Summary: Affordable/simple audio hardware for psychoacoustic experiments

[Daniel Oberfeld <oberfeld@xxxxxxxxxxxx>]

Dear list,

some weeks ago, I asked if someone could recommend hardware for running 
psychoacoustic experiments like the measurement of absolute and 
discrimination thresholds in undergraduate labs in Experimental Psychology.

Now hereâs a summary of the replies I received from the list. Thanks a 
lot to all who replied!


To first give you some more details concerning my question, my idea was 
to use a low-noise audio interface with output power sufficient to drive 
headphones, so that low-impendance phones like the Sennheiser HDA200 we 
use because of its good sound insulation can be connected directly to 
the audio interface. Especially for an undergraduate lab, this solution 
would have the advantage of having as few knobs as possible, and thus to 
avoid the students being tempted to change the settings...
Alternatively, a low-noise headphone amplifier could be combined with an 
audio interface, at the expense of additional cabling and knobs/switches 
that can loose their correct settings.


Volker Hohmann recommended the Edirol UA-25 two-channel bus-powered USB 
audio interface. It includes a headphone output with an analog level 
control. According to the manufacturers specs, the noise is -96 dBu 
(A-weighted, presumably at the line and not the headphones output), and 
the dynamic range is 105 dB. Additionally, Volker mentioned the 
sound-insulated headphones VIC Firth SIH 1 
(http://www.thomann.de/de/vic_firth_sih1_gehoerschutz.htm) and the 
headphone amplifier Behringer HA 4400 (now Behringer sells the HA 4700, 
see 
http://www.thomann.de/de/behringer_ha4700_powerplay_proxl_kopfhoererverstaerker.htm; 
noise < 90 dBu, dynamic range 110 dB(A) according to Behringer). Leon 
van Noorden also uses a Behringer HA 4600 amp and reports good sound 
quality.

Dan Freed had an interesting suggestion: He uses a combination of USB 
soundcard and heaphone amp (HeadRoom BitHead, 
http://www.headphone.com/products/headphone-amps/the-mobile-line/headroom-total-bithead.php) 
and said that the system sounds pretty clean even in bus-powered mode. 
No specs are available.

Piotr Majdak recommended the cheap (45 Euros) Millenium HP1 amp 
(http://www.thomann.de/at/millenium_hp1_headphone_amp.htm; no specs 
available). He uses it in a sound-insulated room and it produces no 
audible noise in HDA200 phones. He solved the problem with the analog 
volume knob simply by fixating it with some glue.

Finally, Ralph Muehleisen suggested to check out the four-channel 
Audiotrak Maya44 (http://www.esi-pro.com/viewProduct.php?pid=62), which 
according to the manufacturer features a high-fidelity headphone 
amplifier and has a dynamic range of 108 dB.



I did some own research on the internet, concentrating on external (USB 
or Firewire) audio interfaces with more than two channels (because I 
often present visual signals synchronized with audio simply by 
connecting low-power LEDs to the additional outputs).

The best (but unfortunately not really affordable) solution seems to be 
the RME Fireface 400 (8 channels I/O, ADAT, etc.). It has the unique 
advantage of featuring two outputs designed for driving headphones but 
at the same time providing the same dynamic range as the other outputs. 
Additionally, the output level can be software-configured for 3, 13, or 
19 dBu, thus providing additional dynamic range, and there are no analog 
volume controls, so that the unit seems âstudent proofâ. Measurements by 
Anselm Goertz published in the german Sound & Recording magazine showed 
a noise level of â93 dBu (unweighted) at maximum output gain and a 
dynamic range of 112 dB (unweighted). So RME once again offers a very 
useful device for psychoacoustics (quite a few people on this list use 
the excellent RME ADI 8-Pro AD/DA converter).

The following devices also seem potentially suitable for psychoacoustics:

* ECHO AUDIOFIRE 4 (4-channel firewire interface). It features 4 
balanced analog outputs, with the nominal output level software 
configurable to +4dBu or -10dBV (no analog controls!), and additionally 
a stereo headphone output with (analog) volume control. 114dB 
(A-weighted) dynamic range according to the manufacturer.

* Lake People G93 headphone amp (http://www.lake-people.com). 
Manufacturer says Dynamic range: > 125 dB (A-wtd), Noise: < -98 dBu 
(A-wtd), THD+N: < 0.002% (800mW / 100R). It can be adapted to different 
headphone impendances and efficiency factors.


To summarize, some of the devices mentioned above could even represent 
an alternative to specialized setups like the TDT system. It would 
fantastic if an evaluation of these systems with a suitable audio 
analyzer (plus maybe psychoacoustic data) became available. So if 
someone here is interested and has access to audio measurement 
facilities, just send me an email.

Daniel


--
Dr. Daniel Oberfeld-Twistel
Johannes Gutenberg - Universitaet Mainz
Department of Psychology
Experimental Psychology

Staudingerweg 9
55128 Mainz
Germany

Phone ++49 (0) 6131 39 22423
Fax   ++49 (0) 6131 39 22480
http://www.staff.uni-mainz.de/oberfeld/