I just completed a DIY sub and I'm testing the response using an SPL meter and "Room EQ Wizard" software. It's pretty cool to move your sub around the room and see how it affects the frequency response curve.

Anyway, I experimented with both the digital and analog crossovers in the 950, and graphed the sub's response curve. This is with the mains unplugged, so "double bass" is not a factor here.

Here are the two (disappointing) observations:

1) The analog crossover is not at 80 as claimed, but is more like 120. It looks like signal sent to the sub has a high pass at about 120, and the mains' signal has a low pass at about 110. This is way too high for my setup.

2) The bass signal when using analog crossover is boosted 5 DB relative to the bass signal when using the digital crossover. This is clearly audible when listening to music. This means if I want to switch from the CD input to the 6-channel input, I need to reduce the subwoofer trim by 5 DB to avoid having overpowering bass.

I ordered an external crossover (paradigm x-30) so I hope will solve my probem. I will tell the 950 that I have no sub, and the x-30 will pull out the lower frequencies. My preliminary experiments suggest that the LFE is blended into the mains at the proper volume.

I noticed a profesional review of the 970 mention that the crossover was higher than 80 as well. I wonder why outlaw does this, and whether it affects the 990 as well.

Other than this issue I like the 950, and I'm really happy with the 7125.

I'm not sure this helps or hurts, but in my 950, the analogue crossover (attached to the 6-ch analogue inputs) inverts the summed LFE output when it is engaged.

I run the subwoofer signal from the 950 through a Behringer DCX2496 which does equalization. It also has a one-touch phase inversion function on each channel. When I listen to CD through coaxial digital input or 2-ch SACD through the 2 channel analogue inputs, I keep the phase normal. When running SACD surround through the 6 channel bypass I invert the subwoofer signal.

I'm not sure why it's implemented this way either, but at no time do I have to adjust the sub level. Is it possible this is what you're seeing? If the sub is inverted and fighting with your mains, it may seem like you need to boost the sub level.

Inverted phase from the analog crosover only? Yikes.

It's possible this is contributing to the effect I get when I listen together with the mains. But this isn't responsible for the reported level jump, because I took those SPL measurements with the mains off.

Does the 110 hz crossover make sense in your case? That's pretty high unless your mains are pretty tiny.

rumatt,

Can you recheck your post? Are you trying to say that the sub channel has a low-pass of 120 and the mains have a high pass of 110? It also sounds like you measured this in your listening room. Did you account for room effects?

For the 6 CH direct, we parsed out (when the 950 came out) that the sub channel has a 120 Hz low pass filter that prevents higher frequencies from going to the sub. This is only a problem for someone running a full range speaker from this channel and listening to a DVD-A that doesn't use this channel exclusively for a subwoofer. An example of such media is Fleetwood Mac, Rumors. There is snare drum and voice in the "subwoofer" channel. The Outlaw ICBM has a 150 LP filter on the sub.

It's been a long time since I've done this, but when I took measurements for the 950 in my system, I'm sure I didn't see the analog crossover rolloff begin until about 90 Hz in my listening room. I also don't remember observing a boost from the 950, though I did have to attenuate the sub channel due to the boost applied by my DVD player.

And yes, the analog input is 180 degrees out of phase on the sub channel.

Ooops, yeah I always get "high" and "low" terminology backwards (whether you allow the high, or filter it out).

I did the measurements in my room, so room effects are in my measurements, but they shouldn't affect the comparison because I was comparing the analog and digital crossovers, and the room is constant for both. Here's what I did to measure the high-pass signal sent to the mains:

1) Disconnect sub

2) Measure frequency response curve to mains when using the 6-channel inputs with the analog crossover enabled

3) Measure the frequency response when using the CD input and the digital crossover. I took a measurement with each crossover value (50, 60, 80, 100, 120, 150).

The graph for the analog crossover (high pass, to the mains) was somewhere right between the digital 100 and 120 curves (thus, I'm claiming 110, which is confirmed by a review of the 970).

To measure the low pass, I reversed the experiment (plugged in the sub, and unplugged the mains). The sub response with the analog crossover was identical to the digital 120 Hz crossover, except the analog was 5 db higher.

I think it's somewhat silly that they would include such a high crossover. I think it's really, really bad that they falsly advertise it. People pay big money for this equipment, and may make a buying decision based on the presence (or absense) of a proper 80 Hz crossover.

Okay, please indulge me a few more questions. Finding the 120 LP filter on the sub channel is not surprising but it doesn't mean that it's the true crossover point. Noting that you saw a drop off in your mains between 100 and 120 is surprising.

Please don't take this as an attack on or dispute of your analysis. I spent quite a bit of time dialing in my 6 CH Direct input and I didn't get the same crossover results as you. I'd like to try and duplicate your results on my system.

Were you feeding the 6 CH Direct input with the analog output of the CD player? Or was it a DVD-A/SACD player? Was the 6 CH Direct receiving a stereo signal hooked into the mains? If it was a DVD-A/SACD player for the 6 CH Direct and a CD player for the digital input, there could be a discrepancy induced by the bass management of the DVD player. If you used a DVD-A/SACD player for the 6 CH direct, do you have it hooked up to a digital input as well? If yes, did you compare that digital input to the digital input for the CD player. I would assume those two to be identical but you never know. In the 950, I assume your speakers are set to small. Have you set them to Large and measured where they naturally roll off in your room with the sub off?

Regarding the measurements, were they taken from the sweet spot or did you take them close to the mains for the high frequency rolloff? Are you using sine waves or full range pink noise?

Thanks in advance for the details.

Bill

Hi Bill,

No problem.

I'm using Room EQ Wizzard software (free) and a USB sound card connected to my laptop to generate the frequency sweeps. I was using the 2-channel analog output of the sound card, sent into the L&R of the 6-channel input.

I'm using a cheap Nady SPL meter to record the sound levels.

I didn't compare the digital input, but my sound card has a digital output, so will try and report back.

The speakers are set to small. When large, they start to drop off at about 45 Hz. I will post some graphs later when I get home. It's pretty cool stuff actually.

I'm not sure how yours measured differently. My data could be wrong, but the fact that a) the curve matched so well with the 100/120 curves, and b) that the this review also mentions the 970 is crossed over at 110, makes me think it's most likely correct.

Since you compared your results to when your mains were set to large, you appaer to have naturally compensated for error in the SPL meter, but it is worth pointing out that SPL meters have a hard time at lower frequencies (the compensation that that SVS has on their site indicates that the widely-used Radio Shack meters start under-reporting low frequencies around 120Hz).

I'm very interested in seeing your graphs. I was going to point out the trouble that SPL meters have with lower frequencies but Gonk (as always) took care of that. The Radio Shack SPL meter is down 1 to 2 dB in the range of interest here so that's probably a non-issue.

I'm not sure if the 970 is the same animal wrt the analog crossover in the 6 CH direct. Regardless, I find it hard to believe that none of us noticed a 110 vs 80 Hz crossover when the 950 came out. We were all over this thing if you check the archives.

I'll be tied up for the next couple of days but I'll check in when I can.

Gonk, if there's one constant on the internet, it is you contributing to this forum. I'm glad to see you're still going strong.

Regards,

Bill