TOPIC:

eisenb11 wrote:

Sorry I missed this reply. Ballpark 80 Hz is usually recommended but I'd be partly guided by the meaured amplitude and phase responses of the speakers and subwoofer. For example, if your main speaker output had a big dip at 80 Hz but your sub didn't then a higher crossover would make sense. I've used as high as 110 Hz with a 48 dB/octabe LR crossover. The risk of going too high is you start to be able to tell that sound is coming from the subwoofer rather than the main speakers.

Ultrasonic wrote:

 

Thanks for the input! This is all pretty new to me.

I attached is a pic of what Dirac recorded last time I ran it (no sub yet). Looks like things drop off around 40 Hz with things looking a little wonky around 60 Hz. There is some weird dip around 150 Hz. I guess the caveat being that since this is Dirac, the graph also has the effect of room modes built into it, so it's not pure speaker performance.

At a glance, it looks like a 80 Hz crossover w 48 dB LR starts the transition around 120 Hz and completely hands off to the sub around 52 Hz. Sounds like a reasonable starting point?

eisenb11 wrote:

Ultrasonic wrote:

 

Thanks for the input! This is all pretty new to me.

I attached is a pic of what Dirac recorded last time I ran it (no sub yet). Looks like things drop off around 40 Hz with things looking a little wonky around 60 Hz. There is some weird dip around 150 Hz. I guess the caveat being that since this is Dirac, the graph also has the effect of room modes built into it, so it's not pure speaker performance.

At a glance, it looks like a 80 Hz crossover w 48 dB LR starts the transition around 120 Hz and completely hands off to the sub around 52 Hz. Sounds like a reasonable starting point?

I'd suggest you learn how to use Room EQ Wizard to make measurements. This (or something comparable) will be necessary to know how to set the time delays to optimise the main speaker to subwoofer crossover.

Ultrasonic wrote:

 

Thanks again. Nirvana! I just added a DB3D to my 805D3 book shelves a few hours ago. Holy moly what a difference - filled in the low end that I was missing. The 805s go decently low, after some playing around, I ended up setting a high pass @ 60 Hz. Adjusted Dirac to stop @ 80 Hz on the low end as the sub has it's own DSP going on. 

eisenb11 wrote:

Thanks again. Nirvana! I just added a DB3D to my 805D3 book shelves a few hours ago. Holy moly what a difference - filled in the low end that I was missing. The 805s go decently low, after some playing around, I ended up setting a high pass @ 60 Hz. Adjusted Dirac to stop @ 80 Hz on the low end as the sub has it's own DSP going on. 

Glad your're happy  .

Is your subwoofers own DSP measurement based? If not then Dirac Live is likely to do a better job, and I'd at least experiment with it at some point.

Ultrasonic wrote:

Glad your're happy 

.

Is your subwoofers own DSP measurement based? If not then Dirac Live is likely to do a better job, and I'd at least experiment with it at some point.

I ended up setting everything up again a day later, so I could go through a more scientific process.

The first time, I set up the sub first - it has it's own DSP and does measurement based corrections. Then I set the a high pass on the speakers and let the full signal go to the sub since it sets its own low pass. Sounded great.

But then I thought about it and I have no idea what it's doing since it's all abstracted away. So I did it again.

Turned off DSP and filters on the sub. Measured the sub in REW and created Auto-EQ files. Then measured in REW again with the speakers on to get the levels close to even. Then I just ran the whole thing (speakers with sub) through Dirac Live. Sounds great too.

I'm not using the fancy DSP in the sub anymore by doing that, but I think that it should be better this way since via the prior method, the sub didn't know about the speakers and Dirac for the speakers didn't know about that sub.