that has been tested up there is the Wellbrook phased ALA 100 delta loops (2 of them). We had 2 of them up simultaneously in Sept, 2007 and found the results stellar before dawn, but not that great a LSR and post LSR. Having said this, John Bryant has very successfully used this array down here on Orcas Island at LSR and post dawn. I've never heard a good explanation on the difference between our two sites (make that 3...also did very well at Grayland), apart from the difference in Latitudes. I'm aware that the QDFA does not like any Beverages close by. Before I take down and stop using my beloved Beverages, I want to make sure that the QDFA works as well during and after LSR!

Dallas Lankford said...

Hi Walter, I understand your concerns about the QDFA performance during sunrise transition and post sunrise. Unfortunately, how well the QDFA performs during that time period may depend on where the QDFA is located. At Kongsfjord, as Bjarne reported, the QDFA did not do well during sunrise transition. But during the tests at Grayland I had nice receptions of Fiji 639, Guam 801, the Philippines DXAM 1017, and others during sunrise transition. The Wellbrook dual loop phased array has also experienced mixed performance at different locations. For example, John Brynt said, "What about this deafness in the extended pre-sunset or post-dawn that we noted at the Queen Charlotte DXpedition? Let me tell you, I was prepared to stay up all morning to quantify the performance of each antenna to the bitter end. I was shocked to find that the deafness noted at QCI just DID NOT EXIST at Grayland. In fact, the [Wellbrook]Arrays continued to outperform the Beverages right through until the bitter end, 90 minutes after sun-up!" Does the QCI site where the Wellbrook dual loop arrays were used have a poor (rocky, or whatever) ground? We are still learning the pros and cons of the QDFA, and this would be useful information to have. Or perhaps it could be some kind of high latitude effect which caused the anomalous performance of the Wellbrook array and the QDFA. Best regards, Dallas

Dallas Lankford said...

I spent considerable time this morning comparing the Wellbrook dual array recordings which Guy made at Grayland with the QDFA recordings which I made at Grayland during sunrise transition. The QDFA was not fully functional when Guy's recordings were made, but it did perform reasonably well in the upper half of the MW band even when "crippled". I listened on seven frequencies above 1000 kHz. On the whole the QDFA recovered more and clearer DX audio than the Wellbrook aray during sunrise transition, though on some occasions the Wellbrook was better. Since John Bryant (excuse me for misspelling your name before) established previously that the Wellbrook array was on the average equal to or better than Grayland beverages (typically about 700 feet long), it follows that the QDFA would have generally been better than the beverages at Grayland. So the most likely reason the QDFA did not perform well during sunrise transition at Kongsfjord is the rocky ground at Kongsfjord. Of course, some other reason(s) could also account for the QDFA sunrise transition anomaly at Kongsfjord.

And because Bjarne had found the QDFA insensitive in the lower MW band, I also spent considerable time listening to the recordings to determine if there were any differences in sensitivities of the Wellbrook array and QDFA at Grayland. Direct comparisons could not be made because the QDFA lower frequencies were crippled while the Wellbrook array was at Grayland. Indirect comparisons of the QDFA recordings from the night of 4/22 after the QDFA was fixed were made with Wellbrook aray recordings from the night of 4/19, and in those cases no QDFA insensitivity compared to the Wellbrook array in the lower MW band was observed. Moreover, on the one night the QDFA was fully operational, it produced a beautiful dawn enhanced reception of Fiji 639 while the Wellbrook array did not hear Fiji 639 on either night it was at Grayland. Based on these things it is possible that the low band insensitivity of the QDFA at Kongsfjord is due to the rocky ground at Kongsfjord. But it is also possible that man made (and/or atmospheric) noise is much lower at Kongsfjord than at Grayland, in which case the insensitivity could be due to the QDFA noise floor.

The QDFA used two 10.8 dB gain push-pull Norton amplifiers, one at the phaser near the center of the QDFA, the other at the Perseus receiver. I do not know what the gain of the preamps in the Wellbrook array heads is, perhaps 17 dB. In any case, the noise floors of the QDFA (with 21.2 dB preamp gain) and Wellbrook array were equal just below the low end of the MW band, while the QDFA noise floor was a few dB higher than the Wellbrook at the high end of the MW band.

Bjarne Mjelde said...

Part of the Grayland equation is that the ALA-100 array is equal to or better than the 700-ft beverage. While I haven't compared an ALA-100 array (2 loops) with the beverages in KOngsfjord, I have made some comparisons with a single ALA-100 and the beverages. No contest. The beverages are far better. So, I'm inclined to believe that a beverage in Kongsfjord performs better, relative to the QDFA, than in Grayland. But all this is theories. What I should do is to get my two ALA-100 loops out and use them with the Misek-Lankford phaser. That would make for a true comparison.

Dallas Lankford said...

Hi Bjarne, As you said, "No contest. The beverages are far better." (than a single ALA-100 at Kongsfjord), but could you provide some explanation? Was the ALA-100 less sensitive (say, in the lower MW band), or did the beverages give greater splatter reduction, or whatever else might be relevant. Best regards, Dallas

Bjarne Mjelde said...

Dallas: It's been a while now since I tested the antennas side by side (not literally), but what I recall is: Directivity (of course) and gain. I do not remember if sensitivity rolled off like the QDFA or not. A single ALA-100 is by design equally sensitive on both ends, so it is of course unfair to compare a single ALA-100 with a dual ALA-100 phased array. But for sheer gain the difference would be modest and for directivity none (or so I believe).

Chuck Hutton said...

This is good news, and I've started construction of a QDFA with the hope of comparing it to a real Beverage (1300') at Grayland this fall. I expect better rejection of domestics. It seems from your recordings that the signal part of the S/N is also distinctly better which I did not expect and testing of that will be interesting.

Chuck

Dallas Lankford said...

Let me remind everyone that nearby antennas will almost certainly ruin the QDFA null; see my Kongsfjord article for more information on that subject. Also, nearby beverages at QCI were probably why the Wellbrook array did not work well there for post sunrise listening.