On 05/03/2015 14:01, Jim Lesurf wrote:


IIRC you're using one of the UCA series - i.e. in the same series as the
one I tested and reported on at

http://www.audiomisc.co.uk/HFN/ADC/USBrecording.html


I'm using a Behringer UFO202. Looks almost identical to the ones in your
review.

but I can't recall if you're using one that includes the RIAA.


Yes, I am.

If you're not already doing so, you may get better results by using a
dedicated RIAA preamp and having longer coax leads from its output to the
Behringer. This would keep the fields from the Behringer and computer
further away from the low level signals from cartridge to RIAA preamp.

A problem with using the RIAA in the Behringer is that you'd have to use
longer signal leads at low level to move the unit away, and so that may not
help - or make things worse! Better to keep low-level signal leads short,
the snag being if that also moves the source of interference closer! catch
22.


That was something I hadn't considered when getting the UFO 202 but moot
point for now, see below.

FWIW I did some more recordings here this morning so can give some values
for reference. Here I get about -72 to -74 dB hum+noise before the needle
drops onto the LP. This is with the 0dB RIAA reference level at about
-15dB. So the nominal level of hum+noise is in the ballpark of -60
dB(RIAA).
However the hum is almost pure 50Hz and some of the value is
noise over a wide band as I'm using a 96k sample rate. In practice, the
noise goes up as soon as the needle finds the groove. Dominated by LF
rumbles for the best LPs as the system goes down to below 10Hz.

Here this is with the 'PC' next to the deck and the ADC just under it. But
the 'PC's power supply is a brick down on the floor.

FWIW The RIAA preamp I use does have a 10 Ohm lift resistor in the
grounding. I found this made a measureable improvement in hum in most
circumstances of normal use.


I tried a 10 Ohm resistor and it did give a small reduction of around 1
dB according to Audacity, so I will keep it. Every bit helps.

I also tried an (enormous 500VA) isolating transformer in the turntable
mains feed and this reduced the hum by quite a large amount.

Total reduction in hum from the 3 changes: ground lift, shielded
isolation transformer and electrically shortened turntable leads amounts
to around 10 dB overall reduction of the 50Hz component. The isolation
transformer contributed much larger reductions of the more troublesome
higher order harmonics and general hash. Some were eliminated entirely.
Compare the original he

http://www.nu-ware.com/Misc/Screenshots/Dual505A.png

With this:

http://www.nu-ware.com/Misc/Screenshots/Dual505B.png


FWIW2 I've just been doing some theory estimates of thermal noise for other
reasons. (Comparing MM with MC as a result of a daft assertion in a
magazine! Old campaigners can probably guess the source given that the
assertion showed 'innocence' of the physics of thermal noise. :-) ) So in
context this is an interesting comparison between ideal world and reality.


Reality wins out, every time :-)

My last tussle with thermal noise was modification of a ccd (not cmos)
camera for astronomical use. Peltier chips for refrigeration of the ccd,
image stacking to average out thermal noise etc. etc.

I'm happy with the Dual 505 hum level for now and will be moving on to a
workout of my test LP. No doubt I will be pestering you all with more
questions. Thank you so much all for your ideas and comments so far,
Jim, Java, Trevor, Mick.