Hello again everyone. Sorry for the extra long post, but wanted to
make sure not to leave out anything, so as to hopefully minimize the
need for further follow up questions on these facts.
SYSTEM DESCRIPTION
Just as a refresher, this is a 6-channel VHF MPT1327 Trunking system.
It consists of 6 Motorola MTR2000 100W repeaters, fitted with a
circulator on the TX output stage, and optional Preselector on the
back of the unit. Also part of the system are a factory tuned Telewave
M101-150-6TRM TX Combiner (Cavity/Dual Isolator type), and a Sinclair
RM201-112S1B RX Multicoupler fitted with a BP Preselector (with 1MHz
pass-band, and factory tuned to 165.625), a "Hi Level" Bipolar preamp
(Angle Linear brand / anglelinear.com), and a 12 port power divider.
Between the preamp output and the power divider there is a 2dB pad
fitted. Since we only use 6 output ports of the 12 available, the
remaining ones have been fitted with 50-ohm terminators on them.
Antenna system consists of two Decibel DB228 8-dipole arrays, mounted
on a 90ft self supporting tower. RX antenna is at the top, TX is
mounted from about mid-tower down. Vertical separation between
antennas is not much due to tower height limitations, and each
antenna's length, but each DB228 is mounted on a separate adjacent
tower leg. Feedlines are 7/8 hardline running on separate tower legs
up until the point where they start to get horizontal to be routed
into the building next to the tower. At this point they run parallel
to each other for about 40ft. Grounding kits have been installed on
each cable at the top and bottom of each run at the tower. A
Polyphaser VHF50HN Surge Protector is going to be fitted soon between
the equipment rack and the end of the 7/8 cable runs with a 1/2
suplerflex jumper. Also another grounding kit will be installed just
prior to where the two 7/8 cables enter the building. For the time
being the ends of the 7/8 lines are fitted each with a double Female
N-connector (in place of the Polyphaser unit, and connected to the 1/2
suplerflex jumpers. The tower being used was in service for a
cellsite, but since has been abandoned for some years. Since the tower
structure is in good shape, only the grounding system proved to
require some attention. Measurements show that the 4 grounding points
show 995, 85, 170, and 218 ohms. The grounding system is going to be
completely redone on our next visit to the site. Adjacent to the tower
are three satellite up/down links. Grounding system on one of them
shows less than 20-ohms (with 4 rods), so it should be no problem to
get our grounding into spec. The only other system fairly close by is
another VHF repeater (plain-Jane GR300), but so far apparently it does
not seem to be any cause of concern. It is located on a separate tower
about a mile away from ours, and has about the same height. Operating
frequencies of this system are 157.700 and 162.700 (but not sure which
is TX).
Combiner to repeater TX port jumpers are made of Belden 9913, and
between RX Multicoupler and RX ports we were using 100% shield RG58.
Since at least one person expressed concern over the usage of said
cables, we changed the RX jumpers to RG142, and then we found a way to
temporarily hook up the combiner directly into the TX ports in order
to remove all 9913 jumpers. See tests bellow for more info on this.
Operating channel frequency pairs start at T160.125 R165.125 (for CH1)
with 200KHz increments to T161.125 R166.125 for CH6.
SYMPTOMS
When the system was installed, under certain circumstances we noticed
an unwanted (weak) signal appearing in one or more repeater receivers.
It barely opens the squelch on the affected repeaters, but it is
enough to cause problems. The signal manifests itself as an indication
of the Carrier LED of each affected repeater coming on, even if the
particular channel is not in use. The interference was consistent with
our own operations, and did not seem to depend on other external
factors, as the other repeater system mentioned before. We monitored
activity of this system, and it did nothing to change our interference
when it was active.
Basically the interfering signal shows up as described above when
several of our own TX signals are present. This normally would happen
when the control channel repeater (usually CH1) is on the air, and a
call is placed into the system. Interference showed up most
prominently when CH5 or 6 where assigned to the call as traffic
channel, but would also happen with other channels. When a call is
placed, this would bring up contentiously another repeater TX for the
duration of the call. The interfering signal would mostly only show up
whenever the PTT of the radio that initiated the call would be keyed.
This happened both with the portable being keyed right in front of the
equipment rack, and also from 300-400 feet away. Although sometimes we
where also able to get the interference to show up when several
repeaters where transmitting at the same time, and without need for
the portable to be keyed at the same time.
We were also able to replicate the same condition when the same two
repeaters where keyed manually (the ones formerly used for control/CH1
and traffic channel/CH5 or CH6) with their mikes, and a conventional
portable radio programmed to the traffic channel repeater's RX. When
all three were keyed, at least one or more of the Carrier indicators
on other idle repeaters came on. Since the easiest condition that
enabled us to see the interfering signal was the method described
before, we used it throughout the rest of the tests that where done
and are listed bellow.
TESTS
One of our first tests was to look at the combiner output with a
spectrum analyzer trough a "Lossy T". It looked pretty clean with only
two or three smaller peaks on the sides of the main carrier, all
between 40-50db down or more.
Note: between tests, unless specifically noted, equipment was returned
to it's normal layout before performing the next test.
1. Swapped TX and RX antenna connections at equipment rack (Tx antenna
to Multicoupler, and Rx antenna connected to combiner output.
Interference WAS present (no variation from the normal connection).
2. Disconnected RX antenna from the system (input of RX multicoupler.
Interference was NOT present.
3. Disconnected TX antenna from the system (output from combiner), and
connected a dummy load. Interference was NOT present.
4. Connected a base station antenna that was already present on the
tower (used with a fixed radio on the premises) to the TX combiner
output, and a mobile antenna placed just outside the window of the
radio room to the RX multicoupler input. Interference was present.
Swapped both antennas with same result (Interference present, no
variation).
5. With the mobile antenna still as RX, hooked back up the system TX
antenna, then put a 6dB pad between the Preselector input of the RX
multicoupler and the input port, then moved to between the Preselector
and preamp input, and lastly, moved it between the preamp output and
the power divider. There seemed to be some slight variations on how
many or which channels where affected with the interference, but
otherwise this test was not really very conclusive. So basically the
interference was still present throughout the test. Then hooked mobile
RX antenna directly into the power divider (in order to exclude the
preamp), but interference was still present.
6. While the interference was present we tried grabbing, wiggling,
moving, etc on all rack cables (control, power, and RF) with no
variation. We did notice that sometimes we could find a position in
which the portable could be keyed inside the radio room and not
produce the interference, but it really was not very conclusive.
7. Removed all connections between the MPT controllers and equipment
and the repeaters. Turned off all of the MPT equipment. Removed all
grounding wires from the rack equipment. Interference WAS still
present (used manual TX keying with repeater mikes and the
conventional portable).
8. As mentioned before, all RX jumpers between the repeater RX ports
and the multicoupler output ports (formerly RG58) where replaced with
RG142 jumpers. No Change and interference was still present. These
RG142 jumpers where left on the system for the remainder of the tests.
9. Since the combiner has pretty long cables between each input port
and the corresponding isolator, we were able to pull the input cables
outside of the combiner rack and reach their corresponding repeater TX
port directly without the use of the formerly mentioned 9913 jumpers.
All but one channel (CH2) could be reached, so it was left unconnected
and this repeater was turned off. Still the interference was present.
10. Completely disconnected all combiner cables from the repeaters.
Then removed both TX and RX system antennas (DB228) from combiner
output and multicoupler input. Hooked up one antenna directly to CH5
TX port at the repeater, and the other at the CH6 repeater TX port.
This way the combiner was completely out of the loop. Hooked the
mobile antenna outside the window to the RX Multicoupler input.
Interference was present. Disconnected mobile antenna from
multicoupler and interference was NOT present. Bypassed the preamp in
the RX multicoupler with a double female N-barrel connector AND turned
off the power to the preamplifier. Interference was still present. As
a side test just to see how hard RX had become, we determined that at
this point about 1uV was needed at the multicoupler input to open up
the repeater squelch (Preselector and power divider without the
preamp).
SOLUTIONS?
In view of all the previous test it would seem that the problem would
be related to the RX path. Basically the whole TX path (combiner,
antennas, and cable run) was taken out and the interference continued.
Since at one point even the multicoupler preamp was also bypassed,
this would lead me to think that the Intermod could be happening right
at the repeater receivers due to signal overload. Since our
multicoupler uses the bandpass type Preselector, and our antennas are
fairly close to each other, it could be possible that enough signal is
still getting into the receivers to overdrive them. Please feel free
to correct me on this and add any comments and suggestions. Since we
are no longer at the site, at the moment there are no other tests we
can perform. But if something comes to mind, please let me know
anyway, as it is possible that we hopefully have already done
something similar, or we will be able to do it upon our return to the
site. Otherwise we'll to have and wait until we have a chance to get
back to the site in 2-3 weeks time (depending on how long it takes for
the additional filtering equipment that is going to be ordered to
arrive). It is our intention to be able to arrive at the site will all
the needed equipment (filtering, etc) required to solve the problem
for good, so we would of course very much appreciate any information
that would enable us to order the right equipment, so as to hopefully
avoid any unnecessary or possibly ineffective equipment purchases.
1. Since our Multicoupler Preselector is of the bandpass type, we
could probably gain more isolation between RX and TX by using a
Pass-Reject type Preselector tuned in a way so maximum rejection
occurs on the 160-161MHz Tx portion. I've looked over several vendor's
catalogs, but all they have to offer are the BP type. If anyone has
any specific suggestions (including a brand and model) of a
pass-reject Rx Preselector please let us know.
2. If we are right in our preliminary assessment, our primary concern
would be to try to keep our the TX signals as much as possible from
our receiver. It has been suggested to use staggered notch filters
(with reactive stubs to alter the notch shoulder response to one side
of the frequency center as desired in this case) on the RX pad, and I
guess that would mean one notch for each of our TX frequencies, so we
would need 6 pass-reject cavities ahead of the RX multicoupler input.
Again, if anyone has any comments about this, or specific suggestions
(including a brand and model) please let us know.
3. It was also suggested to additionally also put another set of notch
filters (also with reactive stubs) on the TX output path. Each of
these would be tuned to a RX frequency channel, in order to reduce any
TX noise on these frequencies. Right?
4. Another solution involves using a Sinclair Q2330 6-cavity Res-Lock
duplexer (said to provide up to 100dB isolation) and to use a single
DB228 antenna. Not sure if this would be the ultimate solution, but if
viable it might be easiest, although then we would be left with one
unused antenna. Still we could then go the route that suggested to
split up the Tx combiner into two TRM3 units. Than one antenna would
be used in duplexed mode for RX and TX of even number channels
(2,4,6), and the second antenna could be used for the odd numbered TX
channels (1,3,5). Any thoughts on this?
5. Finally, it was also suggested that it would be recommendable to
put an harmonic filter (lo-pass) between the Tx combiner output and
the TX antenna, mostly as a "just-in-case" measure. Appreciate any
comments and suggestions on this, and also, if you have a preference
towards any specific brand and model, please let us know.
CONCLUSION
We would of course hope to solve the problem with the least amount of
pain. But ultimately we need a solution, no mater what. The system has
to work as planned or all our efforts will be wasted. Not to mention
the monetary factor. In view of that:
- Possible solution #1 looks like the easiest to implement, but will
it be enough? Where can we get a pass-reject Preselector that would
fit our needs (preferably a RX only Preselector, for size and cost
considerations would be best).
- The #2 notch filter solution (at least on the RX chain at first)
seems to be the more elaborate, but could probably get somewhat
expensive, and will also probably take up a lot of additional room. If
#3 would also be implemented, that would certainly be more problematic
in both terms. But if this is the solution that has the best chance of
success, than we will have to bite the bullet. Not sure if one has to
order separate pass-reject cavities for each frequency, or is there
something available that would be more practical to use? If you have a
preference please let us know the brand and model you would think
would be best in this case.
- Solution #4 seems like the half-way one in terms of cost and
complexity. It would not take up too much additional space, and it is
relatively easy to implement. Besides it would also bring better
performance to the TX chain by splitting up the combiner. But will it
work only by itself?
- Is #5 really needed? We understand that a Cavity-Isolator combiner
in itself should provide adequate 2nd harmonic attenuation. Still we
would like to hear opinions on this one too. It will not cost much,
but as said before, is it worth in this case?
- Another option we still have open is to change frequencies. As it is
now nothing is written in stone yet. If changing frequencies could
help, as some of you have suggested, I would hope to do it on the RX
side, as doing so on the TX would certainly require retuning the
combiner cavities and we rather would avoid that. But again, if it
would be a worthwhile lead, we have no problem doing so too, provided
that the combiner is really "flied tunable".
- Also wanted to inquire about the TX jumpers between the combiner and
the TX repeater ports. We were using 9913, but it seems this could be
a no-no, even though they don't seem to be creating any problems at
this time. Still we would want to replace them with another coax with
better shielding. Someone recommended RG214. Is this a good option? Or
should be use 1/4 (or 1/2) supleflex? although it can become quite a
pain to fit these semi-rigid cables into the existing layout. What
other more "flexible" cable options could we consider?
Thanks again to all persons that so far took the time and effort to
assist us. As commented in earlier messages, it is always very
comforting to count with such a great spirit of cooperation when one
is in the middle of nowhere, and stranded with this kind of problems.
It is also our hope that by posting all this info to the group will
enable other persons to benefit from this experience when maybe in the
future anyone might have to face a similar situation.
Alex
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/Repeater-Builder/
<*> To unsubscribe from this group, send an email to:
[EMAIL PROTECTED]
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
