repeater/duplexer question

[johnny1225]

Could someone help me. I have 2 frequencies into a hospital, using Mototrbo repeaters and sinclair filtering cans. 1 frequency is security the other is engineering. I would like to remove the security cans and use a Q3220 sinclair duplexer and leave the cans for engineering. I was wondering if someone could tell me how to add the duplexer to the system without hurting engineering. I know there is a way to do it because alot of our customers have it this way in there system.

thanks

john

[Bill_G]

A repeater is a repeater whether it's trbo or not. Both of your repeaters should have duplexers on them now.

Or are you trying to say you have a single trbo repeater with two users on different talkgroups on it? In which case, you probably should do nothing. There should be a duplexer there already. Maybe that is what the Sinclair cavities are.

[johnny1225]

there are two Mototrbo's (1) for security and (1) for engineering that are hooked up to sinclair cavity filters to the antenna system. The sinclair filters let you have two repeater systems on one antenna system. I am trying to add a duplexer to security trying to get more power. The sinclair cavity filters have 3.2 db of loss running through the cans where the duplexer has only 0.8 db of loss. I am trying to disconnect the security cavity cans and add the duplexer into the system with the engineering cans still in the system.

thanks

john

[Bill_G]

Okay - so you have two repeaters connected to one antenna through a set of cavities. Do you have the model number of these cavities? Look those over carefully. I think you'll find they are set up as duplexers already with a secondary purpose as a combiner. There has to be two ports for the rcvrs and two ports for the xmitters. I've seen some ingenious roll-your-own designs using pass-reject cavities without isolators to keep costs down. Depending on the freq separation, 3db loss through something like that would be normal, and in fact fantastically low. Adding a duplexer will present impedance imbalance that you will have to account for. You have to consider both repeaters in this, or both will suffer. You will not be able to connect a duplexer to the existing system and magically regain 2db. 2db is nothing. Even 3db is nothing when trying to increase building penetration depending on where you are trying to fill.

What is the real concern here? Is there a sudden (or gradual) loss of range, or simply a desire to increase in-building coverage? Hospitals get built one hallway at a time, and what used to cover the entire campus well doesn't reach into the tunnel level of the new addition. Perhaps the best solution is to relocate the security repeater to a different rooftop with it's own antenna, and put duplexers on both repeaters. Are they operating in trbo mode, or analog? We've seen some significant performance improvements in trbo mode using the same antennas that were used for years previously with conventional repeaters. It's pretty amazing.

[johnny1225]

basically bill, it is a tunnel that has been built. i am trying to add more power to the system to get the coverage. i have around 43watts at the repeater and 1200 feet of T-rad 600 cable from time microwave. The cavities are sinclair C3037 450-470 mhz cavities. do you think i need more power for this long run of 1200 feet leaky cable. or do you think the cable has a problem. i have just a 1/4 wave antenna on both sides of the tunnel. the cable loss for this cable is 2.2db/100ft but is leaky

thanks

john

[johnny1225]

bill the mototrbo is in analoge. the frequencies are security 452.3625/457.3625 tpl 123.0 engineering is 452.1750/457.1750 tpl 123.0

[Bill_G]

From the Sinclair website -

C Series Pass-Reject Expandable Multicoupler, 7" Cans

• Three bandpass cavities and one reject cavity
• Combines transmitters or receivers with inter-channel spacing down to 1 MHz

How many of these are being used - just one set or two? (I'm hoping at least two)

How are they hooked up? Do both xmitters go to one set, and both rcvrs go to the other?

How do they join to the single antenna line - through a TEE?

Was this an existing installation with in-building antenna distribution (leaky coax), or are you adding the leaky coax to feed the tunnel?

Are you adding a power splitter to feed an outdoor antenna as well as this in-building line?

The first thing I see is you have less than 200khz between the two channels, and this model specifically states minimum 1Mhz channel spacing. Is the original vendor available? Without being in front of your cavity array, and able to look this over, take some measurements, etc, it wouldnt be fair for me to say this looks and sounds wrong. I have seen some ingenious designs that worked well despite my reservations. There may be some very good reasons it is built this way. I would put some questions to them if they are around. But, this confirms my advise to not just throw a duplexer at it, and hope for the best. If this is not working for you, then both repeaters need to be considered.

As for the power you are applying - You are inserting approx +46dbm into a 1200ft leaky line with approx 26db of loss yielding +20dbm at the termination. That's a smokin hot signal. Even with 4db loss through this array, you are still super hot. I wouldn't insert into the line more than a -20db to just feed the tunnel. This would work with -80db inserted if your only interest is to feed the tunnel with a perfect signal end to end. If you want lateral building penetration, then I'd bump it up to -20db depending on the depth of penetration you wanted.

Do you have any outdoor antennas that need to be considered?

[Bill_G]

One more thing - with the freqs so close together, I would seriously consider changing the tone PL's to something different on each channel. I'm suspecting there is some mutual interference and rcvr falsing going on that needs to be addressed. Having four different PL's for input and output on the two channels would not be an unreasonable precaution.

[johnny1225]

Bill, there are 16 cans 4 for tx on one freq. 4 for the rx on the rx freq. that would be security then 4 for tx and 4 for rx on engineering freq. There are two antenna's on the outside of the hospital roof. there is also a SRL 301 sinclair antenna on i think the 1st or second floor of the hospital a total of 3 antenna's. The rest of the cabling is not leaky cable, the only leaky cable is the 1200 feet which leads to the tunnel and then down both sides of the tunnel, with a 1/4 wave on both sides. This system has been in for years about 10 years or so through another vendor. They are the ones that put the original system in and who has picked those frequencies. I understand what you are saying about the freq's being so close together i will do something about that.. That is why i thought that could have been the problem for the no coverage in the tunnel. What i did is i took a Q3220 duplexer on securitys repeater and turned off engineering altogether, to see if there was a change in coverage, but there was no improvement. The system is connected with tee's at the repeaters to the antenna cabling which i don't like, but used a 3 way splitter and still no coverage in tunnel. Where can i go from here do you have any ideas for me bill. We also did a test with a repeater right at the tunnel with only the tunnel hooked up and it worked like a top. Do you think i don't have enough power or do you think i am over powering the system. Any other idea's.

Thanks you bill for your help, i really appreciate it my friend.

John

[johnny1225]

bill, when you say you would send -20db down the line what do you mean? -20db is 10 uW's is it not??

[Bill_G]

Got an Anritsu or similar to sweep the tunnel line? That's what I would do next. There is something going on with that cable.

And yes, -20db is all that should be needed to serve the tunnel if everything worked well. With approx 26db loss in the length and a -61db coupling loss at 6 meters away, you should have no less than -107db the entire tunnel which should be well above the performance threashold of a good portable. Since -46db will be hitting the terminating quarter wave, actual rsl will be higher. If you insert +30db / 1W, you will have more than enough.

[Bill_G]

John - thinking about this over the weekend, one thing you should do is draw this distribution system out. If the hospital doesn't have a copy buried someplace already, you need to map out which cables go where. Based on what you've written so far with lines and antennas on both sides of the tunnel, and two antennas on the roof -

(a) there are either two distribution systems - one for the UHF and one for an unknown system (very possible)
(b) they have split the UHF system by user with separate distribution for eng and sec (not likely)
(c) they have split the UHF system into separate rx and tx distribution (highly possible)

Doing a little searching, I see you've asked about diagnosing this system before: http://batboard.batlabs.com/viewtopic.php?f=1&t=77503

So, this problem occurred some time ago which adds weight to the need to map this out for future reference. If it used to work, and worked well, then the hospital needs to invest in your company's services to create a system drawing. Their option is to deconstruct it, and go with a more conventional design like your company did with the additional repeaters. The danger there is interaction between eng and sec because their freqs are so close together. The other three systems may need to be considered as well.

[Ali Abbassi]

At what point in the RFDS did you connect your test RPTR , your problem seems to be between transmitter to the starting point of the leaky cable ,perhaps a bad jumper ,