Is SVS PB2000 an upgrade to these DIY subs?

[Sylar]

I can try to model something quick and dirty with WinISD tomorrow if you wish.
.

Thanks for the offer. This will be a future project in the coming months. So ill put this on hold for now.
For now my current subs will do, with the MiniDSP.
Yea, will create a new thread for the rattling drivers. Maybe in diyaudio or avsForum or here.

 

[Sylar]

Coming back to this...

If you sit closer to your rear sub than the front one, add 1ms delay to the rear per foot difference. For example, if your rear sub is 6 feet away and your front sub is 11 feet, set the delay on your rear sub to 5ms within your minidsp. I think that's how it works anyway. It's been a while since I messed with time aligning my subs.

Both my subs will be at the back. One left and one right. Pretty much equidistant from listener (+/- 1-2 ft).
I guess I can ignore time aligning my subs?

Anyway, having your subs time and volume aligned

Whats Volume alignment?

If your receiver has Audyssey XT32 with SubEQ HT, then it might make more sense to run Audyssey before correcting with miniDSP. But since Audyssey without SubEQ HT only provides LFE correction for a single receiver output, you'll get the best results from Audyssey if you get your subs as unified as possible with miniDSP first.

I have Marantz SR7005. I don't think it has Audyssey with SubEQ HT.

 

[rojo]

Volume alignment just means both your subs play at roughly the same volume. If your subs are equidistant from your primary listening position and each has its gain set at the same level, then you shouldn't have anything to worry about there.

 

[TLS Guy]

TLS, check my work please.

The CSS SDX12 (cheaper by the pair) + the APR12 passive radiator fully weighted with 900g of ballast appears to model like this @100W in OP's 3ft³ boxes:

Group delay is below 12ms till about 20Hz (below which is rises exponentially), and the driver's maximum excursion is never reached until over 900W at 20Hz, which would be about 106dB.

My apologies for the tardy response. I had make a trip to St. Louis for a high school robotics competition. I came back to spring work. Our longtime neurosurgeon in Grand Forks died last week. He had been retired since 2009. His funeral was Monday. One could not have wished for a better friend and colleague.

Anyhow today is a rainy day, thank goodness, as the fire condition have been extreme. We have been under red flag warnings day after day.

In my view you have added far too much mass to the ABR and ended up with a poor sounding extended bass alignment, like so many commercial designs.

It needs 6 of the 45 GM rings and not 20. This gives a much smoother response below 100 Hz. It has as much output as yours at 20 Hz and more above 20 Hz. You can tell my alignment is correct as there is a nice sharp null in the impedance response curve right at driver Fs.

I used a more sensible drive power of 250 watts. No 12" driver will take 900 watts no matter what the manufacturer says.

This will actually be a potent nice sounding sub.

The design file is attached.

 

[rojo]

In my view you have added far too much mass to the ABR and ended up with a poor sounding extended bass alignment, like so many commercial designs.

Yeah, I recalculated in post #17 and figured 540g of ballast would be better. I think our ideas are closer now, although I went with a bit more extension to move the rise in group delay down to subsonics. Your model's group delay climbs sharply at 25Hz.

What's the significance of the impedance dip at the driver's Fs? Does the model I made indicate poor performance because the dip is another few Hz below Fs?

Anyway, last I saw, OP was waiting for MiniDSP to get back with him about a plugin he purchased long ago and needs to re-download (post #37). Otherwise, repositioning his subs seemed to help with overall flatness, just not the output he wants. I doubt inserting the MiniDSP will do much to boost output though. Indeed, I reckon the opposite will happen.

 

[TLS Guy]

Yeah, I recalculated in post #17 and figured 540g of ballast would be better. I think our ideas are closer now, although I went with a bit more extension to move the rise in group delay down to subsonics. Your model's group delay climbs sharply at 25Hz.

What's the significance of the impedance dip at the driver's Fs? Does the model I made indicate poor performance because the dip is another few Hz below Fs?

Anyway, last I saw, OP was waiting for MiniDSP to get back with him about a plugin he purchased long ago and needs to re-download (post #37). Otherwise, repositioning his subs seemed to help with overall flatness, just not the output he wants. I doubt inserting the MiniDSP will do much to boost output though. Indeed, I reckon the opposite will happen.

The alignment with the impedance null at Fs and two nice peaks either side, indicate optimal driver control and in general will have the best bass. In addition you generally get the best sounding bass when F3 is a few Hz above Fs. I'm always suspicious if F3 is at Fs and really suspicious if it is below.

I'm not bothered by group delay below tuning, that will happen. The important thing is to get tuning for the best bass response. That happens when Fs is controlled properly. If you look at my design, you will note a cone displacement null right at driver Fs showing optimal loading and control of the driver.

 

[rojo]

Cool. Well, if the OP goes for my suggestion of replacement driver + PR, he can try it both ways and see which sounds better without having to buy anything else. He'll have variable tuning.

 

[Sylar]

I'm back!
I finally got the plugin. Used the REW Equalize section and selected MiniDSP.

Used the following settings in REW.
Speaker type - None
Individual Max Boost - 9 dB
Max Overall Boost - 6 dB
Flatness Target +/-3 dB

Have attached images of what the corrections look like as simulated by REW after selecting MiniDSP as the Equilizer. I ignored the phase switched measurements. Every two images are the same measurement, one with the initial measurement and one without.
(Dark Green is my earlier measured response and Light color is the predicted response applying MiniDSP equalization)

Wonder how much more rectification would occur after running Linkwitz Transform & Audyssey over this.

 

[Sylar]

Cool. Well, if the OP goes for my suggestion of replacement driver + PR, he can try it both ways and see which sounds better without having to buy anything else. He'll have variable tuning.

TLSGuy also suggested the same CSS SDX12 driver in the current Box (I guess with a passive radiator. Cant leave it open ).
I have to check out if I get that driver back here (India). A quick search did not give me much.

I noticed while running REW and MniDSP equilization that, it barely attempted to overcome the nulls. Corrections happened otherwise. That's interesting. So attempting to rectify using an inverse response with corrections of over 10-12 dB doesn't look like it is even going to attempt it. Especially when the null's have sharp Q's.

 

[rojo]

To flatten those nulls you should lower the target level before calculating the corrections. That way there will be more cuts than boosts.

If you email CSS they'll probably respond within a day or so. Bob and Jim are good with communications and answering questions.

Are you sure we can't tempt you into considering 18" subs? If more output is what you're looking for, larger subs is the most obvious way to get it.

Sent from my LG-VS980 using Forum Fiend v1.3.1.

 

[Sylar]

If you email CSS they'll probably respond within a day or so. Bob and Jim are good with communications and answering questions.

Will check this out. Maybe come back to this later.
How do the Dayton 12/15" Reference compare to the SDX12? We have a Dayton distributor back here.

Are you sure we can't tempt you into considering 18" subs? If more output is what you're looking for, larger subs is the most obvious way to get it.

With you guys around..
"When did you get tempted" seems to be a more accurate question .

I am pondering between upgrading to either of 2 15"'s or 2 18's. Sealed Dayton Reference is on top of the list as of now, unless I get better suggestions here. All of my equipment was suggested by you guys here, and no regrets so far .

Upgrade is coming, but not in the next 6 months. Have to live with my current subs until then.

To flatten those nulls you should lower the target level before calculating the corrections. That way there will be more cuts than boosts.

In my previous post, In the 6th image (BL-BR), there was a 5dB drop around 54Hz with a wide Q. Out of the 6 filters, one filter was being used at 110Hz. I decided to free that one up and modified filter range to 10-95 (from 10-120). I also reduced the target level by 3dB. I was able to fix the 5dB attenuation around 54Hz.
However, the massive nulls at 73Hz, 93Hz and 85Hz doesn't seem to get any better without messing up the rest of the response.
I also attempted dropping target level by 10/15/20 dB, didn't help much.
I modified the Target level by around 3dB and filter range 95/100Hz and here's what I got (Image #1,2,3 @ FR-BR & 4th @ BL-BR).

 

[TLS Guy]

Will check this out. Maybe come back to this later.
How do the Dayton 12/15" Reference compare to the SDX12? We have a Dayton distributor back here.


With you guys around..
"When did you get tempted" seems to be a more accurate question .

I am pondering between upgrading to either of 2 15"'s or 2 18's. Sealed Dayton Reference is on top of the list as of now, unless I get better suggestions here. All of my equipment was suggested by you guys here, and no regrets so far .

Upgrade is coming, but not in the next 6 months. Have to live with my current subs until then.


In my previous post, In the 6th image (BL-BR), there was a 5dB drop around 54Hz with a wide Q. Out of the 6 filters, one filter was being used at 110Hz. I decided to free that one up and modified filter range to 10-95 (from 10-120). I also reduced the target level by 3dB. I was able to fix the 5dB attenuation around 54Hz.
However, the massive nulls at 73Hz, 93Hz and 85Hz doesn't seem to get any better without messing up the rest of the response.
I also attempted dropping target level by 10/15/20 dB, didn't help much.
I modified the Target level by around 3dB and filter range 95/100Hz and here's what I got (Image #1,2,3 @ FR-BR & 4th @ BL-BR).

I would not worry about those narrow nulls. You will not hear those. Where those graphs made with the mains operating or the sub alone? If the mains were operating interference in the crossover range is a strong possibility.

 

[Sylar]

I would not worry about those narrow nulls. You will not hear those. Where those graphs made with the mains operating or the sub alone? If the mains were operating interference in the crossover range is a strong possibility.

Good to know that the narrow nulls would not make much of a difference.

Funny that you asked about the crossover, I was thinking about this last night and realized my crossover settings did not cut out my mains. If I remember it right, those measurements were made with the crossover at 40Hz. I have modified my crossover a few times since I took those measurements. So kinda lost track.
Either way this was my test run attempting to understand the whole process which is all new to me.

 

[Sylar]

Where those graphs made with the mains operating or the sub alone? If the mains were operating interference in the crossover range is a strong possibility.

I re-ran the measurements.
6 Positions. Polarity Normal & Reversed.
Crossover at 100Hz.
Audyssey Off.

Got some really good results. In a couple of positions am able to get a pretty flat response upto 80-90Hz after running MiniDSP EQ in REW.
Will go through them more carefully, play around a bit in REW and post the results tomorrow. I think the miniDSP will be up and running in couple of days

 

[Sylar]

Remeasuring with crossover set @ 100Hz got me good results. Thanks to TLSGuy for bringing this up. I would have probably forgotten this and continued .

Which position do I go with? FL-BL & FL-BR both look good.
You had earlier mention cutting is easier than boosting while attempting to get a flat response. Now that I've got a pretty flat response, is there an option to like boost by 5-15 dB? Is this a task that is delegated to the amp?

Just wondering, during measurement maybe I should set the crossover much higher like 250Hz, to minimize mains influence even further (Will probably do this finally when I decide where the subs will be placed).

 

[rojo]

The 3rd and 4th graphs look the best. Not sure which configurations those are, but all of them look promising really.

You know, to measure your subs with REW while minimizing your mains' impact on the measurements, you could connect your pc directly to your sub amp, bypassing your avr entirely. Or, did you ever get ASIO working?

Sylar said:
...is there an option to like boost by 5-15 dB? Is this a task that is delegated to the amp?

First, increase the volume of the lfe channel on your receiver. If you've got it +9 or higher already, then I'd start turning the gain up on the amp. You could also turn down the volume of the lcr and surrounds, the turn the main system volume up. Boom. Relativity.

I'm curious. At what level does your receiver default the volume of your subwoofer channel when you rerun Audyssey setup? If Audyssey setup suggests an initial level of -12dB, you might already have your sub amp turned up too high.

Sent from my LG-VS980 using Forum Fiend v1.3.1.

 

[Sylar]

My first attempt running MiniDSP and I get a fairly loud hum. Scratched my head a bit. I kept the miniDSP on top of the crown amp. The hum simply wouldn't go away.
Finally Decided to measure, and ran REW with RTA. There was a spike at 50Hz. Input power is at 50Hz. Switched off the sub and it was gone.

How do I get rid of this?
Ground the DSP to the power supply ground connection?
Can I somehow ground this to the CrownAmp? I'd prefer this.

You know, to measure your subs with REW while minimizing your mains' impact on the measurements, you could connect your pc directly to your sub amp, bypassing your avr entirely. Or, did you ever get ASIO working?

I could never get that darn thing to work. Lost count of attempts.

First, increase the volume of the lfe channel on your receiver. If you've got it +9 or higher already, then I'd start turning the gain up on the amp. You could also turn down the volume of the lcr and surrounds, the turn the main system volume up. Boom. Relativity.

Will try this.

I'm curious. At what level does your receiver default the volume of your subwoofer channel when you rerun Audyssey setup? If Audyssey setup suggests an initial level of -12dB, you might already have your sub amp turned up too high.

I haven't run the MiniDSP followed by Aydyssey yet.
Once I run them will come back to this.

I usually keep the sub am at 50% and the receiver at -20dB while measuring.

 

[rojo]

Sylar said:
My first attempt running MiniDSP and I get a fairly loud hum. Scratched my head a bit. I kept the miniDSP on top of the crown amp. The hum simply wouldn't go away.
Finally Decided to measure, and ran REW with RTA. There was a spike at 50Hz. Input power is at 50Hz. Switched off the sub and it was gone.

How do I get rid of this?
Ground the DSP to the power supply ground connection?
Can I somehow ground this to the CrownAmp? I'd prefer this.

Ah, ground loop. The best way to get rid of this is to have all devices in the signal chain -- PC, AVR, MiniDSP, and sub amp, all plugged into the same power circuit. You might simply experiment with plugging into different outlets and see what happens. You could also try a ground loop isolator inline with your rca cables, but you'll make the purists cringe.

Sent from my LG-VS980 using Forum Fiend v1.3.1.

 

[Sylar]

Ah, ground loop. The best way to get rid of this is to have all devices in the signal chain -- PC, AVR, MiniDSP, and sub amp, all plugged into the same power circuit. You might simply experiment with plugging into different outlets and see what happens. You could also try a ground loop isolator inline with your rca cables, but you'll make the purists cringe.

I was able to get rid of the ground loop hum. This link helped me understand the issue.
http://www.rane.com/note110.html

I fixed it by wiring the MiniDSP differently. It has 3 wrings on input and output. '+', '-' & 'S' (S -> Shield wire). On the Input side I had to short '-' & 'S'. On the Output I had to use '+' & 'S' (Not '+' and '-').
This is for unbalanced config only! MiniDSP manual has more info on this. Pg 16 & 22.
http://www.minidsp.com/images/documents/miniDSP Balanced 2x4 and Kit - User Manual v2.2.pdf

With this my problem went away. I have not touched the AVR or even the CrownXLS, I am still using the same RCA coming from AVR to DSP (which was connected to Crown earlier), and now RCA to Crown comes from the DSP.

Earlier, Crown had no issues being connected to AVR directly. So I'm thinking, Crown RCA '-' must be rigged to signal ground, and the AVR has a common signal ground to all its '-' connections.
Connecting AVR RCA '-' to DSP '-' which is shorted to 'S' in the input side. On the DSP Output side, 'S' connected to Crown RCA '-' obtained a common signal ground between these multiple devices!

 

[Sylar]

Attempted correcting using the miniDSP. It didn't do what I was expecting it to do. There are nulls even after correcting.
Did I miss out something? Could there be something wrong with my corrections file for the Umm-6?