Flatpack Duplexer Fun

quote"I should add that if I increase the split, then things
look better (although not great). If I go with a 7Mhz split, I
can get the insertion loss down to about 2dB and the VSWR
looks good.

I do a flatpack duplexer pre-alignment... With an example 452
& 457 MHz pair, (5 MHz offset) targeted. I might move all the
low notches all the way down, well below 450 and the converse
of moving all the 457 notches up well in to the high 460 segment.
This type of approach can help to minimize miss-adjusted interaction
of excessively (initial set) close spaced cavities or cells. You're
not tuning the notch cavities to any specific frequency well below
and/or above the desired new location, you're just getting them
out of the way to a known visually (test equipment) confirmed starting
point. Once you have the known starting point, each side can then
be "brought back" and aligned at the new, desired location.

The only other thing I haven't tried yet, is to change the High
and low side. Maybe they miss labeled at factory? I don't
see how that will make a different since these cans should
be symmetrical.

If the flat-pack was labeled for repeater service, the high and low
side will be different from the same duplexer marked for duplex
base-station and/or mobile telephone type of service. Trust no
labels and wash you feet at least twice a month.

The flat-pack duplexer might not be symmetrical and I wouldn't
expect it to be for reasons I right now care to deal with. In
theory, the inter-connecting coax lines and the internal probe
makeup will normally not be the same length.

Not to go off on a tangent, but Wow, the email coming
into yahoo groups looks like its out of sequence.

"Yahoo is now part of Oaf", I mean "Oath" the Verizon Group...
it's time to see how a large cell carrier can further ruin Yahoo
while trying to extract your personal information for sale.

This above email was from earlier today before I did
the low/high swap and test. -Jeremy

My last RB Group post took more than than 24 hours to
show up in the listings.

cheers,

Pearl Jam

Jeff you are 100% correct. They are not symmetrical, I don't know what I was thinking since the frequency response swings up or down based on the side of the duplexer.

Here is a snapshot of the response when I put things back on the SA/TG. You'll see the insertion loss...

If I push the split up to 7Mhz, then its not bad. Its kind of like they designed this for a wider split but sold it as a 5 Mhz split.

quote"I should add that if I increase the split, then things look better (although not great). If I go with a 7Mhz split, I can get the insertion loss down to about 2dB and the VSWR looks good.

The only other thing I haven't tried yet, is to change the High and low side. Maybe they miss labeled at factory? I don't see how that will make a different since these cans should be symmetrical.

-J"



Not to go off on a tangent, but Wow, the email coming into yahoo groups looks like its out of sequence. This above? email was from earlier today before I did the low/high swap and test.

-Jeremy

would anyone be able to tell me what brand of duplexer this is?
Thanks John
N4GRW

Decibel DB4068 or DB4072 depending on which bandsplit it is.

--- Jeff WN3A

The only other thing I haven't tried yet, is to change the
High and low side. Maybe they miss labeled at factory? I
don't see how that will make a different since these cans
should be symmetrical.

If they're anything like any other non-Chinese flat pack UHF notch duplexer
I've ever worked on, they are not symmetrical...

--- Jeff WN3A

would anyone be able to tell me what brand of duplexer this is??

开云体育

Thought of what?
Inquiring minds want to know why you did not quote enough of the message your are replying to so that you
can make sense of your reply.
Yahoo does not deliver messages in the order received.

73
Glenn
WB4UIV

Thanks - I should have thought of that.

Tom?

-- 
-----------------------------------------------------------------------
Glenn Little                ARRL Technical Specialist   QCWA  LM 28417
Amateur Callsign:  WB4UIV            wb4uiv@...    AMSAT LM 2178
QTH:  Goose Creek, SC USA (EM92xx)  USSVI LM   NRA LM   SBE ARRL TAPR
"It is not the class of license that the Amateur holds but the class
of the Amateur that holds the license" 

So I double checked, and I should have thought about this before, but the frequency response looks ok for which side is high and which side is low given the roll off of the 'pass' frequency. So the label with the frequencies was indeed incorrect, but the 'low' / 'high' label was ok, which is what I was using for the 'pass' reference.

Note the attachment of the 'low' side. Notice how much the notch is encroaching on my pass frequency.

If I go to 7Mhz split, then its not as bad. Its like this notch filter was not designed for this kind of split... yet it was sold for 5Mhz split.

-J

The LOW side would be tuned to reject the higher frequency and corresponds to the label IF the notch was tuned to the 467 frequency. The HIGH side would reject the 462 frequency.

The split is the same as factory...just a few Khz down.

You sure the high and low sides didn't get swapped? The labels and the
frequencies don't match - high is low, and low is high, in the pic you
posted.

--- Jeff WN3A

Looking at the photo, I think the sides are reversed. Compare the frequencies to the "High" and "Low" labels.

Could the high side and low side be reversed? Maybe it came wrong from China.

I should add that if I increase the split, then things look better (although not great). If I go with a 7Mhz split, I can get the insertion loss down to about 2dB and the VSWR looks good.

The only other thing I haven't tried yet, is to change the High and low side. Maybe they miss labeled at factory? I don't see how that will make a different since these cans should be symmetrical.

-J

Folks,

I have an old eBay Chinese flat pack that I put on the bench to tune for a temporary use repeater. I can notch it down to 90dB just fine. But the insertion loss is horrible. I'm getting around 5dB of loss. I thought something was wrong with the SA/Tracking Generator and plugged it in to the service monitor to test before/after power output and sure enough, 5.4dB insertion loss. SWR is horrible for the actual pass frequency.

I don't think this got dropped (at least not that I know of). I hate to bash a cheap mobile duplexer from China, but I wondering if this has been other's experience of these knock off brand mobile duplexers?

The split is the same as factory...just a few Khz down.

Both sides are within a dB of each other.

-J

开云体育

开云体育

开云体育

Yes, I believe there is.? I think you can select "CSQ", or something similar... but it's been so long since I programmed mine, I seriously can't remember where the selection is.