|
91 XG Modified
So I made lemonade out of lemons today. I ordered a couple of 91XGs through Amazon last week. I was excited when I got home from work, tore the Box open and found at the UPS handling system destroyed part of the antennas. The booms were bent. When I called Amazon to get the antennas replaced they said they would send two more and I could keep the bent ones.
I took that smashed up boom from the 91XG and I straightened it. It cracked in a couple places, not all the way through. I took a piece of half inch "C" channel and slid it in to the cracked booms. Better than new now. I also made a new support boom out of my leftover 1 inch square stock with the .125" wall. It's overkill, just the way I like it! Between the "C" Channel inside the long booms and the under boom support, an amazingly strong antenna. I think it will hold up well in the wind. I did several modifications as I went along. My first mod was to take an extra Center section, remove the front section and add that in between the original Center section and the front section. The signal was Stronger on some channels. Then I added an extra front boom. The results were dramatic. I was very happy with the results. I still have another short rear section that could be added for now I'm happy with what I've got. I'll let the pictures tell the rest of the story. https://i.ibb.co/mJKJNsQ/20200531-211251.jpg https://i.ibb.co/z8xGMwC/20200531-211342.jpg https://i.ibb.co/kBsbNv0/20200531-211306.jpg Here's the modification to the support boom. The riveted end is for stability. https://i.ibb.co/f0kXB91/20200531-205225.jpg This end can slide in and out, I found it necessary as the spacing on the directors on the 91 XG is tight: https://i.ibb.co/qd0TXcL/20200531-205205.jpg https://i.ibb.co/PcHjz0H/20200531-204901.jpg Boom sag before adding the enhanced boom support: https://i.ibb.co/4pjVs1z/20200531-172253.jpg The last two pics just show the "C" channel that I inserted to reinforce the antenna booms: https://i.ibb.co/r0kFyG8/20200531-202338.jpg https://i.ibb.co/HPKNPjt/20200531-202049.jpg https://i.ibb.co/HB3Zv2b/20200531-202031.jpg ] |
91XG Spectrum analyzer results
Well, here's the Spectrum analyzer shots. In summary that signal strength went way up. Later this week when my two replacement 91xg scum in I will test them against this modified 91XG antenna
https://i.ibb.co/5njy0Qt/91-XG-Vs-Mo...irectors-1.jpg https://i.ibb.co/gyBLvcH/91-XG-Vs-Mo...ction-only.jpg [url=https://ibb.co/k3mVT8q]https://i.ibb.co/KNKcgqV/91-XG-Vs-Mo...tors-night.jpg[/url |
Nice work, Bob.
Quote:
Wonder if Home Depot or Lowes has it? Quote:
A few years ago, did a similar thing with MCM 30-2370's, but used round tubing inside the boom. Think it was 3/4 inch square tube to join and reinforce the cradles. It is located in KY, and has seen some ice and high wind. . |
Thanks!
We'll call it a 135XG.
Quote:
Quote:
http://forum.tvfool.com/images/editor/attach.gif I got it at Home Depot, its for edging or framing plywood. 8 feet was $11.00 with taxes. It's 1/2 inch and slid in quite easily into the 91XG booms except the ones that had gotten bent: https://i.ibb.co/x80ZPLq/20200527-17...y-1008x477.jpg Quote:
. |
Super "163 XG"
I believe I've achieved my final modification on my Super XG. As you remember I had Taken 2 91XGs and added sections from each other to make a longer "135 XG". It worked well and I probably could have just stopped there.
I couldn't help myself. I saw the extra rear section closest to the dipole left over from the other donor antenna. I looked at spacing and sure enough if you removed every other reflector you could achieve the five and a half inch spacing of the front sections. I actually only had to drill a few holes for the mounting clamps and another because the director spacing wasn't optimal. After flipping this rear section around and playing with the reflectors I was able to add the rear section to the front. We will call it the "163 XG": https://i.ibb.co/rHnBJbk/20200607-093657.jpg Results were good. I think I have reached the point of diminishing returns. I also know I'm not testing under optimal laboratory type conditions. So I'll do back-to-back scans I tried to keep the test within 10 minutes of each other. Then I'll wait a while and do another back to back scan or even do it the next day. I'll test one and then the other and then the other guy back and forth scales with the same I can say to myself , yeah they are consistent. I always use the same components. Same preamp, same down cable, even the same UVSJ. https://i.ibb.co/g3g0Dd3/Super-XG163...ing-TEST-2.png Next day: https://i.ibb.co/mywpksb/Super-XG163...st-morning.png There was quite a difference between yesterday and today. Although both days were clear, there may have still been some atmospheric conditions contributing to the strength of the scans I did this morning. Look at the 163XGs signals in less than 24 hours: https://i.ibb.co/QXPxfrn/Super-XG163...erent-days.png |
Quote:
Quote:
Still, every small, or not so small, amount of gain adds up. Quote:
Have noticed that stations that are LOS (in Dallas) are most stable between hours of about noon and mid-afternoon. For stations that are considered '2-Edge' and 'tropo', they (in KY) are generally weaker in that same time range. Quote:
----------------------------------- You are discovering the tradeoffs of lengthening Yagis vs stacking. Each has some advantage. For example: Ideally, stacking gain could approach 3dB for all the channels the antennas were designed for. Lengthening a Yagi generally increases gain the most at the highest channels. provides less increase in gain for lower channels. Contrary to popular opinion, doubling length of a high-gain broadband Yagi does not double gain (3dB) for all channels. For theoretical reasons, doubling length of a normal high-gain Yagi will not double gain for even the high channels. ----------------------------------- Depending on individual circumstances , there may be good reasons to prefer a long Yagi over a stacked pair, or vice-versa. You are doing some interesting projects. . |
Those are some supersized antennas Bob!
|
Long antennas in the wind
Quote:
https://www.dropbox.com/s/xay184dha9...21405.mp4?dl=0 |
Thanks Tripelo !
Quote:
Quote:
Yes. The gain is there and I am happy with the mods...For now :) Quote:
https://www.rabbitears.info/searchma...tudy_id=103278 Quote:
Quote:
Such was the case :) ----------------------------------- Quote:
----------------------------------- Quote:
I do have 3 brand new 91 XGs ready for testing and one other I can assemble. I am tempted to play with a quad array but.... I don't know yet, I am going to perfect my dual stack first. I want to attempt the twin lead lossless balun though. Believe me, if I could have made the 1/4 wave transformer work, I would have. I will admit, that was really painful. I think the twin lead endeavor is my best hope for low loss stacking. I will keep you all informed! https://i.ibb.co/34bQmkR/FB-IMG-1591310264869.jpg Special ordered the ferrite for the loss less balun two different types, one for VHF and one for UHF: https://i.ibb.co/HCYpky0/20200609-145158.jpg This is all new to me. Any tips? |
Quote:
Looks like you have several strong signal stations to choose from. Quote:
Quote:
There is no such thing as a lossless balun* (See Note at bottom). Quote:
Quote:
Quote:
Quote:
In general, the more ferrite you use, the less loss. One of the main goals is: To impede signal loss down the outside of the coax shield by raising impedance as high as practical on the outside of the coax shield. For example: If one is using only two ferrite cores, it may be best to use one at the junction and one about a quarter wavelength down the coax. If three cores, maybe two at the junction and one a quarter wavelength away. There are other combinations. Have implemented the separated cores in test installations. Haven't actually tried to measured improvement results, but have run computer simulations that indicate it is better than putting all the cores at the junction. -------------------- If you go with a quad stack: - Need to decide of what configuration you want: 1. Diamond Configuration 2. H Configuration 3. All 4 horizontal 4 All 4 Vertical As usual, some configurations may be more desirable than others. Be interesting to see how you handle some of difficulties related to use of twin-lead. Twin lead is a conductor, depending on location, lengths that are longer than about quarter wavelength can resonate and affect antenna patterns. . ------------------------ Note, baluns and loss: Some antennas can be built such that the the transformation of 'balanced to unbalanced' occurs in the antenna. Log periodic antennas can be built in this manner. Some specialized folded dipoles can be configured to make the transformation. As such, these transformations are more efficient than most any other method. There are many other types of antennas that provide for an unbalanced (coax) connection and require no external balun. |
Great information, thanks Tripleo
Tripleo, thank you for your detailed information. It's exactly what I need. Again as you see I have an eye for detail and I will report each and every change I make. Not just for my benefit or yours but for anybody who may try this. I know it sounds simple but for me it's like working on a new brand of car with new tools and a new scanner.
For now I plan to just try the two antenna stacking side by side. I want to do two 91XGs with the 300 ohm twin lead separate from two 30 - 2476 VHF antennas with a similar. I mean side by side and using the twin lead. When I am done and I am satisfied I have made good progress , I hope to join them at a combining preamp. I've got a good friend making a low noise VHF UHF combo amp. So, to stack each array antennas I believe I have ordered the correct ferrite cores. I have 10 of each number, which are Fair-Rite "Ferrite cable cores 61 Z= 280 ohm at 250 megahertz. I believe those are for UHF. The other ones are the Fair-Rite "Ferrite cable core 43 round cable cores Z = 280 ohms at 100 megahertz". I think those are for VHF. The plan was to put four on the RG6 coax of the VHF array and four on the RG6 coax of the UHF array with six leftover of each. I'm 100% sure I should have no problem soldering the twin lead to the coax. Hey, a question came up should I be using dual Shield or quad Shield RG6? Another friend Sev, is going to do with the similar or the same. The plan was to twist the cable 1 twist per foot or was it two? I'll check my notes. Then put it inside 1/2 " CPVC tubing to keep the rain off of it. That's the plan I would like to hear any input that you or anyone else has. This is new to me, and I am just learning how to do this. Oh, and I like your idea of spreading out the ferrite cores. I don't know what the distance would be when you say 1/4 wave, I don't know the calculations. If it's easy for you, I'd like to hear your input on the lengths I need to work with. I don't mind learning all these calculations but math isn't my strong point. Sure, I know decimals fractions and multiplication / division. But I don't know formulas math and I parted ways at algebra and trigonometry.. Thanks, hope to hear back from you before the weekend when I plan to do this. We're supposed to have some beautiful cooler weather perfect rooftop weather. I'll start with my scans to make sure the two antennas are equal. Which I'm sure will take me several scans to perform an average. |
Ferrite, Wavelength, etc.
1 Attachment(s)
Quote:
The choke balun does not make impedance transformation as regular 300:75 Ohm baluns would. For example; two 300-Ohm antennas paralleled would provide about 150 Ohms impedance at the junction. Then a 150:75 Ohm transformation is needed. Then the ferrite choke balun could be applied. Another example; The twin lead from four 300 Ohm identical antennas could be be paralleled to provide about 75 Ohms impedance, then the ferrite choke balun applied. Quote:
Haven't looked at your particular ferrites. It is best to select for maximum impedance at the frequencies of interest. For UHF, In future you might want take a look at Laird (PN 28B0616-000), it is solid and fits over regular RG6 (not quad), provides more than 500 Ohms impedance at 500 MHz. It has fairly high impedance at upper VHF as well. Maybe the Fair-Rites you have are similar, one could compare graphs. http://forum.tvfool.com/attachment.p...8&d=1591958091 The Laird ferrites are available at DigiKey. Think Mouser has them also. Laird also has some other sizes that are convenient, such as to fit over Connectors, etc. Quote:
Quote:
Optimal twist depends on frequencies to be rejected. It is usually not critical. Quote:
The speed of light is about 300 million meters per second. The frequency is in Hertz, that's cycles per second (cycles being wavelengths). The length of a wavelength is the distance traveled in one second, divided by the number of cycles in one second. For example: At 535 MHz, that’s 535 million cycles per second, that’s also 535 million wavelengths per second. Divide 300 million meters per second by 535 million cycles per second The ‘per seconds’ apply to both numerator and denominator so they cancel. The ‘million’ apply to both numerator and denominator, so they cancel. That leaves meters as the units. 300/535=0.56 meters There are 39.37 inches per meter. 0.56 meters ~= 22 inches One fourth of 22 inches ~ = 5.5 inches There are two numbers to remember; Speed of light and number of inches in a meter. It is not necessary to apply a velocity factor. Since signal of interest is traveling on the coax shield, then conductor is outside the coaxial cable and the signal travels at nearly the speed of light. If inside a coaxial cable, the velocity of propagation slows. The slowed signal can travel at speeds in the range of about 66% to maybe 85% for common coax. For twinlead the slowing effect is less pronounced, maybe in the range of 80-95%. Because the ferrites are an inch or more in length, that occupies a significant portion of a wavelength at UHF, so the precise location of a quarter wavelength is not critical. For upper VHF, multiply the above number by 3 might be close enough, or recalculate to get about; 15 inches for quarter wavelength. Or, Plug frequencies in to this (or other) online wavelength calculator: http://www.procato.com/calculator-wavelength-frequency/ . |
Quote:
The harness he is attempting emulate is based on the one Calveras built some time back. I am working on assembling a couple of quad arrays in diamond form using the same method. The beads selected are 61 for UHF and 43 for VHF if I recall correctly off the top of my head. From the Mouser catalog. These were the sizes suggested by Calveras. His harness was 4 phase matched lengths of twin soldered to a length of Belden RG6 with 4 ferrite beads back to back starting at the soldered juncture. For my purposes I will be using 4 DB4E panels for 1 quad array and 4 yagi's from Newark that are the HDB91x sold on solid signal I recently came into a NOS Channel Master Spartan 3 0064DSB with the 300 Ohm input. I am hoping this will allow me to forgo the need for the beads and coax as the output on the Spartan is 75 Ohm. |
1 Attachment(s)
Good morning Sev.
Quote:
The core and windings are seen in the lower right of this image (near input terminals). http://forum.tvfool.com/attachment.p...9&d=1591972624 Could be CM used a slightly better core than used in most commercial baluns. However, the balun has loss. Removed the core and windings, replaced with half-wave loop saw signal about 1-2 dB strength improvement. . No offense Sev, the post was quick and for a while was thinking about Bob's earlier posts. |
Quote:
By the by. Perhaps you would know. As anybody ever added the missing components on that board to switch it to a 0264? Would have have a picture of the 0264 board? I have yet to come across one. |
CM-0264 Preamp
1 Attachment(s)
Quote:
Main item is additional balun. PCB traces are already there. Need to cut an etch connecting UHF & VHF, etc. Probably have to solder in a small jumper. Maybe touch up tuning on input coils. Quote:
http://forum.tvfool.com/attachment.p...1&d=1591980748 . |
Ferrite cores
Tripelo, thanks.
I am happy that you have given me this info :) My head is spinning but I will recover. Here are the Ferrite cores ordered from Mouser: UHF: https://www.mouser.com/ProductDetail...smXY4HC8Iew%3D VHF: https://www.mouser.com/ProductDetail...tkRaXC%2F3o%3D Man, they are so very careful about the packaging. I guess they are very brittle. Are you thinking I only need two Ferrite cores or, the more the merrier? At least one at the junction and at least one at the 1/4 wave length, 5.5" down? I have 10 of each of the above. I need to stop and get some more shrink wrap |
Quote:
Quote:
Quote:
Quote:
Four of them together at 1.126 inches length each would cover about 4.5 inches. Splitting for quarter wave leaves only a small space between. The signal can divide some portion to pass through the interior of the coax at 75 Ohms impedance, some to pass through 1600+ Ohms (plus the inherent impedance down a coax shield outer conductor). The point is, the higher the impedance, the less signal passes on the outer shield. Signal on the outside of the shield is for practical purposes 'lost'. There are practical limits, cost, weight, difficulty handling and securing a ferrite laden cable. In some cases, two or three is about all a connection at a junction can handle. Diminishing returns, revisited. . |
Diminishing Returns and
Quote:
Quote:
And what about the VHF? That should be a longer 1/4 wave right? How should I configure the ferrite on that coax? Quote:
In my case, the cost is low so that's not an issue. So, 2 or 3 is enough? 4 is OK, but wont do much? As in "diminishing returns"?... . |
Quote:
Two ferrites, next to the junction and two ferrites centered over the 5.5 inch mark should be fine. That would mean the edge of one ferrite on the antenna side of a 5.5 inch mark and another one next to it down the coax. Quote:
Quote:
If you use three, then place two at the junction and one at 15 inches. Quote:
After two or three ferrite cores on a cable, likely the signal strength improvement available with more ferrite is in the low tenths of a dB range. . |
Quote:
Looks like the L7A is the bridge. Or no? https://scontent-atl3-1.xx.fbcdn.net...fc&oe=5F0899E6 |
Quote:
It appears L7 is meant to move to L7a location for the 0264 version. Your preamp is probably newer. . |
The "Less Loss" harness begins...
Quote:
Wow ! OK, I have to say I am pretty fired up! My ADD (Antenna Distraction Disorder) is kicking in. You have made it very clear for me- a novice. This will help others too I'm sure, if they take the time to read it. Oh, the twists in the twin Lead, Sev was telling me, it's up to me how many twists, just make sure they are the same amount per side. That may be the trickiest part. |
Quote:
Seems like I always pick something complicated to do and it's hard to get input that way. So, I really appreciate it. |
Quote:
Bare copper vs coated copper. It also would appear they tuned the G1 next to the trap adjustment. It has a silicon strip across it to hold the coils in place. If somebody had a CNC machine to mill the traces. That board would be pretty easy to replicate after inputting the necessary data. Then it would be a matter determining the ferrite core. Or install half wave loops and sourcing the rest of the components. It would be a nice challenge for a hobbyist. The units have 3.0 and 2.2 dB in noise. I wonder if it would be possible to determine and optimize the components creating the noise. Jeff Kitz has shown that it is possible to create low noise preamp. Then there is Winegards LNA-100. Also very low noise. I would imagine a portion of the noise is being generated by the PSU. The 3 capacitors @ C7, C13 and C28 would be suspect as well. |
Quote:
"OK, I understand about the ferrite to form a choke balun around the coax, but there are two questions I have in mind: 1. If you connect two 300 twinleads together in parallel, that would give you 150 ohms. Wouldn't that be a 2:1 missmatch with the 75 ohm coax? 2. When you connect the two twinleads in parallel, they must be in phase. It's like connecting two batteries in parallel: + to + and - to -. How will you know which way is correct?It will work, but there will be some loss because of the mismatch which causes an SWR of 2:1. It is difficult for me to predict how much additional gain two antennas connected that way will give you compared to just one antenna. There are two problems to be concerned about when you connect together in parallel the two 300 ohm lines from two antennas. The first problem is the mismatch loss because two 300 ohm lines in parallel will give you 150 ohms, which you will then connect to a 75 ohm coax line. You aren't going to like to hear this, but you would need a 1/4 wave matching section of 106 ohms between the 150 ohm point and the 75 ohm coax for a perfect match. The second problem is getting the two 300 ohms lines in phase (or polarity if you prefer) when they are connected together in parallel. You have a 50/50 chance of getting it right. If they are connected out of phase the main lobe will split in two like this" https://imageevent.com/holl_ands/sta...f5v5sec1.goose Ooops... looks like I have another problem to solve? Now that we're in the 11th hour I'm so glad that I did not execute my plan fully. I'm kind of glad it was chilly Friday after work and I had to do some deck work around the house. That kept me off the roof! Anyhow what should I do? What do I do to achieve the correct match? |
Quote:
Channel Master and some other manufacturers have not provided accurate noise figure numbers. Numbers often provided are quotes from transistor manufacturers specs. There is more to noise figure than transistor specs, for example: - The excellent input filtering of the 0264/0064 and other older CM preamps, contribute to the noise figure. - The loss in the ferrite balun adds to the noise figure. Quote:
The first UHF transistor is a BFG67 (1st transistor is usually main active source of noise) ------------------------- The 0064 & 0264 are good and reliable preamplifiers. The design, except for input baluns, is virtually the same as for the old CM7778. The old CM7778 was cloned to the RCA_TVPRAMP1R. http://forum.tvfool.com/showpost.php...76&postcount=2 . |
Quote:
I haven't tried it in situations like this thread is addressing, but I suspect it might be worth experimenting to lower noise. One advantage of non-polarized CAPS is that is doesn't matter which way you solder them in. Also, they're not too expensive as compared to regular CAPS and can still be bought at places like Mouser Electronics: https://www.mouser.com/Search/Refine...zed+capacitors You Techs here are above my pay grade, but I hope this might be of some help. |
Quote:
This image popped up on google. A thread with you and Pete Higgins. https://forum.tvfool.com/showthread.php?t=13530 Do you know of any alternative componets that would lower the noise figure. Or based on the design it is baked in? https://forum.tvfool.com/attachment....7&d=1376151612 One of the members a long time ago mentioned to me that the RCA was a clone of the 7778. The first preamp I picked up actually. So based on Calveras's testing. The CM's and RCA's had a noise figure very similar to the Blonder Tongue Galaxy III's. Assuming that B.T. was being truthful about their noise levels. I have come across a few B.T. 300ohm dual inputs with the 75ohm output but passed no them because there did not appear to be much love for them due to their advertised noise figures. Any idea if those figures are actually accurate or not? To clarify a statement back on post #14 you stated you removed the ferrite bead and replaced it with a half-wave loop on a Spartan? OTAFAN posted an interesting idea concerning replacing the polarized caps with non polarized. Thoughts? Modding the board should not be very difficult. Though I would like to find another Spartan to test on. |
Quote:
Interesting idea. Need more data on it to see what the possibility is. I knew guys years ago that were audiophiles. They were just as bad if not worse than the car guys trying to squeak out every very last bit of HP out of their engines. |
1 Attachment(s)
Quote:
Quote:
It might lower noise figure a couple of tenths of a dB. Prrobably need to retune. The transistor substitution works fine, did that when replacing balun. Quote:
The older 0064/0264 have about the same noise figure as the BT Galaxy series. Quote:
http://forum.tvfool.com/attachment.p...2&d=1592234723 The coax is RG-180 (95 Ohm). The impedance is not ideal, but in some ways better than 75 Ohm coax. It could be better if the impedance for the loop was 150 Ohms. But, such cable would have to be custom made and it would probably be too large to fit in the case. Quote:
Some audio circuit paths require passage of bipolar signals (plus & minus). Audio signals are very low frequency relative to RF (long wavelength, calculate wavelength for say 20 Hertz), therefore large capacitors are needed. All bipolar signals in an RF amplifier pass through small ceramic capacitors, they are essentially nonpolar. The polarized capacitors in RF amplifiers are for filtering the DC (not bipolar) supply voltages. The noise that is bothersome in preamplifiers is usually in the same frequency range as the desired signals (MHz). . |
Quote:
I assume that 150ohm Belden Twinax would not be appropriate for use for the half loop? |
Quote:
It probably could be made to work for a half wave loop balun. but would require some unconventional considerations. In general, coax is better for the half wave loop. --------------- Referring to the 0064 with half-wave balun in previous image. The input & output of VHF portion was disconnected and powered down. There is some noise figure advantage in not having a VHF circuit connected to a UHF circuit. Anyhow, that balun would not work well for VHF. --------------- FYI, will not be logging in frequently for a while. . |
Quote:
Hope all is well on your end. |
| All times are GMT. The time now is 4:32 PM. |
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2025, vBulletin Solutions Inc.
Copyright © TV Fool, LLC