UCD400HG - lowering gain

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi

I have 3 of these with their SMPS400-64 just arrived, my first go at diy class d amping, upgrading my modded NVA A80s (to be used for rears)/Wilson Benesch Arcs, and a velleman K4004B/diy Audiom 15 VX based sub, with years of problem free running. Output from a Chord Dac64, no preamp, using a Meridian 518 as digital vol/imput switcher. It is preferable to be driving the amps at near unity as digital volume controls work best when not doing too much attenuation (even though the 518 does some very clever noise shaped dither work). So looking at the UCD400's there's some documentation on the differential imput gain being controlled by one resistor for 26db gain, R3. Any ideas anyone on changing this to around 2k (around 20db gain) to enable higher line level imputs? I've tried contacting Hypex a couple of times without reply...any thoughts?
 
I have already tried answering your email a couple of times, though there is some delivery issue and they keep bouncing.

Practically all the info you would need on this subject can be found on our website. In the document you can find how to calculate the correct resistor value for the desired gain. But also what the drawback is of lowering the gain, the CMRR will drop by the same amount as the gain is lowered. Other than that there are no real drawbacks, so you're free to experiment with different gain settings.

Please do keep in mind that making modifications to the modules will affect warranty.

The document can be found here:
http://www.hypex.nl/docs/gainappnote.pdf

Frank Veldman
Hypex Electronics
 
Many thanks, yes I did look through the various documents you have issued, I just wanted a general feel that I wasn't going to run into any other issues by effectively altering your designs, exactly as you have in your reply there thanks. The resistors are now in with a very steady hand..I think it should do the job nicely. Not sure what the detectable sonic implications of higher common mode signals being passed through are...maybe a little more noise? I am running balanced outputs from the DAC so will see. R3 is now 2k which should lower overall gain to 20db from 26db I calculate.
 
FIRST TEST....the gain is bang on where I want it, 75 on the Meridian 518 digital volume 'guage' (88 is unity, 88-100 digital gain but very likely to clip) is very loud..pretty much exactly the levels of the NVA A80's but smoother sounding with less crowding - and the NVAs were stunning as they were. I think running the UCD400HG's like this leaves masses of headroom...I suspect at the same listening volume volume there's far lower THD produced. The sound is stunning, and with no running in straight from the box! Now to set up the third UCD amp for the sub...
 
the third amp for the sub, a wall mounted Audiom 15 VX (which needs a new rubber surround!) was better with standard gain/r3 after trying the lowered gain version. The filtered bass comes from a Meridian 568 which provides very fine tuning and doubles as a very fine cinema dts processor for dvd/bd or the tv spdif optical, and a digital switcher with the sub out used as conventional lfe. For stereo 44.1/16 using the DAC64, additional analogue phono outs feed the 568 (as an ADC then dsp) from 42hz down for the sub filter. You have to manually adjust the volume settings for either stereo sub or DTS LFE and select input for the 568, in stereo filtered sub mode, the sub output from the 568 is quite low so the UCDs gain needs to be as standard (very high for DTS output though, and the LFE adjustment on the 568 has to be set -10 db or so, hence the new rubber surround!). The huge advantage of digital filters/crossovers is no phase variations over the roll over point, and it shows (as in Meridian's DSP speakers - all digital xovers).
It amazes me how much detail there is in the lower octaves that this set up reveals - a very visceral experience, the UCD driving the audiom with no signs of strain.
 
Hi guys,
searched all over the place (and six feet under with such old thread...) but can't find a clear answer.
What is clear is that if I lower the gain (using Rg 3.3kΩ with Rf 2.2kΩ, 0102 MELF thin metal film) by say 6dB then I lower CMRR by 6dB, from 75 to 69dB. These are calculations, nothing else right now.
But what in real life that CMRR will do? More hiss (there's none!) at speakers? More fridge and light/appliances switch noises getting in? I'm only willing to get more play at pots and less digital attenuation at my ES9018 balanced DAC, which I admit will sound better... maybe only on paper...
BTW, only one resistor to order/solder... worth it? Yes, I am lazy but feedback from the real world is welcome before doing things for nothing. Or not.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.