Ground issues should not be a problem. I'd say in 99% of the instances, the
output of the computer sound card is unbalanced to ground. Maybe 1% are
unbalanced above ground and there is a difference. And radios are also
unbalanced to ground. Few have a balanced input for either Line or Mike.
If, and they should, share the same neutral and the same 3rd pin ground as
the computer then there is NO difference in potential, thus no current
flowing on the shield and thus no hum or ground loop issues.
Exception to the above is when the radio has a separate external ground that
IS NOT bonded to the AC mains ground. This of course is in violation of the
NEC and could likely pose a dangerous condition. And a ground loop is most
likely to exists, that is current flowing on the ground. Remember a ground
should be at 0 volt potential and since it has some value of R, and if
current is flowing, then it is not at 0 potential. A couple of mV above
ground is all it takes to make a microphone input look as though it is
receiving some form of signal. Most of the time it is hum, buzz or hash. E
= I * R
One issue with using the sound card speaker output and trying to feed it
into a microphone input is a major difference in levels. The Omni VII
microphone spec is 1 mV for full output. As an example, the Paragon
microphone spec is 5 mV (-62 dB). The speaker output is typically a couple
of volts, depending on sound card level, while the microphone input is
looking to see a few millivolts. FYI - 1 volt = 1000 mV. Turning the
sound card level down does decrease the data signal from the soundcard but
DOES NOT decrease the noise which is largely thermal noise and hum, mostly
120 Hz , from the computer power supply even though the computer supply is a
switching supply. It is likely to have the AC line bypassed to chassis thus
the 60 Hz and related harmonics of same exists. Thus the S/N of the data
signal is poor and the noise is amplified by the high gain of the microphone
stages along with the signal. If your radio has a Line Level input, use it
and not the mike input. If your radio does not have a Line Level input then
correctly you should make a simple resistive attenuator, either a Pi, T or L
configuration, your choice, to exhibit some 20 to 30 dB of loss. Turn up
the level of the sound card thus improving S/N ratio.
As to using transformers, many of the less expensive and small transformers
have poor or rapidly declining low frequency response thus acting more like
high pass filters. This has been known to appear to solve many hum and buzz
related issues when correct application of signal levels is more in order.
If you should be so fortunate as to have a computer sound card that has both
Line Level input and output and a radio that has both Line Level input and
output, such as the Omni VII, the direct connection between the radio and
computer is all that is required.
If you want more dialogue on the subject, let's take it off line with a
E-Mail direct to me. I'll be glad to provide you with resistive numbers for
various attenuators.
Take the time to read and study the paper that Jim, K9YC, has suggested.
It is well worth the read. Better yet, print it out and read it at your
leisure.
73
Bob, K4TAX
----- Original Message -----
From: "Jim Brown K9YC" <k9yc@audiosystemsgroup.com>
To: "Barry N1EU" <barry.n1eu@gmail.com>; "Discussion of Ten-Tec Equipment"
<tentec@contesting.com>
Sent: Saturday, December 22, 2007 2:08 PM
Subject: Re: [TenTec] Fw: question Digital matching-- rig to sound card
On Sat, 22 Dec 2007 18:20:43 +0000, Barry N1EU wrote:
The xfmr isn't for impedance matching, it's for ground isolation.
Right. And tranformers should not be necessary if you follow the
simple and inexpensive techniques in my RFI tutorial -- bond all the
gear together with big copper, plug everything into the same outlet
(or outlets in the same quad box), and use cable for the
interconnects that has a beefy copper shield (like RG58).
http://audiosystemsgroup.com/RFI-Ham.pdf
73,
Jim Brown K9YC
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|