AVS Forum banner

Status
Not open for further replies.
1 - 20 of 29 Posts

·
Registered
Joined
·
42 Posts
Discussion Starter · #1 ·
A discussion regarding SF Bay Area DTV stations I started in the HDTV Reception Forum has evolved into some discussions on the Double Bowtie Antenna and then about baluns, so I thought I would start a thread in this hardware forum to get some input on the topic of the baluns. (I apologize if this has been discussed before, but I did a search and couldn't find anything regarding the issue I'm about to bring up on the baluns).


It seems the RS Double Bowtie antenna is very popular in this community due to its low cost and excellent performance. And since it has a 300 Ohm twin lead cable interface, a 300 Ohm to 75 Ohm Coax balun is required.


It also seems a lot of people simply use inexpensive baluns such as the RS #15-1140 or the RCA VH54 .


You can find specs for the 15-1140 on RS's website , and the insertion loss figure given for the UHF band is specified as 7 dB (!!) max. This means you can lose up to 80% of your power using one of these baluns! Someone at station KYES (UPN5) in Alaska has also pointed out how bad these baluns can be and his measurements also show losses as high as 80% for these types of balun. He also states that the Channel Master balun is suppose to be a "good" one. He asked Channel Master about it and received an email from Wayne Massengill @ CMNC.com, stating that "The loss is 0.6dB VHF and 1.5 on UHF." That would be pretty impressive if it was true across the entire band. Of course, it could be best case at a single point in each band, and the rest of the band is much worse.


My questions are: How well known is it that these baluns are so bad? Has anyone seen or does anyone have specs on these common types of baluns besides RS's? Has anyone measured Channel Master's (or can anyone do it) to see how good it really is, and if it really is bettter than the other inexpensive ones?


It would seem if you are using the RS or RCA or similar type balun, you are better off making your own co-ax balun ( http://www.atechfabrication.com/prod...oax_baluns.htm and http://n-lemma.com/calcs/dipole/balun.htm .. Thanks to Sean Kelly for pointing these out). If you design the balun to be matched at mid-band between the lowest and highest UHF station you are interested in receiving, I would imagine even at the band edges, your overall insertion loss is much lower than than the cheap commercial baluns.
 

·
Registered
Joined
·
42 Posts
Discussion Starter · #3 ·
By the way, I included the link to the A-Tech baluns just to show what a co-ax balun might look like. Without any specialized skills you should be able to make them yourselves. The leads for the 300 Ohm balanced side can be much closer together than the ones shown. Make sure you have good electrical connection between your outer conductor connections. I personally wouldn't pay $30 for one.
 

·
Registered
Joined
·
1,525 Posts
I have found that the actual length should be matched to the antenna. The calculated length of 7.43" for 650MHz with .82 vf is a good starting point, but I have found around 6.00" to work better on most of the antennas I have tested them on. This is measured from the point where the shielding begins. You need to seal them with silicone to prevent water absorption.
 

·
Registered
Joined
·
2,814 Posts
Ri-Chee:


Thanks for the post. This is very interesting for me but my problem may be slightly different. I'm having trouble with hum in my HT, which is caused from a cable TV connection. I've had the cable company look at it and have told them about a device which converts 75 ohms to 300 ohms and back to 75 ohms. This apparently will get rid of the hum, which some people call a ground loop. I'm in a brand new housing development and don't understand why the problem even exists, but it's there nonetheless. The cable company seems to have no clue about these things. If I unplug my cable from the wall outlet, the hum goes away.


My question is: Do any of the baluns mentioned in your post do the 75>300>75 ohm conversion? I’ve also tried my local Radio Shack and they have no idea what I’m talking about.


Thanks in advance for any tips.
 

·
Registered
Joined
·
129 Posts
Quote:
Thanks for the post. This is very interesting for me but my problem may be slightly different. I'm having trouble with hum in my HT, which is caused from a cable TV connection. I've had the cable company look at it and have told them about a device which converts 75 ohms to 300 ohms and back to 75 ohms. This apparently will get rid of the hum, which some people call a ground loop. I'm in a brand new housing development and don't understand why the problem even exists, but it's there nonetheless. The cable company seems to have no clue about these things. If I unplug my cable from the wall outlet, the hum goes away.


My question is: Do any of the baluns mentioned in your post do the 75>300>75 ohm conversion? I’ve also tried my local Radio Shack and they have no idea what I’m talking about.


Thanks in advance for any tips.
I've had this problem in my setup before too. All you need is two

slightly different 75>300 ohm baluns. They sell both at the "shack",

they just don't realize you need two different things.


One has a F-Connector female and a 300 ohm pigtail. This is usually

sold to let you connect an "old tv" to a 75 ohm cable, or as mentioned

in this thread, for using 75 ohm cable on an antenna. (Radio Shack

# 15-1140)


The other one you need has a F-connector male and two screw terminals. This one is made for connecting a 300 ohm antenna lead

to a "new tv", i.e. one that only has an 75 ohm input. (Radio Shack #15-1253).


(You can see both on Radioshack.com, under the "indoor" tv stuff).


All you do is screw the 300-ohm pigtail to the connectors on the other one, and Voila!, you have a inline isolation transformer.


There will be more insertion loss on this dude, but since you're on cable,

you'll never notice it. This thread is worrying about insertion loss on

very weak signals...


All a Balun is a small transformer. By connecting two together like

this you completely isolate the incoming ground from your setup, which

is what is probably causing your hum.


Craig
 

·
Registered
Joined
·
4,129 Posts
...or you could do what the cable dorks undoubtedly neglected to do: bond the ground at the cable entrance with the home's common ground at the electrical entrance... this assumes that they even had the courtesy to ground the cable. If not, you need to. This is also safer than - effectively - lifting the ground wire as would be the case with 75/300/75 baluns.


The discussion of high/low insertion loss baluns is very interesting. I had not even considered that as an issue. Since I live on the fringe, this could be very valuable.


Thanks guys!
 

·
Registered
Joined
·
2,814 Posts
Man E:


Thanks for the grounding tip but they already checked all that out. They actually found some evidence of hum coming from the boulevard main cable source.


Liked your Silent Angels and The Manny Page links.:)
 

·
Registered
Joined
·
3,993 Posts
Quote:
Originally posted by cneth


All a Balun is a small transformer. By connecting two together like

this you completely isolate the incoming ground from your setup, which

is what is probably causing your hum.
That's what I had always assumed, but this link from Ri-Chee: http://www.rason.org/Projects/balun/balun.htm , depicts an alternative balun construction that uses an autoformer to crudely match the input and output impedances, and it does not completely isolate the grounds. You can determine whether a matching balun is an isolated transformer or an autoformer using an ohm meter.


Holland Electronics makes a line isolator that uses capacitors, but it sells for about $40. If I recall correctly, its part number is CCI-1.
 

·
Registered
Joined
·
1,525 Posts
I have never been able to kill the ground loop from my cable. The grounding is correct, I tried the two balun trick, but nothing works except disconnecting the cable. I also tried inserting only the center coax wire and not the ground shield, but that made the picture look terrible. I don't have audio hum, I get scrolling faint gray bars on a black screen. If you do a search on this, there is a thread that discusses a black box that cost $100, I think the name of the box is "magic box", but that name alone made be run the other way.
 

·
Registered
Joined
·
4,129 Posts
Thanks Jim. The "hum" is not coming from the boulevard. Me thinks they were tired of chasing the problem and gave you lip service. A ground loop hum is a difference in potential. You have already verified that there is a difference in potential between your cable and your mains. It's just a question of where.


Rule #1: Minimize. Create the smallest testable configuration possible.


I had a situation where splitting A/V equipment across circuits (breakers) caused a slight hum. I reconciled to a single (30A!) breaker. With everything in my outfit - satellite dishes, OTA antennas, cable, PCs, audio/video gear, whole house distribution - I have zero hum. If your gear is not spread across multiple breakers, then I am suspicious of the outside configuration. If it is split across breakers, try to get a minimal setup on just one breaker that first excludes and then includes the cable.


Having said that, if they had properly bonded from their static discharge block's grounding rod to your home's mains grounding rod it is unlikely that you would be having this problem. My cable cabal, my mother's, my brother's, ... didn't do it right either.


Good luck!
 

·
Registered
Joined
·
2,067 Posts
Back to the original subject.... Has anyone done comparisons between the various baluns?


I have used several of the cheap radio shack, RCA, or other $1.99 baluns. I bought one of the Channel Master baluns when I read here that they were better. But, I did not see a noticeable difference between the cheapie and the channel master.


Maybe I'll try to put together a coax balun next. Any tips on how to make one?
 

·
Registered
Joined
·
532 Posts
I have to admit that $30 for a coax balun is a bit steep, as much as I respect A-Tech, given what materials are involved. Then again, given what I charge a client per hour, it just might be worth it to buy a premade 6 incher than make my own.


Besides, I'm a software engineer. They don't allow me to touch a screwdriver, much less a soldering iron. :)
 

·
Registered
Joined
·
1,273 Posts
Guys,


Please note that there are significant differences in the operating principles between the type of TV baluns one buys commercially, and the "coax baluns".


The coax baluns are tuned for one frequency. They'll operate on one channel and a few nearby, but they'e not intended to operate over the whole band.


OTOH, the commercial baluns are of the broadband, or "transmission line" transformer type. Theoretically, they work OK at any frequency above a certain minimum, which is determined by the inductance in the device.


RS could be using a core material that's lossy at UHF; hence the poor numbers.


From what I've measured, the CM and RCA have less than 1 dB of loss.


Note also that some of these baluns may be of the autotransformer-type, and thus not provide any DC isolation between the two ports.
 

·
Registered
Joined
·
42 Posts
Discussion Starter · #16 ·
Thanks for your response, BarryO. Someone actually making measurements are what I was hoping to hear.


What frequency bands did you measure the the CM and RCA baluns over (both UHF and VHF?), and how did you do those measurements?


What started this discussion was the fact the RS gave such bad specs on their balun, and KYES (linked to in my original post) actually measured a similar type, and came up wth losses as much 80%. Yes, the coax baluns are designed to be tuned at one frequency (and when well made, the VSWR should be less than 1.1 at that frequency). But by their nature, a match using a passive half wave length of transmission line with the same impedance as the lower impedance side, the worst case loss of the balun should be a mismatch loss between 75 Ohms directly connected to 300 Ohms, which would be only about 36% loss (slightly less than 2 dB loss). Though this loss is not low, it is still much better than losing 80% (greater than 6 dB) from the RS balun or the ones measure by KYES. If the RCA and CM commercial baluns actually have less than 1 dB of loss, then these are definitely the ones to go with.


Also, co-ax baluns are transmission line baluns (using a half wavelength of transmission line to peform the match as well as the current balancing). I'm pretty sure most of the small baluns are transformer type baluns (i.e a simple transformer is used for the match and balance). What exactly is a "transmission line" transformer type?
 

·
Registered
Joined
·
42 Posts
Discussion Starter · #17 ·
I haven't studied the circuits of the transfomer baluns carefully. but just thinking off the top of my head, using inductance to match impedance would also not intrinsically be very broadband since the impedance due to inductance varies with frequency (Z =jwL whre w = radian frequency).


Regarding RS baluns being so lossy possibly due to having high core losses: I'll have to go back and review my magnetics. But from what I remember, core losses are mainly due to hysteresis losses and eddy current losses. I can't imagine that standard materials used for cores would be that lossy at UHF from hysteresis and the fields are so weak in this application for eddy currents to be any sort of factor. Is there some other mechanism I'm not remembering?
 

·
Registered
Joined
·
192 Posts
Quote:
Originally posted by Ri-Chee


Also, co-ax baluns are transmission line baluns (using a half wavelength of transmission line to peform the match as well as the current balancing). I'm pretty sure most of the small baluns are transformer type baluns (i.e a simple transformer is used for the match and balance). What exactly is a "transmission line" transformer type?
The coax type baluns dont actually perform the impedence matching use the "transmission line" transformer principle. They actually do something that is much simpler to understand. Imagine that your 300 ohm load(source) is replaced with two 150 ohm resistors in series, with a center tapped point to ground. Then the question stands as to how to feed these two 150 ohm ports from a 75 ohm port. It would be quite simple if one could simply connect the two leads together and drive them in parallel, but you cant because they expect(or deliver) equal but opposite voltages. However it does still bear scrutiny that the two 150 ohm ports in series are 300 ohms, and in parallel would be 75 ohms.


So we employ a trick: Take the 75 ohm port and divide it with a T-connector (or other device) into two lines, also 75 ohms each - but with one of the lines a half wavelength longer than the other, and with both of them a muliple of a half wavelength overall. (zero length also being an allowed multiple for one of the lines). The extra half-wavelength in one of the lines will provide the required 180 degree phase shift so that the 75 ohm coaxial port will see two 150 ohm ports in parallel (and unbalanced), while the 300 ohm side will present as two 150 ohm ports in series (also balanced).


This impedence transformation trick is not frequency sensitive, but it does become frequency sensitive when you use transmission lines, i.e., you could just as well use a 1:1 transformer to drive the phase reversed leg and presto, you have one of the DC passing transformer type baluns others are speaking of.


You could just as well turn your 300 ohm antenna into a 150 ohm antenna by putting connecting one of the leads to a 150 ohm resistor to ground. You'll lose 3 db that way, but the remaining lead will be a 150 ohm unbalanced source. You'd have a 2:1 VSWR here, which as I recall results in only about 11% reflected power, so the total loss this way would be something like 3.5 db. Still better than the RadioShack specs.


If you want to insist on keeping everything balanced, then you would use a resistive pad to match the 150 ohms to 75 ohms. You dont just put a 75 ohm resistor in series with the coax, because while that makes the coax look like 75+75=150 to the antenna, it makes the antenna look like 75+150=225 to the coax, which can do screwy things if you have reflections elsewhere and stuff like that. There is some formula for so called "minimum-loss resistive matching" somewhere. Anyway, source matching with a single resistor from 150 to 75 would give an additional loss of 3 db, which including the resistor in series with the other antenna lead would give you a total of about 6db of loss. Just slightly better than the RadioShack spec. huh?
 

·
Registered
Joined
·
42 Posts
Discussion Starter · #19 ·
Quote:
Originally posted by glgorman



So we employ a trick: Take the 75 ohm port and divide it with a T-connector (or other device) into two lines, also 75 ohms each - but with one of the lines a half wavelength longer than the other, and with both of them a muliple of a half wavelength overall. (zero length also being an allowed multiple for one of the lines). The extra half-wavelength in one of the lines will provide the required 180 degree phase shift so that the 75 ohm coaxial port will see two 150 ohm ports in parallel (and unbalanced), while the 300 ohm side will present as two 150 ohm ports in series (also balanced).
Hey glgorman,


Thanks for your great explanation. It's all clear to me now. I'm annoyed at myself for not being able to figure this out. (As the saying goes... everything is obvious once you know the answer :) ) For the balun operation, it makes things easy to understand by thinking of it, as you described, from the 75 Ohm side, and saying you want 2 equal and opposite signals, which in the case of the co-ax balun, you simple split the center conductor and add a half wave section. And with regards to the impedance matching, the key that you pointed out, is to imagine the balanced 300 Ohm line as 2 150 Ohms in series with a center tap ground (which is the way to model a balanced line load.... it's been too long since I taught any of this stuff). Half a wavelength is one complete rotation on the Smith Chart, so the impedance you see at the end of a half wave line is identical to its other side, giving you 2 150 Ohm impedances in parallel at the 75 Ohm side. Thanks.


Now that I understand this, it's pretty straightforward to compute the VSWR loss as a function of frequency due the added length no longer being exactly half-wave. (I'll do this as soon as possible... though with a 18 month old daughter, it may not be for a day or two. Anyone else, though, is invited to beat me to it.)


As you go off center frequency, the two arms are no longer exactly 180 degrees out of phase. There are losses associated with that. I can't think of how to treat that off the top of my head. Any insights into that, glgorman?
 

·
Registered
Joined
·
1,248 Posts
The losses can get as high as 100% This will occure when the phase hits any multiple of 360 degrees. At that point you are driving the same signal into both sides of a differential line. you have 100% common mode signal with 0% differential.
 
1 - 20 of 29 Posts
Status
Not open for further replies.
Top