Re: Signal Generator Source Impedance

On Tue, 30 Apr 2024 16:23:31 +0100, Ian Jackson
<ianREMOVETHISjackson@g3ohx.co.uk> wrote:

In message <4k2t2jdaq5jotm57mvss543qmsk060stnq@4ax.com>, Jeff Liebermann
<jeffl@cruzio.com> writes

On Sun, 28 Apr 2024 18:23:13 +0100, Cursitor Doom <cd@notformail.com>
wrote:
>

Gentlemen,

>
Nobody is gentle on Usenet.
>

The service manual for this radio I'm re-aligning states to use an RF
signal generator with a source impedance of 60 ohms.  I have never
heard of such a beast. I have a 50 ohm RF sig gen and a 70 ohm one. I
thought that would be all I'd ever need, but now this spanner gets
thrown into the works. I'm not having a good day so far!
Do I need to make up a suitable impedance transformer, or can I
proceed with the 50 or the 70 ohm ones I have, whilst making some
allowance for that (and if so, how should that allowance be made?)
>
CD.

>
There actually is such a generator:
<https://fairradio.com/product/sg-376/>
"1-1.25 V into 60 ohms (IF) with attenuation to 100 db in 1 or 10 db
steps."
>
Such generators appeared when the manufacturer wanted to produce a
generator that worked reasonably well with both 50 and 75 ohms.
Hint:  (50 + 75) / 2 = 62.5 ohms
My guess(tm) is that users were losing (or stealing) the minimum loss
pads required to go from 50 to 75 ohms with one front panel connector:
<https://www.analog.com/en/resources/technical-articles/catv-minimum-los
s-pad-for-75937-measurements.html>
The mismatch loss from 50 to 75 ohms is only 0.4dB, so you're not
likely to see much of an effect from using the "wrong" impedance
unless you're working with microwave frequencies.
>
A much bigger problem is that 50 and 75 ohm BNC connectors use
different center pin sizes.
<https://www.vintage-radio.net/forum/attachment.php?attachmentid=30120&d
=1258015846>
I wrecked the 75 ohm receptacle on a CATV sweep generator by force
fitting a 50 ohm BNC plug.  Color coding my cables have largely
prevent a recurrence.  Green tape or nail polish for 50 ohms and
violet for 75 ohms.
>

Despite over 40 years in the CATV industry, I've never seen any 50 or 75
ohm BNCs that weren't mechanically mateable. The mating parts of the
metalwork are usually pretty-well identical.
>
OK, with some 50 versions the point of the male pin might be just a
little bit blunter, but there's little chance of it splaying the female
receptacle. Otherwise, the diameter of the pin is the same. But note
that with the 75 ohm the hole at the rear end of the pin and receptacle
might be a rather tight fit for the inner of 50 ohm coax.
>
What makes one connector 50 ohms and the other 75 is that the amount of
the PTFE dielectric is considerably cut back in the 75 (making it as
air-spaced as possible). IIRC, the 50 is the more constant-impedance,
which makes it more-usable to higher frequencies).
>
On the other hand, the N-connector is a very different beast. The pin of
the 50 ohm has a much greater diameter than the 75 (which may be what
the photo shows). If you insert a 50 ohm male into a 75 ohm female, you
will almost certainly cause irreparable damage to the female (in some
companies I believe it's still a hanging offence), while a 75 male into
a 50 ohm female won't male contact.
>
Finally, the inner parts of a female BNC and an N are very similar, and
you can usually mate either with a 75 ohm N (but not with a 50, as the
pin diameter is too large).

>
>

Same danger with SMA and 3.5mm connectors. At first glance they appear
the same, but they're not - and mating them can cause *very* expensive
damage if it's a high-end RF cable.