NE-FL D-STAR System

The Journal of the North East Florida D-Star Repeater Network

Archive for category Antennas, Feedlines and Duplexers

Antenna Upgrades for EMCOMM-1

Oct 26

Posted by mikelee in Antennas, Feedlines and Duplexers, Go Kits, NET Operations | No Comments

The recently completed MS-150 event taught us much regarding what functions are most necessary – one of the lessons is that YAGI’s are often as important (or more so) than just omni-directional antennas. Many say that the GAIN from the Yagi’s is little more than a high-gain vertical, like the Diamond X-500. But in NetControl mode when communicating using established repeaters, the advantage of a Yagi isn’t just in the forward gain, but also in the REJECTION to the backside of the antenna, as well as the side (often exceeding 20db). This makes for less interference and a quieter NetControl operation.

..
I’m a fan of Arrow Antenna’s, solid-element yagi’s – they’re reliable, well constructive and (relatively) inexpensive – but recently, they’ve had availability problems due to a serious illness in the owner’s family. So I recently purchased 3 mono-band 2m beams from Cushcraft (A148-3S) – well constructed and reliable, like all Cushcraft products. Their 3-element beam is about $65 – about the same price as Arrow’s 4-element yagi. Either way, you can’t go wrong.

..
Remember – the advantage of a yagi is not just in Gain, but in REJECTION as well..!!

Technique – Tuning a Duplexer, On a Budget

Jul 14

Posted by mikelee in Antennas, Feedlines and Duplexers, Test Equipment | No Comments

With our advancements in the world of D-Star Hotspots (thanks LARGELY to Steve, KB4DNQ), some of you will undoubtedly be looking at creating your own hotspots – whether simplex (easiest), or full-duplex (most capable). If you ‘re thinking of full-duplex, then you’ll need a duplexer. UHF units are the cheapest, due to their small size.

Some of you have asked   “… how difficult is it to tune repeater duplexers..?” As in all things in life, the answer is “it depends”.

If you have a spectrum analyzer with a tracking generator, as well as a return loss bridge, the procedure is straightforward and takes maybe 60 minutes.

If you have just basic tools, it’s a bit more difficult and a lot less certain (you can’t SEE the results, only HEAR them). While the basic method can get you in the ballpark, it can sometimes be deceiving, in that you could be tuning into a cavity’s harmonic “sweet spot” and therefore not getting optimal results. There is no substitute for using the right tools for the job, BUT… when you don’t have (or didn’t BRING the right tools), it’s always good to be able to fall back to basics.

So here goes, using a signal generator, an attenuator, a power meter:

You can create a signal generator from an amateur radio and many feet of RG58 (which will help attenuate the signal).  Signal samplers can be had for $50 if you look on eBay, with even more elaborate units for around $100 which are laboratory qulity, more on that later.

A couple hundred feet of the worst RG-58x is about the right amount of feed line and attenuation to generate the required signal (more is better – and the lossier the better). If you need help, see the Amateur Radio Handbook for detailed instructions.

Almost all repeater duplexer are band pass, band reject. Most of the newer designs use two band pass cavities for transmit and two band pass cavities for the receiver and each cavity has a reject circuit, or notch. Each cavity should have two RF ports – if your cavities only have one RF port then you have a notch filter which is not going to get you the attenuation you need. If your cavities have two RF ports but do not have a notch adjustment then you have only a band bass cavity – you are out of luck, since there will not be enough rejection of the other side of the repeater (transmit or receive).

The signal sampler is a critical component for this procedure. RF Parts offers a very nice signal sampler, which will provide between 20-80 dB of attenuated signal. Another source is eBay – use the search term “Step Attenuator”. I got a couple of beautiful units manufactured by JFW (Model 50DR-055) which were “Factory Seconds” with only minor (almost un-noticable) cosmetic belmishes for less that $100 each. The units are rated from DC-2,200Mhz, 50-ohms, N-connectors, attenuation range from 0-30DB in 1-DB steps.  If you own or maintain a repeater system you can’t live with out one.

Tune the TRANSMITTER PASSBANDS - If the duplexer has not been tuned before for your frequency, you may want to tune each transmit band-pass cavity first, and then tune them together.

  • Connect your signal source to one side of the transmit cavity – you will be injecting TRANSMIT frequency
  • Connect your watt meter between the signal output of the band pass cavity (be sure the receiver band pass cavities are not connected) and the load (antenna or dummy load)
  • Inject a signal at the TRANSMIT frequency
  • Adjust the cavity for maximum RF power.
  • Repeat the process for each bandpass cavity
  • When all individual cavities are tuned,
  • Connect all transmit bandpass cavities together, then
  • Repeat the procedure on the cavities in SERIES and adjust for maximum RF power

Tune the RECEIVER PASSBANDS - Again, if this is the first time the duplexer is being tuned, you may want to tune each receiver band pass first, and then fine tune them together.

  • Connect your signal generator to one side of the receiver cavity – you will be injecting RECEIVE frequency.
  • Connect your signal sampler between the output of the receive cavity and the load.
  • Set the signal sampler’s attenuation to maximum (high DB),
  • Open the receiver’s squelch, and turn-up the volume.
  • Inject an on-frequency signal ample enough so it is noisy but can be heard. A 1 kHZ tone applied to the injected signal will help you hear the quality of the injected signal.
  • Now tune each receive bandpass cavity  for the clearest signal. You may have to increase the amount of attenuation on the signal sampler to keep the signal a bit noisy as you improve the bandpass tuning, doing this will allow you to hear the improvement made by your tuning adjustments.
  • When you’ve achieved the maximum signal clarity (listen for the 1khz tone quality), then
  • Proceed to tune all subsequent receive cavities
  • When all receive cavities are independently tuned,
  • Cable the duplexer for operation – cable all of the cavities, two (or three) on the receiver side and two (or three) on the transmit side.
  • Reconnect the signal generator and signal sampler – but this time for the SERIES of the receive cavities.
  • Repeat the tuning process with all of the receive cavities in series. Note that the cavities will interact with each other, so changes on one cavity will affect the signal quality coming from the other (this can be an extremely frustrating process – and is where MOST of the time is spent).

Tune the TRANSMIT NOTCH FREQUENCIES  - this eliminates your receiver signal from the transmitter input.

  • Connect your transmitter to a SEPARATE ANTENNA – not to the duplexer..!! You will be transmitting on the RECEIVE frequency.
  • Connect your receiver to the transmitter cavity where the signal generator used to be
  • Listen on the receive frequency thru the transmitter cavities (in series) and adjust the attenuation so you can hear a noisy signal. Adjust each of the transmit notches in order to REDUCE the received signal (rejecting the received frequency)

Tune the RECEIVE NOTCHES - This eliminates the transmit signal from the receive input.

  • Connect your transmitter to a SEPARATE ANTENNA – not to the duplexer..!! You will be transmitting on the TRANSMIT frequency.
  • Connect your receiver to the receiver cavity where the signal generator used to be
  • Listen on the transmit frequency thru the receiver cavities (in series) and adjust the attenuation so you can hear a noisy signal. Adjust each of the receive notches in order to REDUCE the received signal (rejecting the transmit frequency)

Always tune the notches LAST, since the notch adjustments track with the bandpass adjustments.

OK, I did just what you said to the letter and I still have a problem. Well, bummer. What are my options..?

If you still have desense in your system and the desense is constant, then find a spectrum analyzer and a signal generator and tune the duplexer by the factory procedure. There are a number of folks from Daytona thru Jacksonville who may be able to assist (cheap – for a bottle of beer..!!)

If you are using a re-purposed Commercial duplexer, you may need to change the interconnecting cables between the cavities, since each of those cables are cut-to-frequency. Although usable, it will not be optimal without exchanging. If you are using Tx/Rx Systems duplexers, we can provide you with the exact length of both VHF and UHF interconnect cables for amateur service.

Another possibility is that your duplexer does not offer the required attenuation or frequency separation needed. For a 100 watt transmitter and a receiver with a sensitivity spec of .5 uV you need about 88 dB of attenuation.

The following table lists the typical attenuation required between a transmitter and a receiver which will result in no more than a 1 db degradation of the 12 db SINAD sensitivity. This information is based on a receiver with .5uv sensitivity (which is pretty crappy).

Transmit Power 2 Meter
600 kHz. Frequency Separation		 70 CM
5 MHz Frequency Separation
25 Watts 82db 55db
50 Watts 85db 58db
100 Watts 88db 61db
250 Watts 92db 65db
350 Watts 98db 70db

Jumper Cables for Repeater Service – and Old Wives Tales

Jun 5

Posted by mikelee in Antennas, Feedlines and Duplexers | No Comments

So what’s the big deal..??!!

Coax is coax, right..??  The only factor that we should be concerned about is the LOSS or attenuation of the selected cable at the operating frequency. And since most repeater jumper cables are only about 5-6 feet long, we should be able to use virtually anything, including RG-58 which only has loss of about about 0.3db at 100mhz, versus LMR-400 which has attenuation of about 0.07db. Okay, so the difference is 0.23db – but that’s nothing, right..??!!

From the context of the discussion, you would be absolutely right. The loss differential between the two identical length cables wouldn’t make a hair’s bit of difference. And in commercial 2-way service of mobiles or portables, it would be completely acceptable (not preferred, just acceptable). Remember that most 2-way service is SIMPLEX (or half-duplex), with the radio either transmitting, OR receiving at any given moment. This makes it easy, since we don’t have to worry about desensitizing our own receiver when we transmit.

But repeater systems are inherently different – almost all being FULL-DUPLEX in nature (unless it’s a store-and-forward style, Simplex Repeater). In full duplex mode, the receiver is ALWAYS receiving, even when the transmitter is “ON”. Therefore, we have to be very sensitive (sorry for the pun) as to what enters the receiver’s passband. Given that receivers are not (due to cost constraints) extremely selective, there is always the fear that energy from the transmitter might enter the receiver front-end and desense it by making it deaf, or in worst case situations, cause it to create intermod products which affect both your repeater and potentially other radio systems in the immediate area. The technical term for this is:  BAD

Most amateur repeater owners (and many independent commercial operators) do not realize that the quality of their coax jumpers from the transmitter and receiver to the duplexers are CRITICAL for optimum and trouble free operation. In this case, it is NOT about loss characteristics of the coax, but about the shield-coverage of the cable. Consider for a moment, that both RG-58A and RG-8A/U have specified shield coverage of 95-96% – that allows for signals from a transmitter to permeate thru 4-5% of the outer circumference of the cable (the WHOLE length of the cable). In the case of a receiver, that allows a HUGE amount of signal from the transmitter to pass into the front-end of the receiver – again, causing desense or worse.

Take a look at the picture below – in this case, RG-58. The shield (or braid) is quite loose – you can see thru portions of it, since you can see the white dielectric material thru the braid. Great for 2-way simplex since it’s light and flexible, but horrible for repeater work.

Now look at the specs for RG-8. Most folks would say, “hey… BIGGER cable, so it MUST be better” and again, for 2-way simplex, they would be right. But look closely at the specs below..  96% shield coverage, so it’s only marginally better than RG-58 for repeater work (loss differential is negligible at a length of of 4-6 feet).

Now – look at Heliax. It does not matter which size, since all are constructed in the same manner with virtually identical materials. Helix has a SOLID, non-interrupted shield which is extruded copper (Cu) placed over the foam dielectric material, then surrounded by the PVC or Teflon jacket material. The shield has 100% coverage with no gaps – even to the point of the connectors on both sides. In fact, the connectors are engineered specifically for each size of cable and contain materials which resist corrosion between the cable and the connectors, which is known to cause intermod. I’ll bet you didn’t know that sub-standard coax or connectors could cause intermod, right..?? Take a look at the link below and you’ll see just how significant this problem could be.

andrew-braid-over-foil-imd.pdf

So this is Heliax – Solid, 100% shield coverage, a PAIN to work with unless you have the right tools, connectors and training (there is actually an Andrew Heliax Certification that installers must get in order to work on Cellular sites)  - 100% commercial standard and 100% trouble-free if installed correctly. Is it expensive compared to “normal” coax cable – YOU BET..!!   LDF4 (the most common Heliax) runs about $2+ per foot and the connectors are about $25 each. But you gets what you pays for, as they say. Remember our mantra from our original D-Star presentations – “We will install to commercial standards in order to gain reliability and ease of maintenance”. It is no coincidence that ALL major cellular companies use Heliax on their cell sites. It is also no coincidence that WE use Heliax on our systems – it clearly is the best that we can use. Below is a picture of Heliax (LDF4, in this case) – click on the picture to see an expanded view:

Are there other cables which provide 100% shield coverage – absolutely – and we sometimes use them as well. RG-142 comes to mind since it is super flexible, but it has higher loss characteristics and should only be used for short lengths (TxRx Systems uses RG-142 for their duplexer cables between cavities). BTW, RG-142 has a Teflon jacket and is the ONLY cable authorized by the FAA to be used in avionics system on airborne systems.

Well there you have it..
I’ll bet you know more now about Repeater Jumpers that you ever thought you needed to know. But this series of blogs is about EDUCATING you about what we’re doing, not hiding it from any of you. We are about building a cadre of Amateurs who are willing, ready and TRAINED to help support these systems.

Let us know what you think of these articles thru your comments on any of the posts.

The Agilent N9330B Antenna/Cable Analyzer

Jun 2

Posted by mikelee in Antennas, Feedlines and Duplexers, Test Equipment | No Comments

As we all know, properly installing and maintaining repeater systems is not only a time-consuming effort, but requires some pretty sophisticated equipment in order to do it to commercial standards. Yes – many folks go the way of McGyver either because of lack of access to the proper tools, or lack of funds – and most do a respectable job. But not having the right tools just makes it harder, more time consuming and more prone to errors.

Take the tuning of duplexers as an example. There are those who do it with 2 handheld radios and a step-attenuator and just “listen” for differences in signal clarity. I have known (and worked with) guys who insist that this works, but invariably this leads to off-tuning of the cavity filters and potential intermod due to the passing of undesired signals. The right way is to use a Spectrum Analyzer with a Tracking Generator and a good quality Return-Loss-Bridge like the Eagle RLB-150x4a which will go to 2ghz.

The other case is in the verification of Feed-lines and Antennas. Many of us use the Antenna Analyzers from MFJ or equivalents – darn good devices for what they cost. But in the commercial world, we are required to sweep with units which operate with Frequency Domain Reflectometry (FDR)- a technology which has been around for about 15 years and which measures Return Loss using specific frequencies (versus Time Domain Reflectometry or TDR, which uses DC). The FDR technique is MUCH more accurate and allows us to sweep antennas which are in close proximity to other transmitting systems which are currently in operation.

The current leaders in manufacturing commercial-standard Antenna/Cable Analyzers are Anritsu and Agilent (formerly HP Test and Measurement). The Anritsu SiteMaster has been the mainstay of the Cellular Site Antenna/Cable analysis for many years – with most Cell companies having required “Anritsu Certification” by their contract installers of feed-line and antennas. Agilent has stormed on the scene in the last 5 years with (arguably) a better unit.

We have had the Anritsu SiteMaster S113C for over 5 years now, having purchased it second-hand from a fellow commercial services company. It has been and continues to be a workhorse, but it only sweeps to 1.8ghz – not enough to diagnose the newer technology signals which are above 2ghz (like WiFi, WiMax and some of the higher Amateur bands). It is also difficult to read in the dark due to the lack of back-lit keys. Therefore, we just purchased a new Agilent N9330B which will sweep to 4ghz and are putting it into service. The N9330B gives us even more recording capability, so that we can store sweep information and analyze it when we return to the lab – and include those sweeps into our systems documentation (see our post on April-11 and May-20).

So we welcome the Agilent N9330B to it’s new home – may it provide us years of faithful service and enable us to spot problems with our transmission systems WELL before they become service interruptions..!!