" There's no mechanism within any common type of RF connector which causes
any significant amount of loss. Various types can cause some impedance
mismatch, but not enough to cause a noticeable reduction of signal to the
receiver.


Since the OP was asking about frequencies up to 1GHz, I would be more
cautious about losses in connectors. Poorly assembled connectors can show
significant losses at 1GHz. Also I have come across connectors and adaptors
(SMA) that show loss for no apparent reason at a Gig or so., now whether
this is due to dielectric losses or some other factor I am not sure but
mechanically they appeared to be prefect. One particular SMA barrel had a
stable and repeatable loss of over 2dB, but was clean and gauged OK.

Some of the cheap PL259 connectors also use very poor dielectric which show
significant loss at even VHF.

Regards
Jeff

Jeff wrote:
" There's no mechanism within any common type of RF connector which causes
any significant amount of loss. Various types can cause some impedance
mismatch, but not enough to cause a noticeable reduction of signal to the
receiver.


Since the OP was asking about frequencies up to 1GHz, I would be more
cautious about losses in connectors. Poorly assembled connectors can show
significant losses at 1GHz. Also I have come across connectors and adaptors
(SMA) that show loss for no apparent reason at a Gig or so., now whether
this is due to dielectric losses or some other factor I am not sure but
mechanically they appeared to be prefect. One particular SMA barrel had a
stable and repeatable loss of over 2dB, but was clean and gauged OK.

Some of the cheap PL259 connectors also use very poor dielectric which show
significant loss at even VHF.

Regards
Jeff

I'm surprised at these reports, but not too much except for the 2 dB
connector. It must have had some internal corrosion, or perhaps
something like a helical internal contact causing an impedance mismatch
that would look like loss in a matched measurement system. (Some el
crappo UHF barrel adapters have found to be constructed that way.) I've
seen some awfully shoddy BNC connectors at ham conventions and wouldn't
use them for anything at all critical. But the OP asked if there was a
place he could find the losses of various kinds of connectors. I don't
believe there's any place you can go to find the losses of shoddily made
or improperly assembled connectors or adapters. Decently made and
properly assembled connectors of standard types show no appreciable loss
at 1 GHz. If loss is a concern, my advice is to buy name brand
connectors, and especially avoid the cheap stuff you see at places like
flea markets.

Roy Lewallen, W7EL

In message , Roy Lewallen
writes
Jeff wrote:
" There's no mechanism within any common type of RF connector which causes
any significant amount of loss. Various types can cause some
impedance mismatch, but not enough to cause a noticeable reduction
of signal to the receiver.
Since the OP was asking about frequencies up to 1GHz, I would be
more cautious about losses in connectors. Poorly assembled connectors
can show significant losses at 1GHz. Also I have come across
connectors and adaptors (SMA) that show loss for no apparent reason
at a Gig or so., now whether this is due to dielectric losses or some
other factor I am not sure but mechanically they appeared to be
prefect. One particular SMA barrel had a stable and repeatable loss
of over 2dB, but was clean and gauged OK.
Some of the cheap PL259 connectors also use very poor dielectric
which show significant loss at even VHF.
Regards
Jeff

I'm surprised at these reports, but not too much except for the 2 dB
connector. It must have had some internal corrosion, or perhaps
something like a helical internal contact causing an impedance mismatch
that would look like loss in a matched measurement system. (Some el
crappo UHF barrel adapters have found to be constructed that way.) I've
seen some awfully shoddy BNC connectors at ham conventions and wouldn't
use them for anything at all critical. But the OP asked if there was a
place he could find the losses of various kinds of connectors. I don't
believe there's any place you can go to find the losses of shoddily
made or improperly assembled connectors or adapters. Decently made and
properly assembled connectors of standard types show no appreciable
loss at 1 GHz. If loss is a concern, my advice is to buy name brand
connectors, and especially avoid the cheap stuff you see at places like
flea markets.

Roy Lewallen, W7EL


Even at 1GHz, I don't believe that you would get as much as 2dB of loss
even if you joined two pieces of coax by stripping the ends with a
knife, twisting the conductors together, and wrapping them in scotch
tape.
--
Ian

I'm surprised at these reports, but not too much except for the 2 dB
connector. It must have had some internal corrosion, or perhaps something
like a helical internal contact causing an impedance mismatch that would
look like loss in a matched measurement system. (Some el crappo UHF barrel
adapters have found to be constructed that way.) I've seen some awfully
shoddy BNC connectors at ham conventions and wouldn't use them for
anything at all critical. But the OP asked if there was a place he could
find the losses of various kinds of connectors. I don't believe there's
any place you can go to find the losses of shoddily made or improperly
assembled connectors or adapters. Decently made and properly assembled
connectors of standard types show no appreciable loss at 1 GHz. If loss is
a concern, my advice is to buy name brand connectors, and especially avoid
the cheap stuff you see at places like flea markets.

Roy Lewallen, W7EL


Even at 1GHz, I don't believe that you would get as much as 2dB of loss
even if you joined two pieces of coax by stripping the ends with a knife,
twisting the conductors together, and wrapping them in scotch tape.
--
Ian

The items in question were SMA barrel adaptors and as I said showed no signs
of damage, corrosion or spreading of the contacts and interface dimensions
were in spec

It is surprisingly easy to get significant loss at 1GHz and above, 2dB would
be very easy to achieve even when using what appears to be a 'good'
connection.

regards
Jeff

In message , Jeff
writes
I'm surprised at these reports, but not too much except for the 2 dB
connector. It must have had some internal corrosion, or perhaps something
like a helical internal contact causing an impedance mismatch that would
look like loss in a matched measurement system. (Some el crappo UHF barrel
adapters have found to be constructed that way.) I've seen some awfully
shoddy BNC connectors at ham conventions and wouldn't use them for
anything at all critical. But the OP asked if there was a place he could
find the losses of various kinds of connectors. I don't believe there's
any place you can go to find the losses of shoddily made or improperly
assembled connectors or adapters. Decently made and properly assembled
connectors of standard types show no appreciable loss at 1 GHz. If loss is
a concern, my advice is to buy name brand connectors, and especially avoid
the cheap stuff you see at places like flea markets.

Roy Lewallen, W7EL


Even at 1GHz, I don't believe that you would get as much as 2dB of loss
even if you joined two pieces of coax by stripping the ends with a knife,
twisting the conductors together, and wrapping them in scotch tape.
--
Ian

The items in question were SMA barrel adaptors and as I said showed no signs
of damage, corrosion or spreading of the contacts and interface dimensions
were in spec

It is surprisingly easy to get significant loss at 1GHz and above, 2dB would
be very easy to achieve even when using what appears to be a 'good'
connection.

regards
Jeff

I have to admit that I've routinely worked to 870MHz, and recently even
to 1003MHz, but rarely ventured higher. However, I've never found any
excessive loss in a connector which could not be accounted for by a
straightforward 'bad connection', ie usually a total or partial open or
short circuit, whether in the making off of the cable, or in the
construction of the connector itself. Honest!
--
Ian.

Ian Jackson wrote in
news
In message , Jeff
writes
I'm surprised at these reports, but not too much except for the 2 dB
connector. It must have had some internal corrosion, or perhaps
something like a helical internal contact causing an impedance
mismatch that would look like loss in a matched measurement system.
(Some el crappo UHF barrel adapters have found to be constructed
that way.) I've seen some awfully shoddy BNC connectors at ham
conventions and wouldn't use them for anything at all critical. But
the OP asked if there was a place he could find the losses of
various kinds of connectors. I don't believe there's any place you
can go to find the losses of shoddily made or improperly assembled
connectors or adapters. Decently made and properly assembled
connectors of standard types show no appreciable loss at 1 GHz. If
loss is a concern, my advice is to buy name brand connectors, and
especially avoid the cheap stuff you see at places like flea
markets.

Roy Lewallen, W7EL

snip
I have to admit that I've routinely worked to 870MHz, and recently
even to 1003MHz, but rarely ventured higher. However, I've never found
any excessive loss in a connector which could not be accounted for by
a straightforward 'bad connection', ie usually a total or partial open
or short circuit, whether in the making off of the cable, or in the
construction of the connector itself. Honest!


One thing, SMA connectors are rated for 500 connection cycles, I don’t know
what happens after that, but on test equipment 500 connections might happen
fairly fast. I know from experience that SMA is a connector that’s easy to
put on badly.

John Passaneau

On Tue, 11 Mar 2008 16:00:11 +0000, Ian Jackson
wrote:

Even at 1GHz, I don't believe that you would get as much as 2dB of loss
even if you joined two pieces of coax by stripping the ends with a knife,
twisting the conductors together, and wrapping them in scotch tape.
--
Ian

It is surprisingly easy to get significant loss at 1GHz and above, 2dB would
be very easy to achieve even when using what appears to be a 'good'
connection.

regards
Jeff

I have to admit that I've routinely worked to 870MHz, and recently even
to 1003MHz, but rarely ventured higher. However, I've never found any
excessive loss in a connector which could not be accounted for by a
straightforward 'bad connection', ie usually a total or partial open or
short circuit, whether in the making off of the cable, or in the
construction of the connector itself. Honest!

Hi All,

Anyone can do it wrong (come up with 2dB loss), but we would have a
great more deal traffic here if doing it wrong was that common.

I have measured attenuation of near everything from DC to 12GHz at the
standards bench and RG58 with BNCs for a short run (2 to 3 meters)
never presented any problems being wildly imagined here. At the
worst, the connectors (note plural) "might" show 0.2dB mismatch loss -
trivial.

On the other hand, if Jeff is so sure of this 2dB figure, it should
reveal itself in blisters to the fingers for even mild power (100W).
Since this unequivocal evidence is so obviously missing from the
records (or the testimony, as the case may be); then it is very
apparently one of those "someone heard from someone else about their
brother's friend's boss who had a customer who made this claim."

73's
Richard Clark, KB7QHC

Anyone can do it wrong (come up with 2dB loss), but we would have a
great more deal traffic here if doing it wrong was that common.

I have measured attenuation of near everything from DC to 12GHz at the
standards bench and RG58 with BNCs for a short run (2 to 3 meters)
never presented any problems being wildly imagined here. At the
worst, the connectors (note plural) "might" show 0.2dB mismatch loss -
trivial.

On the other hand, if Jeff is so sure of this 2dB figure, it should
reveal itself in blisters to the fingers for even mild power (100W).
Since this unequivocal evidence is so obviously missing from the
records (or the testimony, as the case may be); then it is very
apparently one of those "someone heard from someone else about their
brother's friend's boss who had a customer who made this claim."

73's
Richard Clark, KB7QHC

Well let me put you right Richard, there was no high power, so burnt
fingers, this was noticed in a test set-up on a network analyser, and
subsequently the adaptors in question were tested in isolation. 'Good'
adaptors were also substituted and their loss was measured and was very low
indeed.

The problem was first noticed by a very senior consultant in the lab in
question and then checked by several other people because they could not
believe it either. The connector interfaces were checked and were Ok. No
further time was spent investigating why the problem occurred, just 2 SMA
adaptors in the bin and the rest of the adaptors in the lab checked.

So please don't denigrate other people's comment without asking about the
facts.

My comments were not about properly made connectors, that will have very low
loss, but were a caution that at high UHF and above care has to be taken and
it is very easy to introduce loss. This is particularly the case when
launching from connectors onto pc boards with microstrip or joining on to
coax cable.

Having been involved professionally for over 10 years in the design of
microwave equipment I can assure you that there is an endless list of the
ways that loss can creep into a system at high frequencies.

73
Jeff

Richard Clark wrote:
On Tue, 11 Mar 2008 16:00:11 +0000, Ian Jackson
wrote:


Even at 1GHz, I don't believe that you would get as much as 2dB of loss
even if you joined two pieces of coax by stripping the ends with a knife,
twisting the conductors together, and wrapping them in scotch tape.
--
Ian


It is surprisingly easy to get significant loss at 1GHz and above, 2dB would
be very easy to achieve even when using what appears to be a 'good'
connection.

regards
Jeff


Hi All,

Anyone can do it wrong (come up with 2dB loss), but we would have a
great more deal traffic here if doing it wrong was that common.

I have measured attenuation of near everything from DC to 12GHz at the
standards bench and RG58 with BNCs for a short run (2 to 3 meters)
never presented any problems being wildly imagined here. At the
worst, the connectors (note plural) "might" show 0.2dB mismatch loss -
trivial.

On the other hand, if Jeff is so sure of this 2dB figure, it should
reveal itself in blisters to the fingers for even mild power (100W).
Since this unequivocal evidence is so obviously missing from the
records (or the testimony, as the case may be); then it is very
apparently one of those "someone heard from someone else about their
brother's friend's boss who had a customer who made this claim."

73's
Richard Clark, KB7QHC

I've encountered (defective or broken) SMA hardware that has remarkably
high *apparent* loss. Typically, internal cracks or voids. I had a
batch of Pasternack right angle M-F elbows that had losses that were all
over the place at 1 GHz sorts of frequencies. I'm pretty sure that
internally, there was something not connected, or a spring loaded
something that wasn't making contact.

In this case, the connector wasn't actually lossy, but more of an
extreme mismatch, and when hooked up with a 50 ohm signal generator on
one side and a 50 ohm power meter on the other, it looked like loss.
Actually, it was reflecting the power back to the source.

In this scenario, with 100W, your fingers wouldn't have gotten burned
(but the signal generator would probably shut down with reverse power
warnings)