Integrating subs with Minidsp SHD, should you block main speaker ports?

[Rockdog]

As for my situation, which differs from Rockdog's to some extent, I will definitely be running speakers at full range and likely crossing them over w/new subs at between 50 and 80 hz......to be determined, of course, when subs arrive.

My theoretical plan was to keep the ports plugged, use the PEQ on my RME ADI-2 to flatten out the hump, and use the subs to compensate for the lowered output.

Your prior post on measuring with and without plugs was very interesting and answers some basic questions about what freqs are primarily effected by plugs or no plugs. My 228s are probably very similar if not identical in that to what you measured. Thanks for posting.

I cross at 80 right now. The Soundoctor material is pretty strong about crossing at 80 or above. Don't remember a convincing explanation of why, outside of lowering power requirements and giving mains a break from full range duty. Seems like imaging was mentioned but maybe not. I doubt Revels would be effected one way or another. I've experimented quite a bit with different XO freqs, crossing as low as 40hz, but it all sounds better integrated at 80 or 90 to me. Your mileage may vary.

I have yet to read any convincing reason why leaving ports open would improve integration. With the theory that phase matching is more important than timings, I'm pretty convinced plugging ports makes intuitive sense. I've experimented with Dirac now with both plugged and unplugged, and even with that sophisticated eq, I hear plugged as a tighter and more seamless result. Subjective of course, and not precisely level matched, but a quick A/B back and forth is easy as pressing a button.

Thanks for the discussion. I will hopefully have time to work with REW and experimented with time alignment over the weekend and will post any thoughts.

 

[Rockdog]

By plugging the port, you are lowering the output at those frequencies above the port tuning(25Hz). You'll have to compensate for that decrease with EQ, which will increase distortion. The port gives you lower distortion above 25Hz, but increased distortion below 25Hz. Unless you're running them full range or crossing at 20Hz, plugging the port will increase distortion and limit headroom. It also depends on where your crossover is. If you're crossing at 40Hz, then plugging the port will increase distortion and limit headroom, but if you're crossing at 80 or 100Hz, it likely doesn't make much difference.

Badly designed ports can cause problems(ime), but the Revel ports are well designed and properly flared, so no problem.

Thanks

 

[Rockdog]

As for my situation, which differs from Rockdog's to some extent, I will definitely be running speakers at full range and likely crossing them over w/new subs at between 50 and 80 hz......to be determined, of course, when subs arrive.

My theoretical plan was to keep the ports plugged, use the PEQ on my RME ADI-2 to flatten out the hump, and use the subs to compensate for the lowered output.

Curious why you would run mains at full range? It would seem to me that integrating separate speakers playing the same freqs, in different physical locations as well, would be more of a challenge. If time alignment is important, then attempting that outside of just the crossover frequency with only a low pass (not a crossover) would be impossible as timings vary with wavelength. In timing integration, and even phase matching, its only really possible to do that at the crossover point, as each speaker does it's own thing outside that. Be interesting to hear your thoughts on that after playing with your new setup.

Now, happy to admit I could be completely wrong!

 

[RPG]

Your prior post on measuring with and without plugs was very interesting and answers some basic questions about what freqs are primarily effected by plugs or no plugs. My 228s are probably very similar if not identical in that to what you measured. Thanks for posting.

I cross at 80 right now. The Soundoctor material is pretty strong about crossing at 80 or above. Don't remember a convincing explanation of why, outside of lowering power requirements and giving mains a break from full range duty. Seems like imaging was mentioned but maybe not. I doubt Revels would be effected one way or another. I've experimented quite a bit with different XO freqs, crossing as low as 40hz, but it all sounds better integrated at 80 or 90 to me. Your mileage may vary.

I have yet to read any convincing reason why leaving ports open would improve integration. With the theory that phase matching is more important than timings, I'm pretty convinced plugging ports makes intuitive sense. I've experimented with Dirac now with both plugged and unplugged, and even with that sophisticated eq, I hear plugged as a tighter and more seamless result. Subjective of course, and not precisely level matched, but a quick A/B back and forth is easy as pressing a button.

Thanks for the discussion. I will hopefully have time to work with REW and experimented with time alignment over the weekend and will post any thoughts.

Agreed.....my current thinking is to leave the ports plugged. In addition to that 28hz hump, it appears I have a nice +9b peak right around 88hz---nothing to do with the ports, of course. Because of that, seat of pants would have me thinking that crossing my F208's at 80hz might be a bit too high. Seems a bit counterintuitive to me to EQ that peak down and then add some of it back in with an 80hz sub crossover. The SVS crossover calculator says F208's crossover should be at 50hz and from the Stereophile and Amir's in-room measurements, it appears the -3db point for the F208's is more like around 60hz. I'll probably start at 70hz and go from there. Somewhere in there between 50 and 70 is likely to be good, I'd expect.

 

[RPG]

Curious why you would run mains at full range? It would seem to me that integrating separate speakers playing the same freqs, in different physical locations as well, would be more of a challenge. If time alignment is important, then attempting that outside of just the crossover frequency with only a low pass (not a crossover) would be impossible as timings vary with wavelength. In timing integration, and even phase matching, its only really possible to do that at the crossover point, as each speaker does it's own thing outside that. Be interesting to hear your thoughts on that after playing with your new setup.

Now, happy to admit I could be completely wrong!

I have to admit my "rookieness" here. I think, because of both the equipment I have and my physical setup limitations, I've believed I have no choice but to run them full range. I have to run the subs with speaker-level connections because I cannot run wire from subs directly to amp (which has no room correction in it, anyway). So, subs will run off wires connected to the each speaker's binding posts, connected to a speaker level to line level converter, and then out from converter to sub. So, given that....without an AVR or some add-on dsp box, how would I NOT run the speakers full range?

 

[johnp98]

Curious why you would run mains at full range? It would seem to me that integrating separate speakers playing the same freqs, in different physical locations as well, would be more of a challenge. If time alignment is important, then attempting that outside of just the crossover frequency with only a low pass (not a crossover) would be impossible as timings vary with wavelength. In timing integration, and even phase matching, its only really possible to do that at the crossover point, as each speaker does it's own thing outside that. Be interesting to hear your thoughts on that after playing with your new setup.

Now, happy to admit I could be completely wrong!

I know the Geddes approach to subs is to have as many low frequency sources as in the low frequency range we hear more the steady state than at higher frequencies. Because of this he designed his speakers to be sealed and run 'full range' to have a greater overlap with numerous subs (and then subs also have their own crossover and slope all working toward a smooth combined frequency response.

That line of thinking seems reasonable to me, so I have plugged the ports of my speakers and run them full range and then use crossovers on the subs to integrate them with the mains based on location (one in the front center has a much higher crossover than nearfield subs behind me, as I don't want to localize those).

Hope that makes sense, but also curious what other people think about the Geddes method.

 

[RPG]

Interesting to know, thanks, @johnp98. My belief is that, due to my layout's limitations, I'm pretty much locked into Geddes' approach, regardless. So, I'm pretty convinced that your (or Geddes' way as the case may be) way is, indeed, how I'm going to have to do this. Good to hear that it's worked for you.

My suspicion is that, in a normal home environment, there is more than one way to achieve good sound---or at least the best sound one can get given the room and limitations within which one's system operates in. The measurement tools we have available at least give a fighting chance to explore many different options in terms of placements, crossover, gain levels, EQ, etc. in order to find as optimal a solution as possible given the constraints.

 

[Rockdog]

I have to admit my "rookieness" here. I think, because of both the equipment I have and my physical setup limitations, I've believed I have no choice but to run them full range. I have to run the subs with speaker-level connections because I cannot run wire from subs directly to amp (which has no room correction in it, anyway). So, subs will run off wires connected to the each speaker's binding posts, connected to a speaker level to line level converter, and then out from converter to sub. So, given that....without an AVR or some add-on dsp box, how would I NOT run the speakers full range?

Some subs have an actual crossover on board through the high level speaker terminals, and some just a low pass filter. I don't know about SVS, but my JLs have both a high and low pass to make a true crossover. However, John cites the Geddes approach which is well respected for sure. Heck, maybe I'll try running my mains full range! I have the ability to go either way, but amazingly have not tried that.

Lot of good info here, seems there's more than one way to skin a cat.

 

[richard12511]

Your prior post on measuring with and without plugs was very interesting and answers some basic questions about what freqs are primarily effected by plugs or no plugs. My 228s are probably very similar if not identical in that to what you measured. Thanks for posting.

I cross at 80 right now. The Soundoctor material is pretty strong about crossing at 80 or above. Don't remember a convincing explanation of why, outside of lowering power requirements and giving mains a break from full range duty. Seems like imaging was mentioned but maybe not. I doubt Revels would be effected one way or another. I've experimented quite a bit with different XO freqs, crossing as low as 40hz, but it all sounds better integrated at 80 or 90 to me. Your mileage may vary.

I have yet to read any convincing reason why leaving ports open would improve integration. With the theory that phase matching is more important than timings, I'm pretty convinced plugging ports makes intuitive sense. I've experimented with Dirac now with both plugged and unplugged, and even with that sophisticated eq, I hear plugged as a tighter and more seamless result. Subjective of course, and not precisely level matched, but a quick A/B back and forth is easy as pressing a button.

Thanks for the discussion. I will hopefully have time to work with REW and experimented with time alignment over the weekend and will post any thoughts.

If you're crossing at 80Hz, plugging the port doesn't really hurt distortion imo. Port is also not really contributing much, so shouldn't cause real phase problems.

Which measures better? If you're having a hard time hearing a difference, go with what measures better.

 

[richard12511]

Curious why you would run mains at full range? It would seem to me that integrating separate speakers playing the same freqs, in different physical locations as well, would be more of a challenge. If time alignment is important, then attempting that outside of just the crossover frequency with only a low pass (not a crossover) would be impossible as timings vary with wavelength. In timing integration, and even phase matching, its only really possible to do that at the crossover point, as each speaker does it's own thing outside that. Be interesting to hear your thoughts on that after playing with your new setup.

Now, happy to admit I could be completely wrong!

Running mains full range with speakers like the Salon2 can work sometimes, as it's kinda like adding 2 more subwoofers. It's more difficult, though, and takes more integration knowledge and skill. I know a couple people with JTR 215RTs(which stomp the Salon2 in terms of extension and output) and Captivator 4000ULFs, and both of them cross to the Captivators at 80Hz.

 

[RPG]

Some subs have an actual crossover on board through the high level speaker terminals, and some just a low pass filter. I don't know about SVS, but my JLs have both a high and low pass to make a true crossover. However, John cites the Geddes approach which is well respected for sure. Heck, maybe I'll try running my mains full range! I have the ability to go either way, but amazingly have not tried that.

Lot of good info here, seems there's more than one way to skin a cat.

As near as I can tell, the SVS's only have a low pass. Nonetheless, I've had a little bit of experience integrating the old subs with full range speakers (using the speaker level connection), so I think I'll eventually get there (famous last words, I know!). I gotta' read up on this Geddes dude, so thanks for the reference.

 

[RPG]

I just watched this. Very interesting and informative re. the Geddes "method". Certainly one of a number of ways to do this but, in my particular situation, probably closer to what I will actually have to do as opposed to what I might like to do.

The Geddes Multi-Subwoofer Bass Optimization Approach - YouTube

 

[Rockdog]

I just watched this. Very interesting and informative re. the Geddes "method". Certainly one of a number of ways to do this but, in my particular situation, probably closer to what I will actually have to do as opposed to what I might like to do.

The Geddes Multi-Subwoofer Bass Optimization Approach - YouTube

Just looked at briefly but will sit down for the full look. Very interesting. I'd read about the Geddes approach but it's been a while. One solution is to just spend more money and get about 5 subs! I will say, however, the Minidsp SHD I bought, primarily to replace my analog JL crossover and to add Dirac and EQ, may be some of the best dollars I've ever spend on audio. And, selling the used equipment I actually came out ahead. Not cheap, but not crazy by audiophile standards and it opens up complete control of what we're discussing. After setting phase and level manually on subs, then applying Dirac measurements, it's a big leap ahead in integration. Come on over, I'll pour some nice bourbon and see what you think..

Honestly, it's awesome to be able to experiment with different methods. You have great mains and the subs on the way.. It ain't gonna suck no matter what you do!

Running mains full range with speakers like the Salon2 can work sometimes, as it's kinda like adding 2 more subwoofers. It's more difficult, though, and takes more integration knowledge and skill. I know a couple people with JTR 215RTs(which stomp the Salon2 in terms of extension and output) and Captivator 4000ULFs, and both of them cross to the Captivators at 80Hz.

Damn, that's some seriously low-end capable speaker setup! Think I'd trade the Salons for the mains, but those JTRs list being capible of 128db output! Might be loud in the 50th row at Red Rocks! I'll end up trying my Revels at full range and trying to integrate with two subs. It does make sense to add sub outputs if possible. Ports plugged. Be interesting to see how Dirac handles that.

 

[levimax]

I. There seems no simple way to do it, or at least not a widely accepted way that I can glean from lots of reading here. Audiolense gets high praise for time alignment, but I can't afford another software.

Rephase is free and will do what you want. There is a learning curve but once you figure it out it is very powerful.

 

[richard12511]

Damn, that's some seriously low-end capable speaker setup! Think I'd trade the Salons for the mains, but those JTRs list being capible of 128db output! Might be loud in the 50th row at Red Rocks! I'll end up trying my Revels at full range and trying to integrate with two subs. It does make sense to add sub outputs if possible. Ports plugged. Be interesting to see how Dirac handles that.

I'd take the Salon2 as well, unless I'm at 7m or something. The Salon2 is definitely a better speaker, but the JTR digs deeper(10Hz in room) and plays much louder. One of the best setups I've ever heard(maybe the best multichannel) was a 7.0 JTR 215 RT setup at Axpona.

All the speakers were pretty distant from the MLP, so the super narrow dispersion was wonderful and imaged better than the normal setup you get with wide dispersion setups. Don't know how practical it is though, as usually you need wide dispersion speakers for the surrounds, since they're typically closer to mlp. Oh, and the bass was monstrous, but also so tight and well controlled.