ON0TB VHF repeater 145,6375 MHz technical data:

·         Transmitter AEG Telecar 160 series.  BGY43 Final.  11 Watts before cavities

·         Receiver Kenwood TK780 digital CTCSS 131,8 Hz decoding squelch @ -123 dBm

·         OT0D/ON7NH PIC minicontroller sponsored by Elektro Dermo

·         4 resonator Kathrein state-of-the-art Invar Bp/Br duplexer < 1,2 dB loss

·         ¼” Cellflex cable

·         2 x 3 el Vargarda antenna 12 dBi pointed towards 300 degrees (Brussels)

·         131,8 CTCSS decoding/re-encoding

·         Spare repeater Kenwood TKR740 series 8 Watts

 

ON0TB UHF repeater 438.275 MHz D-Star technical data:

 

·         RCVR= Motorolla GSM360

·         TX : YAESU 7800

·         DVRPTR + RPImet G4KLX Soft

·         1/2 “ Hiflex cable Eupen

·         Diamond dual band antenna

 

ON0TB UHF repeater 439,0125 MHz technical data:

·         Ericsson RS9044 series repeater 60 Watts before cavities

·         Receive sensitivity -124 dBm squelch level

·         PA4DEN repeater controller embedded software

·         Celwave 6 cavity Bp/Br duplexer < 1,5 dB loss

·         1/2 “ Hiflex cable Eupen

·         Diamond dual band antenna

·         Identical spare repeater

ON0TB SHF repeater 1298,675 MHz technical data:

·         Kenwood TM531 transmitter 8 Watts before cavities

·         Kenwood TM531 receiver -122 dBm squelch

·         OT0D/ON7NH PIC minicontroller sponsored by Elektro Dermo

·         Homebrew 8 resonator twin bandpass cavity duplexer < 1,5 dB loss

·         Procom 9 dBi collinear CXL 23-7C/h sponsored by TRYCOM-Oostende

·         1/2 “ Hiflex cable Eupen

·         No spare repeater

ON0TB 23 cm beacon 1296,950 MHz: not yet operational, sponsor wanted for DB6NT beacon.

 

ON0TB 13 cm beacon 2320,900 MHz:

·         DB6NT beacon transmitter 700 mW outdoor mounted sponsored by ON4SH

·         OT0D/ON7NH PIC minicontroller sponsored by Elektro Dermo

·         Alford slot 6 dBi antenna, no cable.

ON0TB 6 cm beacon 5760,900 MHz:

·         DB6NT beacon transmitter 200 mW outdoor mounted sponsored by ON4SH

·         OT0D/ON7NH PIC minicontroller sponsored by Elektro Dermo

·         13 dBi doppelquad antenna @ 300 degrees, no cable

 

 

All the information on this site is copyright . Authorization from the author is required to publish.

 

Repeatertechniek: PIM oorzaak van gekraak in de ontvanger

Pedro M.J. WYNS ON7WP – AA9HX

 

Passieve intermodulatie in Connectoren:

Het gekraak dat regelmatig te horen is op vele repeaters is afkomstig van passieve intermodulatie, veroorzaakt in het verbindingspad van de zender naar de antenne, oftewel in de onmiddellijke nabijheid van de antenne.

Zeer veel belang dient gehecht te worden aan de gebruikte connectortypes:

N connectoren in vernikkeld materiaal met vergulde middenpin zijn waardeloos, alleen de verzilverde types zijn goed.

Een andere vuistregel is “hoe groter de connector (en dus het contactoppervlak) hoe minder de PIM.

Ook kabels zijn belangrijk.  Gebruik geen koperkabels met alufolie zoals de LMR-400.  Best is puur verzilverd koper in binnen en buitenmantel als de RG-214.

 

Onderstaande tabel verduidelijkt veel: (met dank aan http://docs.lib.purdue.edu nanopub 208)

 

Een goede repeater zijn squelch opent bij -128 dBm.  Duidelijk dat het gekraak ver boven het squelchniveau uitkomt wanneer foute materialen gebruikt worden in de antennelijn of duplexer kabels..

De fenomenen treden alleen op bij zendvermogens >10 watt en meer op sites waar meerdere zenders samen opgesteld zijn. 

Het komt er dus op aan in dergelijke repeatersystemen alleen N-connectoren te gebruiken met verzilverde binnen en buitencontacten (Spinner heeft deze).

Bijkomend probleem is dat alle commerciële radioamateurantennes vernikkelde aansluitconnectoren hebben, wat nefast is voor PIM.

De bevestiging van de antenne in de montagekoker is dan ook nog eens zo’n slecht contact met één boutje dat bij de minste wind garant staat voor gekraak op de repeater.

 

Hallaar, 26 januari 2010.

More technical info about PIM in these files:

technicalpart/PIMAppNote.pdf

technicalpart/ARFTG52.pdf

 

All the information on this site is copyright . Authorization from the author is required to publish.

 

De gebruikte Duplex filter, een Celwave-RFS 526WB type, en de overspraak grafiek alsook de SWR en invoegverliesgrafiek per zijde. De duplexer is breedbandig afgeregeld en kan ingezet worden van 438,800-439,300 zonder noemenswaardige verschillen. Het is het beste type duplexer dat ik in mijn carriere al tegenkwam, maar heeft dan ook een prijskaartje van 3000 $ moesten we hem nieuw kopen. Deze is afkomstig van een oud netwerk van Elektrabel en werd voor een hamprijs verworven.

De klassiek in amateurkringen gebruikte Procom filters (of de Taiwanese namaak ervan, professionals noemen trouwens deze Procom's soms smalend "dakgootpijpfilters") hebben beduidend minder notch diepte (70 dB) terwijl deze tot 100 dB gaat, maar de netwerkanalyser was 'te slecht' om dit te meten) waardoor deze bij een 20 Watt al aan de grens van hun kunnen zitten.

De celwave-RFS 526 is ook gemakkelijk om te bouwen naar de 1,6 MHz shift Teven de Grafiek van het (optionele) bandpass filter tegen mogelijke Astrid overload dat momenteel NIET in gebruik is.

 

 

Der für ON0RBO & ON0TB eingesetzte Duplex-Filter Celwave RFS 526-WB-1 wurde breitbandig abgeglichen und kann ohne nennenswerten Messunterschiede zwischen 438.800 MHz & 439.300 MHz eingesetzt werden. Er ist laut Pedro ON7WP, der beste Duplexer in seiner Karriere den er je durchgemessen hat. Die Durchgangsdämpfung beträgt maximal 1.0 dB Die Klassischen für Amateurzwecke verwendeten Duplex-Filter der Marke Procom oder die taiwanesische Kopie, die Fachleute spöttisch auch manchmal als "Dach-Rinnen-Rohr-Filter" bezeichnen, haben eine wesentlich geringere Sperrdämpfung von ca. 70 dB gegenüber dem Duplex-Filter von Celwave der bis zu 100 dB  zwischen TX & RX dämpft .Leider war der benutzte Netzwerk-Analysator nicht gut genug um verlässliche Messungen durchzuführen. Der Duplex-Filter Celwave RFS 526-WB1 lässt sich auch einfach für 1,6 MHz Shift umbauen. Neu kostete der Duplex-Filter ca.3000$, unsere kommen aus einem alten Netzwerk der Electrabel
und wurden zu einem fairen Ham-Preis erworben.

 

 

 

 

 

The search for the Holy Grail Antenna…

 

Preface:

The author is since 1990 committed to repeater system design.  In the quest for the ultimate coverage, he is always looking for better sites, better cable and the best antennas.

When installing his latest multiband repeater system on a TV transmitting tower of 300 meter, he discovered that the common HAM Japanese multiband antenna is not always the best choice, especially not when you have some oversized towers at your disposal…

 

The problems started with the installation of the 1298,250 MHz Machine at 117 meter asl.  The repeater previously was located on a hilltop 50 m asl, on a moderate 26 m tower.  Performance was average, using a Diamond X5000 antenna.  When moved to the higher locations signals were far below expected values, especially when compared to a similar system in Antwerp.

When installing an additional machine on 438,650 MHz similar findings were observed.  Apparently the antenna location was too high so the main lobe was overshooting the target area.

 

Previous articles by some fellow Danish hams showed that some of the Japan- and Taiwan-made (not to mention the Chinese ones..) have quite good matching specifications but show main lobes heavily tilting upwards.  Commercial UHF gain antennas as used in Tetra applications use intentional downtilting by shifting phase between 4 or more stacked dipoles.  Another article is being prepared about dissection of two of these high-tec UHF gain collineairs.

We also know that it is possible to obtain downtilting by using a gain collineair below the design frequency, with a VSWR mismatch as a penalty.

 

Because the author doesn’t like changing antennas every week, he decided to do some controlled test setup to see how above theory is behaving in practice.  So one windy Friday night after work at 16:00 he mounted on one of the company towers with a radio and a bunch of commercial antennas on his back.  The day before he had prepared at test signal originating from his home QTH some 40 km away.  The test included absolute measurements of this known signal as well as receive signal strength measurements of some known repeaters, both in the upper and lower part of the 430-440 MHz band to distinguish the possible difference in performance in relation to the frequency.

 

Enclosed is an .XLS file with the measured results, the used equipment and the path loss calculations.

Also included a .pdf with the VSWR plots of most of the antennas to try to analyse the relation between the matching bandwidth of the antenna and the optimum receive performance frequency.

 


The results ?

 

Antenne Comparison measurements at 90 m agl in Brussels (telecom tower VRT)

 

Path loss:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

23 dBm source power from generator plus PA

 

 

 

 

 

 

 

-4 dB cable loss (30m H100 + Jumper)

 

 

 

 

 

 

 

+5 dB estimated antenna gain (Comet CX901)

 

 

 

 

 

 

 

+2 dB isotropic RX gain reference antenna

 

 

 

 

 

 

 

-116 dB path loss

 

 

 

 

 

 

 

 

 

estimated RX level = - 90 dBm at 0dBd reference antenna

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Corrected Measurement table

 

 

 

 

 

 

 

Type antenne

Merk

Construction

430,125

430,125

430,275

430,325

430,325

439,2

145,638

 

 

 

 

?

ON0LB

ON0GBN

PI4BRD

ON0LLV

ON7WP/B

with ATT

2x4 MAX

2-band

Comet

stacked 5/8

-94

-91

-88

-106

 

-98

-91

V2000

3-band

Diamond

1/2 + stacked 5/8

-91

-88

-88

-104

-96

-94

-85

CA712EF

monoband

Comet

12 x 1/2 coax coll.

 

-84

-86

-103

S8

-93

 

406-430

monoband

Procom

collinear array

-92

-89

-87

-102

-98

-92

 

Ringo

monoband

Cushcraft

collinear array

 

-89

-92

-105

-97

-100

 

440-470

monoband

Kathrein

collinear array

-92

-89

-92

-104

 

-94

 

406-430

monoband

Kathrein

collinear array

-92

-89

-89

-102

 

-94

 

CX901

3-band

Comet

1/2 + stacked 5/8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

It is very astonishing and not exactly what I was expecting, but the Diamond V2000 stood up very firm against the dedicated 5-fold more expensive Kathrein and Procom monoband antennas.  Very nice performing was also the Comet CA712EF coax collinear but here the frequency bandwith limitation is extreme and the 90 km/h wind velocity at the time of the test also showed unacceptable bending of this antenna combined with severe signal strength fading.  Definitely your favourite California sunshine antenna, but worthless for EU-applications.

 

I hope this small finger exercise of mine summarised in this report is of any value to colleague repeater builders, or just the average FM rag chewer.

 

I often laugh at those HF freaks by saying that the difference between the best and the worst 3 band beam with similar boomlength is less than 1 dB.  Well, this is definitely not the case with UHF antennas.  Who said HF was for pussies….

 

Hallaar, January 15th 2008

 

Pedro M.J. Wyns

ON7WP – AA9HX

technicalpart/UHF Monoband comparison.pdf

 

Some measurements:

1270 RX Loss

 

1290 TX Loss

on0tb 2 x 3 el vargarda

on0tb 70 cm ca712ef

on0tb rx

on0tb tx