From: Alun Palmer
Subject: Re: How to check for a 50 or 75 ohm connector
Date: 23 Dec 2002 07:48:51 GMT
References: <email@example.com> <firstname.lastname@example.org>
NNTP-Posting-Host: host220.fastdial.net (18.104.22.168)
DarkMatter wrote in
> On 22 Dec 2002 21:05:18 GMT, Alun Palmer Gave us:
>>The centre pins on 75 Ohm BNC plugs are very slightly larger, so that
>>inserting them in a 50 Ohm BNC socket causes damage by enlarging the
>>hole. I used to work in a lab where we had both.
> Absolutely incorrect.
I speak from actual observation, not from reading a spec sheet. If this
means that the 75 Ohm types didn't meet the spec, then that is a slightly
>>I'm not sure how F connectors got into this thread, but they are nasty
>>rubbish, worse even than Belling-Lee connectors. Only use F connectors
>>if you absolutely have to. They make UHF connectors look like an
>>engineering masterpiece by comparison, which they're not.
> Whatever. Your opinions are probably as good as your
> mis"information" is. Which is nil.
F connectors are cheap mass-market junk.
> The center pins are the SAME
> diameter. The connector you refer to as causing damage is NOT of the
> BNC type.
Technically it may not be, but it looks enough like a BNC to be a problem
> In order to be called BNC legally
Lawyers don't attach much meaning to the word 'legally' - what do you mean
by it here?
> , it has to be able to
> plug into a BNC connector jack. Period.
They did plug in and work, but...
> And that would be ANY BNC
> jack, and that would be without subsequent damage.
You can't change the impedance and keep the same dimensions. If there is a
standard, I'm assuming you're saying it's 50 Ohm only, and therefore the
75 Ohm version isn't a BNC. The trouble is it looks like one, and it
damages the 50 Ohm socket. What is the standard number, then?
>>F connectors leak RF in and out like a sieve. One of my old colleagues
>>did extensive tests of 'transfer impedance' (this has nothing to do
>>with characteristic impedance, but is a measure of screening) and could
>>find nothing under the sun that had worse screening than an F
> The shield level of the cable used is a determining factor in that
> test as well.
Not if the test is done properly. It is possible to get independent
numbers for connectors and cables. Since the F doesn't have a centre pin,
though, I'll admit that does present a unique problem.
> Also, the assembler has to know what the hell is going on in order
> to fashion the things right as well.
He had a PhD, so he wasn't exactly a dummy. He designed his own jigs for
measuring transfer impedance, which gave repeatable figures up to 2 GHz,
as far as I recall. His last name was Smithers, and I am pretty sure his
work was published. For the record, I am an electrical engineer (only a
BSc, not PhD).