Analog Bass Management Anyone?

[davee70]

Marchand high pass filter

It turns out Marchand makes two types of 12 db/oct filters, a Butterworth and a Bessel. The Bessel has a gentler roll-off and phase shift. So, I'm inclined to go with the Bessel over the Butterworth.

 

[TLS Guy]

It turns out Marchand makes two types of 12 db/oct filters, a Butterworth and a Bessel. The Bessel has a gentler roll-off and phase shift. So, I'm inclined to go with the Bessel over the Butterworth.

I would be inclined to go with the Butterworth. The Bessel could well lead to a rise in output in the crossover region, but you might tame it by setting the low pass to a slightly lower frequency than the high pass.

 

[davee70]

I would be inclined to go with the Butterworth. The Bessel could well lead to a rise in output in the crossover region, but you might tame it by setting the low pass to a slightly lower frequency than the high pass.

I was more worried about the phase shift when I ran the calculator that's on Phil Marchand's web site. The phase shift at 80 Hz is 90 degrees with the 12 db/oct Butterworth and ~56 deg with the Bessel. That compares to 45 deg with the 6 db/oct Butterworth. I hope the phase shifts won't be too much of a problem. The digital crossover in the DD-15 allows me to adjust the slope and set the phase angle in certain increments. I guess it remains to be seen what kind of results I can achieve.

 

[TLS Guy]

I was more worried about the phase shift when I ran the calculator that's on Phil Marchand's web site. The phase shift at 80 Hz is 90 degrees with the 12 db/oct Butterworth and ~56 deg with the Bessel. That compares to 45 deg with the 6 db/oct Butterworth. I hope the phase shifts won't be too much of a problem. The digital crossover in the DD-15 allows me to adjust the slope and set the phase angle in certain increments. I guess it remains to be seen what kind of results I can achieve.

You worry too much about small details.

The phase shift for the 24 db low pass filter will be 180 degrees. For the Butterworth total shift will be 270 degrees and for the Bessel 236.

Then don't forget you will have huge phase shifts due to placement issues. So you will end up setting your phase control on the sub for maximum output at crossover anyway. I can tell you it won't be the calculated value, as the whole thing is a witches brew.

 

[Pyrrho]

You can not use bass management in an Oppo player without having the player do a PCM conversion. No bass management is possible within DSD.

On the BDP-95, the Audio Format Setup allows the user to choose DSD over PCM. It says:
"For the analog audio outputs, DSD data is converted to analog signal directly by the internal DAC." (pg. 62)
...

If that is the case, then you ought to set it to convert to PCM so it can do your bass management. That is going to be better than some add on thing that will cost you money, unless Oppo designed it incompetently.

 

[davee70]

If that is the case, then you ought to set it to convert to PCM so it can do your bass management. That is going to be better than some add on thing that will cost you money, unless Oppo designed it incompetently.

Maybe that's why Oppo advises it's a matter of user preference. I've not heard DSD direct before and now that I have, that's my preference at this point.

 

[davee70]

Wrap up

Despite numerous (unrelated) trials and tribulations, I finally have everything together and working in sync on my DSD-direct external crossover workaround. One key was to add a pair of summing amps (ATI SUM100) to combine the L/R front and L/R surrounds from the Parsound P7 pre-amp before sending them to the stereo inputs on the Velodyne DD-15.

I used a pink noise generator to calibrate the ATI amps against my speakers. This required me to add a fair amount of gain, possibly because I am feeding the ATI's from the unbalanced P7 outputs, whereas I am using the balanced outputs to connect to the Parsound A51 amp and my speakers. Initially I forgot to disconnect the high-pass filters (Marchand 46XLR-A) between the pre-amp and amp, which would have made the exercise a bit nonsensical. When I realized what I had done, I disconnected them and redid the calibration, although it only made 1.5 dB difference.

Working on the integration of the D-15 sub and the front speakers, I found that the I could keep the crossover at 80 Hz (the same as the passive Marchand high-pass filters) and that a 24 dB/oct slope worked best. The room has a dropout coincidentally at the same 80 Hz frequency, so there was no "bump" caused by using the Bessel 12 dB/oct slope on the high-pass filters.

I knew I got it right when I listened to to Elton John's Madman Across the Water (Island B0003610-36):

.

The surround mix was done by Greg Penny for the 2004 SACD release. He had plenty to work with as the original was produced to perfection by Gus Dudgon. Anyway, when the bass guitar notes kick in at 2:19 in "Indian Sunrise", the bass sound is fixed in the overhead center position. Indeed, the whole disc sounded incredible and the bass seemed as coherent as I have heard. I guess that's my definition of success.

I am about to add a third ATI SUM100 to mix in the rear back channels, that is when I replace my side surrounds and add a center rear speaker. My front speakers are only 6 feet apart, as necessitated by the positioning of the sub in my room. So the center channel in the A51 will be feed with the rear back L+R from the Oppo BDP-95 player via the P7. The P7 output will also be summed with the surround L+R using the third ATI amp. All three will be connected together with horizontal joiners and DC power loops to make a single unit with one power supply.

In close, I want to thank those of you on the site who gave me assistance and ideas, without which I would never have gotten this far.

 

[davee70]

Post script

The plot thickened when I replaced my surround speakers recently.

I concluded that I needed a higher XO to the sub for the surround speakers than for my front speakers. So having just the one in the DD-15 would not be enough.

I am now awaiting delivery on a pair of Marchand XM46SB two channel low pass filters to replace the one in the subwoofer, one at 80 Hz/18 dB/oct for the front speakers (a tweak from the 24 dB/oct from before) and the other at 150 Hz/12 dB/oct for the surround speakers.

I also added a second subwoofer, a DD-12, to fill in a dip I had in my room. To manage the whole thing, I got a Rane SM 26s mixer to combine the low pass outputs for the L/R front and the L/R surround channels, along with the LFE channel. The Rane has two balanced outputs that are connected to the two subwoofers.

 

[davee70]

Post Post script

I finally received the Marchand XM46SB passive crossovers in December. The 2-channel low-pass filters, one at 80 Hz with 18 dB/oct slope per channel and the other at 150 Hz with 12dB/oct per channel, were intended to blend with the front L/R and surround L/R speakers respectively. (I had selected these values by running the test tone sweeps generated by the SMS-1 system internal to the Velodyne DD subwoofers and experimenting manually with the crossover settings.) Now, with the two Marchand crossovers inserted into the the Parasound P7's front and surround channel RCA-out signal paths, the internal crossovers on the Velodyne DD-15 and DD-12 are turned OFF. Nor am I using any EQ.

Initial checks with the SMS-1 tone sweeps confirmed that the front and rear speakers blended well with the two subs. However, through the surround speakers there was a notable drop-out at ~90-100 Hz, a bit above the 80 Hz 12 dB/oct (XM-46XLR) high-pass filters I am using on them but well below the 150 Hz 12dB/oct low-pass filters. I had observed a similar drop-out without the XM-46SB. I was able to eliminate this by setting the phase shift on the front subwoofer to exactly 180 degrees. This is the configuration I settled upon for two of the DD-15's six "pre-sets".

In listening sessions, I have noted (to my surprise and delight) a subtle yet distinct change in the quality of the bass I am getting in my listening room. It seems to have a more natural quality to it that it is difficult to describe in words. It seems as if the bass has a more fluid, less constrained manner to it. On surround recordings especially, it is almost like it flows through the room.

Sometimes I find I like the 180 deg phase shift presets (on the DD-15 in front) and sometimes I don't depending on the recording. However, usually I find myself preferring the phase shift on surround recordings that are recorded "live" in a concert hall setting, for instance, the way most classical music SACD's are done. I assume this is because there is significant bass content in the rear/surround channels.

To sum it up, substitution of the XM-46 analog crossover for the digital crossover in the DD-15 and DD-12 has made a significant improvement at the low end. There is a definite increase in warmth in the rear channels, which is what I expected, but there is a more general improvement that goes well beyond my expectations.

I must be happy with the latest changes because I am finding myself buying new music, instead of thinking about new equipment.

..... at least until I decide to add a center speaker.