Za	    	Zb		Z coax line
	25.0 ohms	50 ohms 	35.36 ohms
	30.0 ohms	50 ohms 	38.73 ohms
	37.5 ohms	50 ohms 	43.30 ohms
	
	Za	    	Zb		Z coax line
	25.0 ohms	60 ohms 	38.73 ohms
	30.0 ohms	60 ohms 	42.43 ohms
	37.5 ohms	60 ohms 	47.43 ohms
	
	Za	    	Zb		Z coax line
	25.0 ohms	75 ohms 	43.30 ohms
	30.0 ohms	75 ohms 	47.43 ohms
	37.5 ohms	75 ohms 	53.03 ohms
			

Enter any Za and Zb ... Cumulated Za [ohms]                 Zb [ohms]         Z splitter QW line [ohms] ... or chose splitter from menu Type of Splitter                 Opt.- corr.lgth. at 2 ports on λ/4 3 ports on λ/4 4 ports on λ/4 6 ports on λ/4 2 x 1 port on 2 x λ/4 2 x 2 ports on 2 x λ/4 2 x 3 ports on 2 x λ/4 2 x 4 ports on 2 x λ/4     MHz Single Z's at (a)           50 ohms 60 ohms 75 ohms       Zb 50 ohms 60 ohms 75 ohms       Core Ø       Units       mm in nH/Port   D/d ->eqv.lgth 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Square x 1 Square x0.5 Round x 1 Round x0.5 Ignore all Imp. Za Ω     Z splitter QW line(s)         Ω Square Shield inner dim.       Round Shield inner dim.     Option - Approx. compensation for number of ports connector pins inductances: Equiv. QW length wl = For using this calculator Javascript must be activated n ports on λ/4 = short splitter with n ports on single side 2 x n ports on 2 x λ/4 = long or "λ/2" splitter with n ports on either side • Find core's length in "Equiv. QW length" textbox. Er = 1 (air) is estimated for all. For λ/2 double core length. Select "D/d ->eqv.lgt" => "Ignore all" for result exclusive approximated port impedances. • An average dim. N-connectors pin may be around 1.0 to 2.0 nH. • use Select "D/d ->eqv.lgt" => ... x 0.5 to split imperfection between length with right Z and resonance of whole splitter to yield best splitter tube length ... unless you ar planning something special Image: © Dr. Tanja Kurzenknabe

See plot of a build with inductance at ports correction applied, here

Back up page             Back to Splitters Menu

               

	Single Z's	Calc.	     Za 
	2 x 50 ohms	50/2	 25.0 ohms
	4 x 50 ohms	50/4	 12.5 ohms
	3 x 75 ohms	75/3	 25.0 ohms
	...
		

From this we can derive that a 2 x 50 ohms to 50 ohms splitter may be used as 3 x 75 ohms to 50 ohms splitter too.

---

2, 4, 6 or 8 way splitters with 2 x λ/4 lines

The image is showing a 4 way splitter. Just imagine 1, 2, 3 or 4 connectors mounted either side.
These commonly are called λ/2 splitters. In my oppinon this name giving falsely suggests that we
would have a different transformation mechanism here.


A power splitter with 2 x λ/4 lines is a precise mirror image either side of what the half
numbers of antenna ports λ/4 splitter is. Just that the Z at feed point (x3) must be twice the number
since boths sides meeting results in half impedance.

Use same calculator as above. Match to Zb = 100 ohms at port x3 instead when feeding with 50 ohms coax
... any feed line Z must be multiplied by two.

---

Any splitters made from λ/4 lines can be used on 3rd harmonic bands

Every splitter using λ/4 or 2 x λ/4 lines may be used on tripple frequency too.
The transformating lines length is 3 x λ/4 then, which is equivalent to a single λ/4 line followed by
a λ/2 line, which does not change any properties.

A 144 MHz splitter will work fine on 432 MHz. Any 432 MHz splitter will give fine results on 1296 MHz. I have used
a quality 432 MHz λ/4 4 way splitter made by UKW-Berichte with success for feeding four Tonna 1296 MHz Yagis
in many contests.


Back up page             Back to Splitters Menu

               

---

Testing and plotting Splitters

---

•   A 70 cm band Splitter with Tuning Screws by Peter, DJ4TC

Dimensions: 25 x 25 x 1.8 mm aluminuim square tube, 10mm copper tube for core.
With an inner core diameter of 10.0 mm the inner of the outer square tube shall be 21.3 mm.
The used 25 x 25 mm with a wall thickness of 1.8 mm gives 21.4 mm inside. Fair enough ...
The inner round tubes length is 2 x 173 mm.
The outer square tubes length is 392 mm, which Peter reports to be not critical

The Splitter engages 4 galvanized trimming screws M5 at 70 mm from the center each.

Result: 173 mm per side is a little short,
giving -26 dB at 432 MHz and -28 dB on 443 MHz for best S11 exclusive trim screws.
But when trimmend the splitter easily reaches a Return Loss (S11 in dB) better- 40 dB.

---

•   A 2m band Splitter with PTFE Tuning Screw by Wolf, DF7KB

The PTFE screw pushes the core tube slighty out of center, which
enables adaption of the impedance of the transformating path.

Side view: Here we see the PTFE cushion bearing rest for the core tube

Furthermore the second constructive detail:
A brass tube inserted into the copper core tube to adjust the frequency response

.. this resulting in a notable -60 dB at 145.5 MHz

And the measurement equipment at DF7KB

A miniVNA PRO operated with Java SW vnaJ.3.1.9, calibrated with WiMo's set SMA 50 Ohm, SMA short circuit and
open ports. The coax for calibrating the 4 pole measurements is Suhner Sucoflex with SMA connectors.
"I have connected the 2-way power splitter with 2 x 10 dB precision attenuators followed by 2 x SMA 50 ohms terminations.

---

•   A fine tuned 144 MHz 3-way λ/4 splitter made and tested by Jörg, DL2VL

Yet another remarkable produce by DL2VL. A way to yield same low Return Loss numbers as with the 'Tunable Splitters'
I show below and are subject of my article '2 m & 70 cm High Performance Power Splitters' in Dubus 2-2016.
Only that Jörg with his sense for fine mechanical solutions came up with a method of tuning length of the
splitters transformation line.
All photos,plot and build: on courtesy of DL2VL. Click the photo to enlarge.

The resulting Return Loss

And how he does it

First of all we notice the PFTE slices he uses to secure the conductor in the center of the outer square tube.
They do not only act as mechanical elements, they by their number also tune on the electrical length of the inner tube.
With 4 slices at the sum point and 1 at the split side he found the optimum for this pairing of tubes.

On top of that he machined an N-flange bushing that has a spring pin (!).
How to do that?
The pin is drilled hollow, then he places a tiny piece of rubber inside the opening and a sliding new pin follows.
Now what is that good for? ... with this bushing he can shift the position ie. alter the length of the transformation
line by very small amounts to do the final fine tuning. Find details on the photos. Click to enlarge and view them full screen.

Hier die Daten des 3-Way 2m Splitters:

Außenleiter    25 x 25 x 2 mm        561,5 mm lang
Innenleiter    14,0  x 0,5 mm        511,5 mm lang
1-Port-Seite       4 x 2,5 mm = 10 mm PTFE
3-Port-Seite       1 x 2,5 mm PTFE
Impendanz 1-Port-Seite    (49,8 + j0,1) Ohm @ 144,2 MHz
Teilungsverlust            4,86 dB - 4,78 dB = 0,08 dB (S21= -0,08 dB)
Phasendifferenz 3-Port-Seite    <0,8°
(Messung mit Anritsu VNA)

And in English

Here the Data of the 3-Way 2m Splitter:

Outer Conductor    5 x 25 x 2 mm      length  561,5 mm
Core tube         14,0  x 0,5 mm      length  511,5 mm
1-Port-Side       4 x 2,5 mm = 10 mm PTFE
3-Port-Side       1 x 2,5 mm PTFE
Impendance 1-Port-Side    (49,8 ohms + j0,1) ohms @ 144,2 MHz
Splitting Loss             4,86 dB - 4,78 dB = 0,08 dB (S21= -0,08 dB)
Phase Difference 3-Port-Side    <0,8°
(Measured with an Anritsu VNA)

73!
Jörg, DL2VL

---

•   An adjustable 432 MHz 4-way λ/2 splitter made and tested by John, ZS6JON

While the ideal inner diameter of the square outer tube would be 27.08 mm for the given cores diameter here a tube
with 28.6 mm is in use. Further with the tuning screws the frequency can be adjusted in a way that a slightly longer
tube can be used when the impedance of the splitter line is slightly higher than it should be.

For a 12.70 mm core the ideal outer square tubes inner diameter is 27.08 mm. With the available materials of 12.7 mm
core and 28.6 mm square tubes diameter the calculated Z is 53.2 ohms. Instead of the 50.0 ohms, that each
side of a 1/2 λ 4 way splitter should have. With those 53.2 ohms the splitter coax on its own would not
allow a Return Loss better than -32 dB.

The length of the core tube without tuning screws and right d/D ratio would be ~ 347 mm when ignoring all
corrections for bushing pin inductivities or d/D ratio correction. This tunable splitters core measures 362 mm,
that is an excess length o f at least 7.5 mm per 1/4 λ transformation way each side. It is that bit
longer because the capacitances introduced shorten it electrically.

The result is a 70 cm splitter with an unseen before fabulous input reflection of well beyond -50 dB on spot
on 432.1 MHz for EME operation without having to see a mechanical workshop to do diameters and lenghts of parts
to the 10th of a millimeter.

S11 (input reflection in dB) plotted with an Anritsu MS2000 series VNA
and calibrated measurement lines: Return Loss (432.1 MHz) = 54.9 dB

Same as Smith Chart: Z(432.1 MHz) = 50.3 + 0.3 ohms

And a drawing of that splitter, bushings are 'screw-in' type, tuning screws are 5M

---

  Attenzione!

When building tunable splitters please make sure that
(1) the minimum distance between screws and coax core is sufficient for the used output power
(2) seal the screw from environmental influences like rain and air moisture

Aluminuim and stainless steel in some pairings are not very 'friendly' to one another as far as electrochemical
corrosion is concerned. Putting on clear coat paint on the outside seems a good idea.

---

•   An adjustable 144 MHz 4-way λ/2 splitter made and tested by Stef, F4EZJ

Stef built this splitter with a 7/16 DIN bushing at the feed point and 4 N bushings at the antenna ends.

The geometry is as follows:

Outer square tube 30 x 30 mm on the outside, 26 x 26 mm on the inside, length is 1047 mm
Inner core is a copper tube Ø 12 x 1 mm, length is 1019 mm
Tuning screws are M 8 and located at 140mm either side of the 7/16 bushing. Later he added more screws.
Which now help to trim down to close to -60 dB and in any case better -35 dB with whatever ~ 50 ohm loads / antennas connected.

External link to Stef's 'Onedrive' online album with many detailed photos

Stef uses the splitter for terrestrial work as contests and aimed for 144.30 MHz and here we are ...

Note:

Future builders might use M8 screws or maybe 2 screws in a row as for this build the space left between
screws and core tube is not much. More screws help to wider tuning abilities. See the photo above.

---

•   A 144 MHz 4-way λ/2 splitter made and tested by Peter, DJ4TC

Calculated with the above online calculator adjusted 'nH/Port' = 1nH and 'D/d ->eqv.lgth' = 0.5 computing a length
of 509.5 mm. Making the core a 1019 mm (2 x 509.5mm) and the outer square profile a 1060 mm in length he plots a

Return Loss = -47 dB and Z = 50.2 ohms at 144.1 MHz.

For this build he used a 25 x 25 mm outer dim., 1.5 mm wall thickness i.e. 22 x 22 mm inner size square tube
and a Ø 10 mm rod as core and quality N-bushings throughout

---

•   A 432 MHz 4-way λ/4 splitter made and tested by John, ZS6JON

Note the 4 similar termination loads. It is essential to use good loads at each port to get true results ...

Freq. 432.14 MHz, RL = -44 dB, Z = 50.1 - 0.6 ohms, what more can we expect? Very good for EME

---

•   A 144 MHz 4-way λ/4 splitter made and tested by John, ZS6JON

Note the 4 similar termination loads. It is essential to use good loads at each port to get true results ...

Freq. 144.161 MHz, RL = -38.1 ohms, Z = 51.0 - 0.8 ohms, very good for EME

For the above λ 1/4 splitters John used square aluminium tubes 25 x 25 mm with 1.6 mm wall thickness. That leaves
an inner height of 25 - (2 x 1.6) = 21.8 mm. Which suits 15 mm CU tubes as core. This diameters result in 27 ohms for the
transformation line. As the 25 x 25 mm tubes wall thickness is not very consistent it worked out better then that as we
can see from the plots he took. He uses my online calculator for determinating length and Z.

---

•   A 144 MHz 4-way λ/2 splitter made and tested by John, ZS6JON

Freq. 144.161 MHz, RL = -37.5 ohms Z = 51.1 - 0.7 ohms, very good for EME

For this λ 1/2 splitter John used square aluminium tubes 25 x 25 mm with 1.6 mm wall thickness. That leaves in inner
height of 25 - (2 x 1.6) = 21.8 mm. Which suits 10 mm CU tubes as core. As he could not supply himself with that he found
9.7 mm tubes from the air-conditioning and refrigeration industry. This diameters result in 53 ohms for the transformation
line. As the 25 x 25 mm tubes wall thickness is not very consistent it worked out better then that as we can see from the
plots he took. He uses my online calculator for determinating length and Z.


Back up page             Back to Splitters Menu

               

---

73, Hartmut, DG7YBN