Sundown ZV4 18D2 - sealed enclosure

[Droogne]

Put the last panel in (without glueing). How does it sound? Impressive ö. That things moves some impressive amounts of air! Have not yet had the opportunity for some critical listening, but so far it does sound good (appart from amazingly loud). Not boomy at all! Means I'm gonna glue that panel on later this day (you can literally feel all the air escaping from that panel). 

 

Next up: EQ! Will do some measurements after I close it up. I still got my horn sub, so gonna try to integrate it, any tips? Will try a 15-40hz (EQ flat down to 15hz) for the Sundown, and 40-80hz for the horn sub. 

[SME]

Looking good!

Definitely make absolutely certain that the box is completely sealed or else you'll end up with a lot of noise and/or whistling.

As for integrating them with the bass horn, that's likely to be tricky.  For best results, you'll want to do a lot of experimentation, trying a lot of different settings and taking measurements to find what works best.  Start by measuring the horn and the subs without the XO.  For the cleanest reproduction, you'll probably want to add at least a 2nd order high pass filter to the horn.  The subs will likely need at least a 4th order and maybe even a 6th or 8th order low pass filter to narrow the region that they overlap.  An XO at 40 Hz sounds reasonable.  Though without knowing the native response of your horn, it's hard to say for certain.

A crucial parameter to optimize is the relative delay between the subs and the horn.  You may need more delay on the horn than the subs, in which case, you may need to adjust delay(s) on the mains too.  Note that the best relative delay is likely to change depending on XO settings, so you may need to do a lot of experiments to find what works best.  Another thing to try when trying to get the best integration is inverting the phase on the subs.  This will completely change what delay works best, so keep that in mind.

[Droogne]

44 minutes ago, SME said:

Looking good!

Definitely make absolutely certain that the box is completely sealed or else you'll end up with a lot of noise and/or whistling.

I didn't hear anything during my short listening session, but I definitely did when doing the measurement. A remarkably loud whistling noise during the first frequencies of the sweep. I only left it unglued to make sure there was no problem. Do you think it not being sealed could influence the effect of the damping? Cause thats the primary reason I wanted to check it first before I glued it shut. What is a good way to check for the boominess? Pushing the sub to its limits with some pink noise? Or something else maybe? 

44 minutes ago, SME said:

As for integrating them with the bass horn, that's likely to be tricky.  For best results, you'll want to do a lot of experimentation, trying a lot of different settings and taking measurements to find what works best. 

Start by measuring the horn and the subs without the XO. 

I'll do that tomorrow. Already got some pretty amazing measurements (pretty flat down to below 20hz) from the Sundown.

44 minutes ago, SME said:

For the cleanest reproduction, you'll probably want to add at least a 2nd order high pass filter to the horn.  The subs will likely need at least a 4th order and maybe even a 6th or 8th order low pass filter to narrow the region that they overlap.  An XO at 40 Hz sounds reasonable.  Though without knowing the native response of your horn, it's hard to say for certain.

I always used the horn down to 40hz, it doesn't perform well below that point. 

44 minutes ago, SME said:

A crucial parameter to optimize is the relative delay between the subs and the horn.  You may need more delay on the horn than the subs, in which case, you may need to adjust delay(s) on the mains too. 

Could I just add the physical length of the horn, or is it not as easy as that? And how would I measure/calculate the distance/delay settings?

44 minutes ago, SME said:

Note that the best relative delay is likely to change depending on XO settings, so you may need to do a lot of experiments to find what works best.  Another thing to try when trying to get the best integration is inverting the phase on the subs.  This will completely change what delay works best, so keep that in mind.

I hope it doesn't go as complex as that. Trying to sell the horn sub. Might be easier to just leave the horn sub alone and let the pretty capable Sundowns take it up to 80hz. 

[SME]

4 minutes ago, Droogne said:

I didn't hear anything during my short listening session, but I definitely did when doing the measurement. A remarkably loud whistling noise during the first frequencies of the sweep. I only left it unglued to make sure there was no problem. Do you think it not being sealed could influence the effect of the damping? Cause thats the primary reason I wanted to check it first before I glued it shut. What is a good way to check for the boominess? Pushing the sub to its limits with some pink noise? Or something else maybe? 

Are you not able to remove the driver?  Most of us glue the cabinet completely before stuffing it and only install the driver at the end.  If something need to be changed, then we'll remove and reinstall the driver.  If the enclosure was leaky, I don't think you would know for sure if there was enough stuffing.  However, I think you'll be fine with what you show in your picture, plus "a bit more".

6 minutes ago, Droogne said:

I always used the horn down to 40hz, it doesn't perform well below that point. 

Then you may want to cross a bit above there, like 45-50 Hz so that the overlap is better.

7 minutes ago, Droogne said:

Could I just add the physical length of the horn, or is it not as easy as that? And how would I measure/calculate the distance/delay settings?

I hope it doesn't go as complex as that. Trying to sell the horn sub. Might be easier to just leave the horn sub alone and let the pretty capable Sundowns take it up to 80hz. 

You can use the physical length of the horn as a starting point.  Are your mains processed by the same electronics as the different subs are?  Electronics can add delay too.  Either way, for the best performance, you'll want to measure frequency response with different delays and choose the one that gives you the most / smoothest output.

[Droogne]

17 minutes ago, SME said:

Are you not able to remove the driver?  Most of us glue the cabinet completely before stuffing it and only install the driver at the end.  If something need to be changed, then we'll remove and reinstall the driver.  If the enclosure was leaky, I don't think you would know for sure if there was enough stuffing.  However, I think you'll be fine with what you show in your picture, plus "a bit more".

I was not sure if the bolts for mounting the sub would be screwed in tightly enough not to fall in when trying to put in the monsterous sub, which is why I mounted the sub first (so I was able to have access to the bolts from both sides). To be clear, I did mount it in a way that I do have the possibility to remove the driver after everything is glued shut. I also liked the fact that I had way more access to the inside for attaching the damping material to the walls. 

Quote

Then you may want to cross a bit above there, like 45-50 Hz so that the overlap is better.

You can use the physical length of the horn as a starting point.  Are your mains processed by the same electronics as the different subs are?  Electronics can add delay too.  Either way, for the best performance, you'll want to measure frequency response with different delays and choose the one that gives you the most / smoothest output.

The processor are not the same no.. Both from Xilica though. And by measurement do you mean a measurement from all speakers at the same time? 

 

 

[SME]

24 minutes ago, Droogne said:

The processor are not the same no.. Both from Xilica though. And by measurement do you mean a measurement from all speakers at the same time?

If you can, just measure one channel at a time, like the center, but with all the crossover(s) and subs active.

[Droogne]

I can do that for my LCR and the rear surrounds. Those are processed through the Xilica. Will try it out tomorrow, thanks! 

[Droogne]

Ok, second sub is almost done, what do I do next? I see my Xilica doesnt have the option to EQ/cross lower than 20hz  the Audyssey app would supposedly give me that option, but I cant seem to get the app working so I'm not sure what to do now hmm.. Buy a new and capable cross? Or just cross/EQ everything from 20hz? And what kind of measurements should I do except response and delay?

 

[Droogne]

Hey @SME

 

What would you recommend for placement of the subs? They fill in the space between the LaScalas now, but I know that placing them on similar sides is not ideal. Should I maybe move the right sub to the right of the right LaSCala (into the corner)? 

Also, do you have some advice on my question above this one? 

 

 

 

 

[Infrasonic]

These are sealed subs so you don't need to worry about setting any highpass filter or anything. At least not yet. Also "EQ'ing" below 20hz isn't really all that necessary. However, boosting the lowend with a shelving filter would be useful and the corner frequency for that does not require to be under 20hz and likely somewhere in the 20-35hz area. So that's that.

You will have to measure your response and go from there. Setting up a new system will require some trial and error.

Placement also depends on the geometry of the room and where you will have the MLP. Measure any location(s) you can or want. Don't just guess or simulate it. See what actually performs the best.

[SME]

4 hours ago, Droogne said:

Hey @SME

What would you recommend for placement of the subs? They fill in the space between the LaScalas now, but I know that placing them on similar sides is not ideal. Should I maybe move the right sub to the right of the right LaSCala (into the corner)? 

Also, do you have some advice on my question above this one?

Where did you get that idea?  In a symmetric room, placing them in similar locations on each side can help reduce the effects of room modes.  However, in an asymmetric room, it doesn't make as much of a difference.  There are pros and cons to different locations, and every room is different.  For sub performance alone and especially when you have EQ available, I would say that the best location is in corners or more generally as close as possible to the longest walls in the room.  From your picture, that would appear to be the corner on the left (if it were available).  However, another factor is how well the subs integrate with the mains.  For quite a lot of the range, the subs and mains will be having to play together, and placements near the mains can help with that.

I also agree with @Infrasonic that your best bet is to try a few configurations and take measurements.  This can be quite an involved process being that you have two subs and probably want the best integration with the mains too.  Each configuration is likely to have its own optimal settings for the delay/distance on each sub.  The optimal crossover frequency can also vary.

How many configurations options do you have available?

By the way, it looks like your left and right speakers are not the same distance from the center.  Is this on purpose?  Is your main listening position centered between them and it's your center channel that's misaligned?  If so, then perhaps that's OK.  Otherwise, you should probably move the left or right speaker so that they are the same angle relative to where you sit, even if that means bringing the right one more to the inside.

[Droogne]

5 hours ago, Infrasonic said:

These are sealed subs so you don't need to worry about setting any highpass filter or anything. At least not yet. Also "EQ'ing" below 20hz isn't really all that necessary. However, boosting the lowend with a shelving filter would be useful and the corner frequency for that does not require to be under 20hz and likely somewhere in the 20-35hz area. So that's that.

Ok, good to know. Was exactly what I was doing know (and what REW predicted).

5 hours ago, Infrasonic said:

You will have to measure your response and go from there. Setting up a new system will require some trial and error.

Placement also depends on the geometry of the room and where you will have the MLP. Measure any location(s) you can or want. Don't just guess or simulate it. See what actually performs the best.

Haha, I should have known that was gonna be the solution (as it usualy is  ) . Will try maybe one or 2 more placements. 

[Droogne]

5 hours ago, SME said:

Where did you get that idea? 

Now that you ask me... No idea. Read it somewhere once, but I might have misunderstood back then (and remembered it differently because of that). My room is really really assymetrical (an L-shape, with the second leg being the kitchen and extending from the right back.)

5 hours ago, SME said:

 

In a syetric room, placing them in similar locations on each side can help reduce the effects of room modes.  However, in an asymmetric room, it doesn't make as much of a difference.  There are pros and cons to different locations, and every room is different.  For sub performance alone and especially when you have EQ available, I would say that the best location is in corners or more generally as close as possible to the longest walls in the room. 

The left wall is around 3-4m long, but as yo can see, it's not really directly next to the "TV / speaker" wall. It kind of protrudes. Should I try to get the sub completely in that corner? The right corner has a 4m wall (the "TV wall") and a 2,2m wall. 

5 hours ago, SME said:

 

From your picture, that would appear to be the corner on the left (if it were available).  However, another factor is how well the subs integrate with the mains.  For quite a lot of the range, the subs and mains will be having to play together, and placements near the mains can help with that.

 

5 hours ago, SME said:

How many configurations options do you have available?

By the way, it looks like your left and right speakers are not the same distance from the center.  Is this on purpose?  Is your main listening position centered between them and it's your center channel that's misaligned?  If so, then perhaps that's OK.  Otherwise, you should probably move the left or right speaker so that they are the same angle relative to where you sit, even if that means bringing the right one more to the inside.

Haha, I think I uploaded the wrong pic! This was the photo 3 seconds before I moved the whole setup (except the right LaScala which was positioned correctly) to the right. Will try to move the right sub to the right of the LaScala and see how much the boost/effect is. Might also try to place the left one in the cove to the left. I could also try to place the right sub to the left back corner (4m wall and 4m wall), that way I would have 1 in right front and in left back, so also symmetrical in a way no? And also spread out. 

 

 

[Droogne]

For reference. 

 

Red = couch

Purple = TV/center

Blue = Fronts

Yellow = subs as they are positioned now (moved the right one into the corner).

[Droogne]

Also, I took some measurement (lot of different combinations). Anyone interested in helping me out figuring out what they mean, and how I can apply this for a good response in this position? Cant upload it, as it takes in too much space. Below is the response when running both subs without PEQ etc. Both dibs @16hz and 55hz get way deeper when one of the subs is run in opposite polarity. Does this mean they are canceling out each other partly? There is one dip @ 28hz which is only present when they are run with the same polarity. 

It looks like the left sub is almost completely flat down from 7hz to 80hz (now that I think of it, not sure if I used a high enough volume..). The right sub (corner) is flat from 5,8hz up to 30hz, then from 30hz to 60hz it gets a 15db boost, and then down again to 0db between 60-80hz. This translates into that plateau you see below when running both together. What is the best way to take on that 15db boost? Attenuate this part? 

Also, what kind of smoothing should I apply on this? 1/3th? 

same polarity:

 

opposite polarity:

 

[Infrasonic]

Dude. Shrink that dB range. It's a whopping 20dB per division!!! You need to set it to about 5dB per division. It will, of course, make your response look .... much worse. But it's for a good cause! Trust me.

 

Smoothing? It's up to you. Don't go any more coarse than 1/6 which is a good way presenting an entire frequency response to a discussion but not very useful for doing the nitty-gritty, which you're deep into now. I usually go fully unsmoothed (or barely smoothed like 1/24) until I've knocked out the bad stuff. Then I usually go with 1/6-1/12 just cuz it makes it easier to read and towards the end of your journey (being real for a moment) isn't as necessary because... who cares. Posting a frequency response of one's system with no smoothing is pretty annoying and not very useful to anyone unless you're trying to troubleshoot.

So.. at first, use no smoothing. Then after you have done most of the PEQ work, there is little need NOT to smooth.

If that makes sense.

 

Repost with better graphs and we can move onto the next section....

 

 

5-200hz is a great bandwidth to look at while measuring your subwoofer system. You want at least an octave above and below the operating range to make sure things are integrating well (on the top end) and how well the extension is.

[Droogne]

9 hours ago, Infrasonic said:

Dude. Shrink that dB range. It's a whopping 20dB per division!!! You need to set it to about 5dB per division. It will, of course, make your response look .... much worse. But it's for a good cause! Trust me.

Whoa! I think I scrolled out to much to get a good view (which was ofcourse not the best way to achieve this). Looks quite different @ 5db intervals true.

Quote

 

Smoothing? It's up to you. Don't go any more coarse than 1/6 which is a good way presenting an entire frequency response to a discussion but not very useful for doing the nitty-gritty, which you're deep into now. I usually go fully unsmoothed (or barely smoothed like 1/24) until I've knocked out the bad stuff. Then I usually go with 1/6-1/12 just cuz it makes it easier to read and towards the end of your journey (being real for a moment) isn't as necessary because... who cares. Posting a frequency response of one's system with no smoothing is pretty annoying and not very useful to anyone unless you're trying to troubleshoot.

So.. at first, use no smoothing. Then after you have done most of the PEQ work, there is little need NOT to smooth.

If that makes sense.

It does! I used psychoacoustic smoothing to calculate PEQs for my mains (by recommendation), but wasnt sure if that was applicable to subs too. 

Quote

Repost with better graphs and we can move onto the next section....

 

 

5-200hz is a great bandwidth to look at while measuring your subwoofer system. You want at least an octave above and below the operating range to make sure things are integrating well (on the top end) and how well the extension is.

I set my presets to 5-200 to get the best view (without having the inclination to scroll out to see that range, causing a db range jump). I applied 1/12th smoothing. Same graphs as before, and I also added left and right sub seperately. Is this a more helpful presentation? 

Both subs

 

Both subs (1 polarity inversed)

 

 

Left sub:

 

 

Right Sub

[SME]

In your room, the left corner is more beneficial than the right because there is no large opening near it like there is with the right side.  The "other" left corner, against the left wall instead of the front wall, may be even better.

From your measurements the two subs obviously work better if they use the same polarity.  It also looks like some low end bass boost would be a good idea.  Are these measurements done with the sub low pass filter active?  I assume so because there is a lot of roll-off above 80 Hz.

Also, in your second set of plots you are using a linear instead of log frequency axis.  I suggest using a log axis for future plots.  Also, it's helpful to be able to compare multiple measurements at the same time on the "All SPL" tab.

[Droogne]

5 hours ago, SME said:

In your room, the left corner is more beneficial than the right because there is no large opening near it like there is with the right side.  The "other" left corner, against the left wall instead of the front wall, may be even better.

So what would you recommend? One sub in right corner, and one in the "other" left? Seems plausible if I change up the position of the amps/processors. 

5 hours ago, SME said:

From your measurements the two subs obviously work better if they use the same polarity. 

 

It also looks like some low end bass boost would be a good idea.  Are these measurements done with the sub low pass filter active?  I assume so because there is a lot of roll-off above 80 Hz.

I measure "as is", which means through the AVP. I did disable the crossover in the processor. Will "disable" the AVP processor too next time ( by setting the crossover @250hz).

5 hours ago, SME said:

Also, in your second set of plots you are using a linear instead of log frequency axis.  I suggest using a log axis for future plots.  Also, it's helpful to be able to compare multiple measurements at the same time on the "All SPL" tab.

Wasnt sure about the axis, so good to know! Will do them all together next time.

 

For now I'm gonna try the different left sub position and do some measurements. Will also try some PEQs and include those. I know not to boost certain dips, how do I know which ones I can better leave alone? 

[SME]

8 hours ago, Droogne said:

So what would you recommend? One sub in right corner, and one in the "other" left? Seems plausible if I change up the position of the amps/processors. 

I recommend going with what measures best.

Another option to consider is to put the other sub on the left at the rear of the room, as laid-out.  That will likely give you the best deep bass performance.  However, integrating with the mains is likely to be trickier.  It might work OK though if you are using it with mid-bass horn(s) and/or the XO frequency is lower.

[SME]

8 hours ago, Droogne said:

For now I'm gonna try the different left sub position and do some measurements. Will also try some PEQs and include those. I know not to boost certain dips, how do I know which ones I can better leave alone? 

Any boost you do sacrifices headroom, and if you boost in the wrong place, you can end up making the bass more resonant / boomy.

Take measurements at multiple locations in the room and avoid EQing (boost or cut) anything that is not consistent from seat-to-seat.  The averaging feature in REW can be helpful here.

Note that you shouldn't worry about EQ until you've optimized distances.

[Infrasonic]

8 hours ago, Droogne said:

So what would you recommend?

Trying for yourself any physical location that seems practical enough for you. You don't need any one on the internet to help you out with this part.

 

Setting up two subs at the same time can be quite involving. I'd suggest measuring/testing one at a time (took make things easier). You will find an ideal location easier for one. For simplicity sake, you can even stack the 2nd sub on top of the first. Eventually you will find that a spot works rather well, then you can try and integrate the 2nd sub. Doing it this way will probably take more time but you will make less mistakes.

Another tip is to treat two sub locations as one. If you can place the subs in locations that are equidistant then setting up delay/distance/phase will be easier and applicable to both physical enclosures. Likewise, you can EQ them as a "single" unit. This will also aid in the simplicity and less chances for mistakes. Trust me, I've been through it.

[Droogne]

On 5/14/2018 at 7:53 PM, SME said:

I recommend going with what measures best.

Another option to consider is to put the other sub on the left at the rear of the room, as laid-out.  That will likely give you the best deep bass performance.  However, integrating with the mains is likely to be trickier.  It might work OK though if you are using it with mid-bass horn(s) and/or the XO frequency is lower.

Lower crossover might be hard (they are at 80hz, and 60hz for the LaScalas might be hard).  Can you tell me why it would be easier to integrate? 

[Droogne]

On 5/14/2018 at 7:57 PM, SME said:

Any boost you do sacrifices headroom, and if you boost in the wrong place, you can end up making the bass more resonant / boomy.

Take measurements at multiple locations in the room and avoid EQing (boost or cut) anything that is not consistent from seat-to-seat.  The averaging feature in REW can be helpful here.

Note that you shouldn't worry about EQ until you've optimized distances.

Will do. Thanks for the tip!

[Droogne]

On 5/14/2018 at 8:16 PM, Infrasonic said:

Trying for yourself any physical location that seems practical enough for you. You don't need any one on the internet to help you out with this part.

It's a shame I dont have anyone IRL to help with this part ? those subs are heavy beasts ! Also a shortage of cables.. I went with speakon cables in the end, and mine are not nearly long enough to run through the room to test all positions.. Will figure something out, but it might take some time.

On 5/14/2018 at 8:16 PM, Infrasonic said:

 

Setting up two subs at the same time can be quite involving. I'd suggest measuring/testing one at a time (took make things easier). You will find an ideal location easier for one. For simplicity sake, you can even stack the 2nd sub on top of the first. Eventually you will find that a spot works rather well, then you can try and integrate the 2nd sub. Doing it this way will probably take more time but you will make less mistakes.

Another tip is to treat two sub locations as one. If you can place the subs in locations that are equidistant then setting up delay/distance/phase will be easier and applicable to both physical enclosures. Likewise, you can EQ them as a "single" unit. This will also aid in the simplicity and less chances for mistakes. Trust me, I've been through it.

That is actually a very helpful tip! I don't really have the time (and cables or help) to try out all positions, but I do want to configure the setup for now. Starting with 1 placement might be a good way to start. Will help me figure out the other parts like delay, PEQ and integration. Will try that later this day!