On 17/03/15 08:12, Jeff wrote:
It is all here in the rest of the thread but here goes. I have only
three antenna mounting points available on my home for antennas to serve:
a dedicated two meter transceiver for a digital node,
a dedicated Seventy Centimeter transceiver for a D-STAR hot spot.
a separate 120 CM transceiver
the six meter band of multiband transceiver
a dual band 146/446 transceiver

That is five radios that need to be able to use Six Separate VHF/UHF
frequencies with as many as four transceivers

operating at the same time. Two of those transceivers would be
operating as automatic stations so as many as three of

those transmitters may be transmitting at the same time. Two of the
three may even be on the same band given that their

will be an automatic station on both Two Meters and Seventy Centimeters
and I may be operating as a manual control operator

on either of those bands at any given time.

I initially asked for advise on which dual band antenna to use to
serve a 144 Mhz and a 440 MHz digital stations.

Someone then suggested that I use two tri-band antennas in order to
provide all band capability for FM operations on 6, 2, 1.25,

and 0.7 Meter bands while still operating the two automatic stations on
144 & 440. I would use vertical separation to reduce the

de-sensing on the common band stations.

I then asked for advise on any foreseeable problems with using the DCI
filters to effectively separate the signals at the radios

and yet allow the transceivers to share two multi-band antennas to cover
the needed bands with only two mounting points.

I'm saving the third mounting point for stacked Yagi-Uda beams to
provide additional range on 146/440 for the repeaters at the

extreme ends of the service area of the Forecast Office that I am
serving as Assistant Coordinator for Net Management

in their Skywarn Program. I am hoping that between the beams and two
amplifiers with receive pre amps I will be able to take reports when
appropriate from some of the more distant sub nets. I have done this
previously with temporary masts from this location so I am very hopeful
that it will work using more permanently mounted beams. All of that is
so you will know why the third mounting point is not available for the
needed omnidirectional vertical antennas.

I hope that makes the expanding questions clear.

Tom

The first thing that you have to consider is that none of the DFCI
filters will give you any protection when 2 radios are operating in the
same band, and paralleling things up with tee pieces is generally not a
good idea.


The issue with using Tee pieces is inadvertently inserting 'stubs' etc.
If care is taken to avoid this, eg by keeping leads very very short,
then at the frequencies being used, things should be OK. After all, a
duplexer is just a set of filters in one box (ie the connections are
very very very short). In fact, as the DCI filters are bandpass (based
on Tom's post, I've not looked at the specs), they are probably better
in some ways than some duplexers which just use LPFs and HPFs.
(Triplexers tend to use at least one BPF but some only use an
additional HPF.)

On 3/17/2015 5:19 PM, Brian Reay wrote:
On 17/03/15 08:12, Jeff wrote:
It is all here in the rest of the thread but here goes. I have only
three antenna mounting points available on my home for antennas to
serve:
a dedicated two meter transceiver for a digital node,
a dedicated Seventy Centimeter transceiver for a D-STAR hot spot.
a separate 120 CM transceiver
the six meter band of multiband transceiver
a dual band 146/446 transceiver

That is five radios that need to be able to use Six Separate VHF/UHF
frequencies with as many as four transceivers

operating at the same time. Two of those transceivers would be
operating as automatic stations so as many as three of

those transmitters may be transmitting at the same time. Two of the
three may even be on the same band given that their

will be an automatic station on both Two Meters and Seventy Centimeters
and I may be operating as a manual control operator

on either of those bands at any given time.

I initially asked for advise on which dual band antenna to use to
serve a 144 Mhz and a 440 MHz digital stations.

Someone then suggested that I use two tri-band antennas in order to
provide all band capability for FM operations on 6, 2, 1.25,

and 0.7 Meter bands while still operating the two automatic stations on
144 & 440. I would use vertical separation to reduce the

de-sensing on the common band stations.

I then asked for advise on any foreseeable problems with using the DCI
filters to effectively separate the signals at the radios

and yet allow the transceivers to share two multi-band antennas to cover
the needed bands with only two mounting points.

I'm saving the third mounting point for stacked Yagi-Uda beams to
provide additional range on 146/440 for the repeaters at the

extreme ends of the service area of the Forecast Office that I am
serving as Assistant Coordinator for Net Management

in their Skywarn Program. I am hoping that between the beams and two
amplifiers with receive pre amps I will be able to take reports when
appropriate from some of the more distant sub nets. I have done this
previously with temporary masts from this location so I am very hopeful
that it will work using more permanently mounted beams. All of that is
so you will know why the third mounting point is not available for the
needed omnidirectional vertical antennas.

I hope that makes the expanding questions clear.

Tom

The first thing that you have to consider is that none of the DFCI
filters will give you any protection when 2 radios are operating in the
same band, and paralleling things up with tee pieces is generally not a
good idea.


The issue with using Tee pieces is inadvertently inserting 'stubs' etc.
If care is taken to avoid this, eg by keeping leads very very short,
then at the frequencies being used, things should be OK. After all, a
duplexer is just a set of filters in one box (ie the connections are
very very very short). In fact, as the DCI filters are bandpass (based
on Tom's post, I've not looked at the specs), they are probably better
in some ways than some duplexers which just use LPFs and HPFs.
(Triplexers tend to use at least one BPF but some only use an
additional HPF.)


They do not have to be short - but as long as you have a multiple of 1/2
wavelength between the two diplexer inputs, each input will see the very
high impedance of the other input. Rather easy, since you are operating
on the third harmonic.

This could also be a good application for a hybrid ring duplexer. It's
a bit more complicated to build, but should provide better isolation.

In either case, the trick is going to be getting the lengths of the coax
correct. This is where a grid dip oscillator would be advantageous -
actually measure the coax, instead of depending on the listed velocity
factor.

And BTW - duplexers can also be bandpass. Can duplexers are typically
used on VHF and UHF repeaters. They can have very high Q (and therefore
very narrow passbands).

--
==================
Remove the "x" from my email address
Jerry, AI0K

==================

On 3/18/2015 3:28 AM, Jeff wrote:

They do not have to be short - but as long as you have a multiple of 1/2
wavelength between the two diplexer inputs, each input will see the very
high impedance of the other input. Rather easy, since you are operating
on the third harmonic.

Don't forget that it is not a simple as being 1/2 wavelength long, you
have to take into account the input impedance of the filer which may
substantially modify what you are looking at, and may load the other
filter affecting its response. The DCI dual band filters are aligned as
a unit so those sort of effect can be taken into account.

Jeff

The input impedance of the filter active filter would be 50 ohms, so no
loading. And the impedance of the other filter would be very high.

I would not expect matching a 50 ohm antenna to a good quality filter
(which these are) would affect their response significantly enough to
cause a problem. After all, this is the exact type of operation they
are designed for.

You *might* be able to measure some effect in a lab, but in the real
world it won't make any difference.

--
==================
Remove the "x" from my email address
Jerry, AI0K

==================

On 3/18/2015 6:49 AM, Brian Morrison wrote:
On Wed, 18 Mar 2015 06:44:55 -0400
Jerry Stuckle wrote:

I would not expect matching a 50 ohm antenna to a good quality filter
(which these are) would affect their response significantly enough to
cause a problem. After all, this is the exact type of operation they
are designed for.

The stop band performance is dependent on the antenna and filter
impedances out of band, at certain phases this can lead to a
re-entering stop band. I've been dealing with the consequences of this
in my day job very recently, it is a real and significant effect in
some circumstances. It's also not trivial to fix.


True. However - this was specifically made for 144/440 mhz with 50 ohm
impedance. As long as these are the bands he's using and he stays
within a reasonable range of 50 ohms, I don't see what the problem would be.

Again - these units are DESIGNED for this operation. What good would it
be to design and sell such an item if it doesn't work?

DCI has good engineers who know their stuff. Their equipment is top
quality. And while I've never tried this particular unit, I know they
would not let something with your perceived problems out the door.

--
==================
Remove the "x" from my email address
Jerry, AI0K

==================

On 3/18/2015 1:52 PM, Brian Morrison wrote:
On Wed, 18 Mar 2015 10:55:18 -0400
Jerry Stuckle wrote:

Again - these units are DESIGNED for this operation. What good would
it be to design and sell such an item if it doesn't work?

Well, not much, sometimes you can get caught out by something
unexpected when a change is made to fix a problem and does so but then
you discover some unintended consequences.

That's exactly what I have been dealing with in a system where it's not
easy to measure the frequency response of a combined antenna and filter
in a radiated test. But we managed to do it, just took a while to
create an appropriate test system.


It may not be easy to measure when that isn't your main business. But
for a company which is, I'm sure they have the appropriate equipment and
procedures.

--
==================
Remove the "x" from my email address
Jerry, AI0K

==================

On Wednesday, March 18, 2015 at 8:13:58 AM UTC-4, Jeff wrote:
On 18/03/2015 10:49, Brian Morrison wrote:
On Wed, 18 Mar 2015 06:44:55 -0400
Jerry Stuckle wrote:

I would not expect matching a 50 ohm antenna to a good quality filter
(which these are) would affect their response significantly enough to
cause a problem. After all, this is the exact type of operation they
are designed for.

The stop band performance is dependent on the antenna and filter
impedances out of band, at certain phases this can lead to a
re-entering stop band. I've been dealing with the consequences of this
in my day job very recently, it is a real and significant effect in
some circumstances. It's also not trivial to fix.



Indeed Brian, also if you look at the DCI filters there are quite
large returns in the stop bands.

Jeff

What are the effects of this on my proposed operation?

--
Tom Horne W3TDH

On 3/18/2015 1:32 PM, Jeff wrote:
On 18/03/2015 13:54, Tom W3TDH wrote:
On Wednesday, March 18, 2015 at 8:13:58 AM UTC-4, Jeff wrote:
On 18/03/2015 10:49, Brian Morrison wrote:
On Wed, 18 Mar 2015 06:44:55 -0400
Jerry Stuckle wrote:

I would not expect matching a 50 ohm antenna to a good quality filter
(which these are) would affect their response significantly enough to
cause a problem. After all, this is the exact type of operation they
are designed for.

The stop band performance is dependent on the antenna and filter
impedances out of band, at certain phases this can lead to a
re-entering stop band. I've been dealing with the consequences of this
in my day job very recently, it is a real and significant effect in
some circumstances. It's also not trivial to fix.



Indeed Brian, also if you look at the DCI filters there are quite
large returns in the stop bands.

Jeff

What are the effects of this on my proposed operation?


The effects are that it is not a good idea to parallel filters using
coax and tee pieces.

Jeff

Except he is not paralleling filters. The two diplexers feed different
radios.

And how would YOU suggest he hook them up?

--
==================
Remove the "x" from my email address
Jerry, AI0K

==================

On 3/19/2015 5:32 AM, Jeff wrote:

Except he is not paralleling filters. The two diplexers feed different
radios.


If you had been keeping up with the thread that was exactly one the
question that the OP was asking: could he parallel filters using coax
and tee pieces.

I have suggested a solution that only uses diplexers; the addition of
filters will probably not make much difference to the resulting
performance unless the radios in use have extremely high harmonic or
spurious outputs or produce very excessive wide band noise or there is
interference from 3rd party out of band transmitters. What will almost
certainly be more of a problem is the de-sense caused by the in-band
transmissions between the voice and data radios. Filtering that out is
not a trivial problem, at least at the spacings available at 2m.

Jeff



I have been keeping up with the thread, Jeff. And he was asking about
using his diplexers. It was others who though adding filters would be
helpful. I have never suggested this setup.

But what is the suggestion you made? And how is it different than the
one I made? And exactly what is wrong with my suggestion?

And he was not asking about desense caused by using different radios and
antennas on the same band. Tom understands this.

But it seems some people keep wanting to to keep going off-topic with
extraneous information not related to the original question.

--
==================
Remove the "x" from my email address
Jerry, AI0K

==================

On 3/19/2015 12:41 PM, Jeff wrote:

I have been keeping up with the thread, Jeff. And he was asking about
using his diplexers. It was others who though adding filters would be
helpful. I have never suggested this setup.


You obviously have not, because this is what the OP asked:

"Can I anticipate any ill affect from placing a coaxial Tee fitting
across the output of one filter and connecting a jumper to the output of
another filter in order to connect the transceivers to the multiband
antenna? Would there be any additional concerns be raised if one of the
two filters was the dual band model with separate connections for each
of it's two bands on the end being used for input connections? What
about if I make up the assembly using Three separate filters with Two of
them connected to TEE fittings"

That specifically asks about filters not diplexers.

Jeff


In this case the "filter" he is referring to is one section of the
diplexer. Not necessarily the most technical description - but also not
inaccurate.

--
==================
Remove the "x" from my email address
Jerry, AI0K

==================

On Wednesday, March 18, 2015 at 1:32:51 PM UTC-4, Jeff wrote:
On 18/03/2015 13:54, Tom W3TDH wrote:
On Wednesday, March 18, 2015 at 8:13:58 AM UTC-4, Jeff wrote:
On 18/03/2015 10:49, Brian Morrison wrote:
On Wed, 18 Mar 2015 06:44:55 -0400
Jerry Stuckle wrote:

I would not expect matching a 50 ohm antenna to a good quality filter
(which these are) would affect their response significantly enough to
cause a problem. After all, this is the exact type of operation they
are designed for.

The stop band performance is dependent on the antenna and filter
impedances out of band, at certain phases this can lead to a
re-entering stop band. I've been dealing with the consequences of this
in my day job very recently, it is a real and significant effect in
some circumstances. It's also not trivial to fix.



Indeed Brian, also if you look at the DCI filters there are quite
large returns in the stop bands.

Jeff

What are the effects of this on my proposed operation?


The effects are that it is not a good idea to parallel filters using
coax and tee pieces.

Jeff

Jeff

Are you deliberately trying to be obscure? "It's a bad idea" is not an effect but rather a rather vaguely stated opinion. Would you please advise what the deleterious effects of connecting the filters in that way would be.

--
Tom Horne W3TDH