Help making DIY balun from DBT LogPeriodic Antenna to 490-700mHz preamp

[Paul Price]

How do I make a balun?

I live in Copenhagen Denmark and want to receive DBT TV from Malmo Sweden and I have a .3 meter long indoor tabletop  log-periodic antenna(mounted outside on the back porch), but all Swedish TV signals are very weak, and even with an adj. gain antenna preamp (up to 30db adj. but has no LTE filter), I can only see a few hours of one Malmo TV station per week.

I have found that no satellite TV or  local TV cable-tv services do not offer but only 3 of the 8 possible Swedish freeview TV stations, so that is not a good option.

The local Copenhagen transponders give a strong signal and I have no trouble receiving them, but their broadcast is of little interest to me!  A coat-hanger attached to coke cans works well for local signals!

Due to restrictions on what I can attach to my back porch of my apt. building, I cannot mount anything but a satellite dish or a small non-obtrusive antenna on my tiny back porch.

It may be that I am only receiving one strong Swedish TV station, but only occasionally, that is about 30-miles away because of atmospheric skip and the situation seems hopeless, but I have not yet explored all my options

The antenna has no balun and when I tried inserting a standard TV 300-ohm line to 75 ohm balun to connect to the preamp the signal was gone completely.

I am connecting the front of the antenna directly to 75-ohm coax and making connections by tightly winding many turns of silver plated 32AVG stripped wire-wrap wire around the front most element pair, near the center boom (which is plastic).

Surprisingly, when I bought a mushroom Yagi antenna of 4-ft length, there was no reception of any but the local Dan channels, so this desktop antenna is the best solution so far for what  I have to work with.

Reception was much better before LTE/4G service started many years ago. I have found there is now a 20-DB DVB antenna preamp available with LTE/4G filter, but I haven't tried this option yet.

If I was to purchase a LTE/4G filter I would still need to build a balun to connect coax directly to the antenna, would inserting a filter before the preamp help with such a weak signal?

I have some discarded TV/Internet routers that multiplex TV and broadband internet on the same 75-ohm coax cable, and I see within the RF shielded box connecting directly to the 75-ohm F-connector a 3-4mm ext. diam toroidial core used to match the cable to the broadband internet processing chip section of the router.

Please, can some rf guru's help me, give me some help in using this small ferrite core to wind a balun to interface the antenna to the 75-ohm unbalanced antenna preamp?


BTW, I am very experienced in electronics design and have the ability make my own PCB for circuits, but know only so little about RF.

[BigBoss]

Don't bother yourself by designing/realizing such balun.
Use this ( or similar) balun from Mini-Circuits and just after place an Low Noise Amplifier with a single transistor for that band then place them all
on a single PCB by taking care of RF principles.
https://www.minicircuits.com/WebStore/dashboard.html?model=TC4-122-75X%2B

[wasedadoc]

For the same length a Yagi for the correct channel group has significantly more gain than a log periodic.  Get as much antenna gain as you can before adding an amplifier.

[Paul Price]

As I've mentioned, I have already tried using a standard legacy 300-ohm line to 75-ohm coax balun and the result was a complete loss of any Swedish channels.  This failure to work could be due to a very poor quality balun or (more likely IMHO) a mismatch?
The Mini-Circuits is a 4:1 balun which expects to be connected to a feed point on a 300-ohm dipole.

The Mini-Circuits balun you recommend has very good specs(read the spec sheet), but I expect I would still see poor results due to the way I position a feed  on the very front-most dipole elements of the log-periodic antenna. I expect this is creating an impedance mismatch.

Consider this:
The elements on my log periodic antenna are connecting alternating horizontal sides along the plastic boom over the length of the antenna.
The top and bottom elements are crisscrossed and are separated vertically by the width of 1-in plastic boom. The separate planes of the elements cause elements to overlap each other after each element crisscrosses in horiz. position.
 
Where on front element would give the optimal impedance match to the balun? My feed point now is at the point where the element is attached to the plastic boom and then starts to  cross over to the next element behind.

[Paul Price]

wasedadoc,

As I've mentioned in my opening post, a yagi antenna could would not be permitted to be mounted on my back porch because it would be so long that it would be an obvious eyesore.

When I bought a mushroom triple boom Yagi antenna of 5-ft length, there was no reception of any Swedish channels, no channels but the local Dan channels were received, so this desktop antenna is the best solution so far for what  I have to work with.

[mag_therm]

A job for  XNEC2 .
There is a 35 ~ 55MHz LogPer in the \examples that you could scale, saving  some work.

[Paul Price]

mag-therm, Many thanks for the tip.

Unfortunately I  don't have a Linux OS computer to work with to run XNEC2.

The real problem seems to be how exactly to make a Log Per Ant that can reject LTE signals along the elements and where to connect the feed point.

Do you think the XNEC2 could deal with a LogPerAnt with LTE/4G band-reject elements and it would be worthwhile to build a Linux PC?

[Paul Price]

BigBoss:

I am already trying to get signals from what might be the mud area of signal strength as seen on a spectrum analyzer. The Mini-Circuits balun has a 2-db insertion loss and 12-14db input return loss.

Return loss is a measurement parameter that expresses how well a device or line matches. A high return loss is advantageous as it will result in a lower insertion loss.

However, the 2-db insertion loss is still much better than a direct mismatch of a 75-ohm coax to the elements.

But still unanswered:
Where is the best feed impedance match point along the front elements length as a distance from the plastic boom?

Is it practical, considering the very low signal level, to insert an in-line LTE/4G filter prior to the input of my existing ant preamp?

Considering the strong local transponder channel amplitudes, how can I keep these signals from overloading an in-line amplifier and so preventing the amplified weak signals from being handled well by the RF input on my DVB/T receiver?

[mag_therm]

I think there are versions of NEC2 to run on win, mac etc.
Perhaps someone on here can answer that query.

I was just thinking, that the 35 ~ 55MHz example could be run as is, to give you some insight.

Scaling all of the individual elements would take me maybe 2 days  to get working. (Based on my usage so far, and I am a bit slow)
That is because there is no graphic UI, each element is decribed numerically on a "card"

[wasedadoc]

Are the Danish and Swedish transmissions coming from the same, or nearly same, direction?

I don't accept the "Yagi is too long" response.  If you have enough length for a log periodic, a Yagi of the same length is possible and will have higher gain.  If the 5 foot Yagi performed worse on the Swedish channels than the 0.3 metre log periodic then the Yagi was faulty or wrong channel group or pointed in the wrong direction.

[Paul Price]

wasedadoc

My LogPerAnt is a tabletop indoors UHF only model and it is only 20-in from tip to tail and sweptwing airplane shaped.

My porch faces south towards local transponders but where I point to, Malmo transponders, is almost SSE.
When pointed optimally, the angular window of reception is very critical and reception is possible only within 15 deg window of rotation.

The mushroom Yagi was a No LTE/4G filtered three-boom design with the rearmost center element a folded bipole
Both ants were fitted with my 5-30dB preamp sans LTE/4G filter, as close as possible to the ant, to make tests. If I advanced the gain  setting too close to max all reception halted, probably due to LTE/4G overload.

[mag_therm]

Hi Paul,
If your LogPer is store bought, can you just connect your amplifier cable cable to whatever jack was provided?
That should solve the balun problem. Twisted wire connection  may be no good at this frequency.  And there is question whether balun is of use on receive antenna.

If your need to watch tv is greater than need to experiment, then can you look at antenna/amplifier kits used on  farms and distant village etc?

If you are experimenting for higher band interference, there is a possibility for unwanted response at 3rd & 5th on LogPer elements.
In that case 1/4 wave open lines can be connected as stubs to form sharp filters. At that frequency pcb lines could be used.
Years ago in Autralia remote area I made a coax stub filter for VHF TV QRM and it worked well when trimmed to frequency.
There is info on some ham radio sites about stub filters, also about calc and build LogPer.

Now I am only HF here; I found XNEC2 (free) and a nanoVNA (not expensive) both very useful.
NanoVNA usability is greatly improved by a desktop UI, I use nanovnasaver (FOSS) on linux.

[ftg]

Do you know if the Malmö DTV mux is on the 470 - 700MHz UHF band or the 170 - 230MHz VHF band?
Similarly, do you know the frequency your local transmitters use and where they are located.

You mention turning up the gain to 30dB, that might not help, as now the frontend in the TV will get saturated by super strong local signals.
Similarly, if the amplified antenna is specified for 47 - 860MHz, it will also get interfered by 700MHz and 800MHz cellular signals from cellphone base stations.
Antennas with built in LTE filters usually either advertize that they have the LTE filter, or print on the box that the frequency range ends at 600MHz or 700MHz.
Older 800MHz filters might not help enough if the new 700MHz band is in use around your parts.

If you must use an indoor antenna, then it could also be an inconveniently large yagi.

If the outdoor antenna cannot be long, but can be flat, then bowtie or dipole mat antennas also become an option for directional gain outdoors.
Some examples of those:
https://www.amazon.com/High-Gain-Bow-Tie-High-gain-receiving-low-signal/dp/B00MU2JL0Q
https://www.techhive.com/article/583569/antennas-direct-db8e-review.html
https://www.instructables.com/Eight-Bay-Bowtie-Antenna-Array/

There's some gain comparisons between different TV antenna types here:
http://www.hdtvprimer.com/ANTENNAS/types.html

[Paul Price]

I know for sure the local and Malmo Sweden transponders are operating between 500 and 688 mHz. My DBT/T tuner has a SYSTEM diagnostic display of channel freq. signal strength and signal quality(both have a range of 0-100%). I notice that when I am receiving Swedish HD Chan 4 that the signal strength bar indicator slams at 100% while the Signal Quality stumbles around 0 to 11% with the majority of Sig Qual around 1-2%. This tells me LTE/4G transmissions are jamming the ant preamp and the front end of my DVB/T tuner.

The LTE/4G signals are so strong in my apt. that I cannot use the 10mv/div sensitivity of my 200MHz Rigol scope, even with 20-Mhz BWselected! To work with low-freq signals I would have to build myself a Faraday Shielded room.

I ordered a new preamp today with stop band LTE/4G filters that has adj. 5-20dB passband gain only over the 470-700 mHz DVB/T which is the freq. range of the local transponders in Copenhagen and the Sweden transponders.

What is frustrating about this search for Swedish channels it that my DBT/T tuner must be reset by a full freq. scan in tuner Setup to allow any display of info about reception.  This means, for instance, that if I run a scan when the Swedish Chan 4 is very weak, the channel will not have been granted a valid channel selection entry of the tuner's list of channels. In other words, the signal strength has to be strong enough for the tuner to be able to select the channel on the remote control, otherwise the channel number is not valid and I cannot see any signal strength because I cannot set the tuner to the channel of interest.

During a brief 6-hr window of receiving opportunity a week ago, I was able to see 6 Swedish channels.
I find this DVB/T DX-ing quite challenging and fun!l 

Denmark State run TV channels were a decade ago very entertaining to watch, but now, with no paid license for viewing, there are but two Danish channels of any interest and they broadcast very little programming I would bother to watch. Not true for Swedish freeview, where there is a Cultural Channel, a children's channel (that turns into adult programming late night) etc.

What is still unanswered is what point on the frontmost elements to connect a balun to to get a 300-ohm match.

[pienari]

For your question.
Folded dipole needs 300ohm balun.
Buy or made it for coax cable.

Log per antenna not need balun.
Feed impedance depends distance of the frame tubes.

Can you tell what is the distance to Malmö transmitter?.

I personally would build high gain yagi then low gain log per anenna.

[Paul Price]

pienari,
I don't understand why LogPerAnt doesn't need a balun?

Distance to Malmo 30miles, after first 5 city, 20-mile seawater than <5 mile to Malmo transponder.

Installed new LTE/4G filtered 0-20dB ant preamp today, now able at full gain 20dB adj. to get occasional snippets of sound, and on many rare occasions over an hours time, broken pieces of a picture.


What is still unanswered is what point on the frontmost elements to connect a balun to to get a 300-ohm match.

How to make a DIY balun with 5-mm ext width toroid ferrite core? Is my antenna connection really 300-ohm?

I don't understand how a 4:1 balum transformer would not attenuate a weak signal 4:1 even though the balanced input/output would create optimal power transfer?

However, connecting one front element directly where it meets the plastic boom of a multi-element LogPerAnt to the coax outer shield seems utterly sloppy except if the first element pair was an isolated dipole.

Now, my only option to get reception is to improve the quality of the LogPerAnt.

[wasedadoc]

What is your reasoning for wanting this 300:75 balun?  I am under the impression that 75 Ohm antenna, coax cable and receiver input are widespread in Europe.  Do you have specific evidence that either the log periodic antenna or the receiver input is not 75 Ohm?

Have you ascertained that both the local and distant transmitters are using the same polarisation?

Have you considered using two antennae?  One for the local transmitter and another for the distant one.  Put the amplifier only on the distant one and then combine into a single feed to the receiver.  Prevents overloading and allows optimum pointing direction of each antenna.  Very desirable if polarisations are different.

[Paul Price]

wasedadoc,
I want 4:1 balun because a LogPerAnt is balanced and a direct connection of an unbalanced coax line would unbalance the elements of  the antenna if connected directly to the coax and there would radiation and reflective imbalances in the electromagnetic field introduced. Antennas are all about symmetry, max transfer of energy, and predictability of operation, while unbalanced to balanced connections are all about SWR and signal loss and impedance mismatch chaos.

i have no idea what the impedance of the antenna is because this information was in no way revealed to me.

Antennas and their coupling to tuners are designed to be 75-ohm because it is standard for consumer ant amps, filters, cable, tuner inputs, etc.

In any case the impedance of the antenna ultimately depends on  how the antenna is coupled to the coax. A folded dipole is most used as the feed element and usually positioned near the support pole at the rear of an antenna. Its impedance is well controlled by its precise geometry to be 300-ohms. My LogPerAnt has no folded dipole. Polarization is known to be horiz. and I am making the frontmost element pair as the feed point(as it was originally connected as an indoor Ant.

There's barely enough room for one antenna, two would change my back porch into an antenna farm..a little too nerdy even for me.

[wasedadoc]

https://hamwaves.com/lpda/en/index.html#:~:text=Every%20so%20often%2C%20HF%20LPDAs,75%20%CE%A9%20characteristic%20input%20impedance.

Lots of info there.

But IMHO you are asking for the impossible.  I live in a village 26 miles from the transmitter (not a low power relay) with open country in between.  Myself and all neighbours with terrestrial DTV aerials have external "roof height" Yagis with 48 elements or more.  I installed my own but the vast majority were installed by professional fitters.  You can be sure they would have fitted smaller ones if they would have been satisfactory.

An amplifier close to the antenna is useful when the cable between antenna and receiver has a lot of loss.  But if the cable is OK and the signal at the antenna is weak an amplifier cannot do magic.  Signal/noise ratio is the issue.

[pienari]

A log periodic antenna does not need a balun.
I've never seen them used.
The fitting impedance is determined by the difference between the pipes used for fixing the elements.
Log per is low impedance 50-100ohm.
Perfect matching is impossible to achieve because the impedance curve is wavy and therefore cannot be good but sufficient to function.

https://www.changpuak.ch/electronics/lpda.php

Adding an amplifier to the input point is challenging.

Malmö's dvb-t transmitter must be 100W erp? which is really small, maybe some supplement transmitter?.

Depending on the terrain, I think that the signal can be received 20-30km away with a good yagi antenna with a gain of more than 10-15db and when a mast amplifier is used.
The most important thing is to use a DVB-T measuring device to find the best SNR ratio (not the best signal).
The best SNR probably cannot be found in one place, but the antenna must be moved sideways and vertically.

This is how I have successfully installed a few antennas in Kuusamo Ruka's DVB-T2 in Finland
26km from a 400W erp UHF transmitter in a challenging terrain with big changes in height.

[Paul Price]

pienari,
It is interesting to note that I have looked at pictures of DBT/T antennas from Amazon and a multitude of other suppliers and all of them feed the coax to TV or preamp from a folded dipole at the rear of the antenna, some 1/4 lamda or so from the last pair of element. Never an instance of direct connection to the front elements.

Do consumer antenna mfg's know something about antennas that the ARRL or other mfg's ignores?

The design of many commercial antenna use  fractional 1/8 to 1/4 lamda dipoles as elements, while others use small diam rods into a metal boom..why is one approach possibly better?

My very compact LogPerAnt uses flat stamped aluminum flat triangular elements tapering from about 3/4-in wide with their leading edge perpendicular to a plastic boom, crisscrossing from left to right along the antenna. These stamped elements are connected along the whole length of the antenna zig-zagging from left to right from the frontmost to the rearmost element with the feed point on the front most elements. Top and bottom elements that mirroe/completely overlap each other in the vertical plane, but separated by the 3/4-in plastic boom.

On the other hand the dipoles suggested by the formulas have the small diam alum rod elements screwed into a solid rectangular boom.

What's going on here? 
All the design formulas assume aluminum rod of 3mm or so diam..yet my wide-element abstract antenna design works so well compared to a 3-boom LPA commercial ant I bought or attempts I've made to construct traditional thin diam rod element based LPAnts.

I am wondering if I could somehow construct a test fixture with a fixed 400Hz modulated 600Mhz CW anatrophic transmitting dipole horiz antenna fed by a 600 mHz 1-Watt oscillator/final, Across a foot or so of separation on my workbench build an antenna piece by piece by experimenting with different element configurations and spacing to make my own antenna, but based on the design of the modfied LPA one that is working now, but adding more elements? This would require a gain measuring 600-Mhz receiver (superhet or super-regen?)

Sounds like fun for an antenna-obsessed hobbyist I've become!

[pienari]

Are you trying to make some hybrid antenna?.
I have seen some small plastic frame log perodic antenna indoor use.

Im talking antenna like this.
https://www.iskra.eu/f/pics/5978/UHF_TV_Antenna_P-2845_m.png

If you have space for the antenna why not stack 4.
It can be done easily with combiners.
4 to 1 and then mast amp.

I have used Elcom ASS1200 sweep signal generator / signal analyzer to test few 446MHz yagi beams.

[wasedadoc]

pienari,
It is interesting to note that I have looked at pictures of DBT/T antennas from Amazon and a multitude of other suppliers and all of them feed the coax to TV or preamp from a folded dipole at the rear of the antenna, some 1/4 lamda or so from the last pair of element. Never an instance of direct connection to the front elements.

Those are Yagis. One driven dipole. The other elements are parasitic.

[ftg]

I notice that when I am receiving Swedish HD Chan 4 that the signal strength bar indicator slams at 100% while the Signal Quality stumbles around 0 to 11% with the majority of Sig Qual around 1-2%. This tells me LTE/4G transmissions are jamming the ant preamp and the front end of my DVB/T tuner.

This might also mean that some local DVB-T channel is on the same frequency and much stronger than the Swedish broadcast.

A really annoying problem to solve, especially if the local transmitter is nearby or in the same direction as the Swedish one.


Most coax fed logperiodic antennas are built to have the characteristic impedance of the coax, so no 1:4 balun or anything like it is required.
Some TV yagis do have folded dipole feeds, as those are also DC-short and thus good against static electricity.
The balun in those is usually built into the feed.

[pienari]

Finally i found this.
https://www.teracom.se/privat/tv/?SearchQuery=&PositionQuery=location
I tested today small Iskra P-20WF logper antenna with preamp.

Antenna was 2m hi from the ground.

I chould receive all dvb-t2 channels from Kalix transmitter.
Kalix is about 180km from my qth.