Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Matt]

I have two unused channels on my power amplifier.  Together, at 4 ohms, they can provide 370 watts RMS. 

I'm trying to figure out how to use this power for a nice passive, sealed, subwoofer (as opposed to buying a $300 plate amplifier for the subwoofer with the same amount of power).

I currently have six RCA cables coming out of a computer.  5 go to the power amp to drive speakers.  The 6th RCA cable goes to a powered subwoofer.

I was thinking I could Y the subwoofer RCA and hook it to two channels on the power amplifier.  From the amplifier, I could connect each subwoofer channel to a voice coil on a dual-voice coil 4 ohm subwoofer.

But I'm concerned that the output level will be too low on the subwoofer with this approach.  I might be able to get around this by adjusting the subwoofer level in Media Center, but I don't want to have to turn all the other channels down a huge amount.

Does this idea even make sense?  Any tips?

Thanks.

[jmone]

There is only one way to find out!  Why don't you try it with one of you speakers, just swap say the RH for LFE to get an idea.  As you suggest, balancing the pressure level may be fun (most active subs have their own volume control for this reason).

Did I mention that you just fell into the "I think I'll build my own" speaker trap?

[Matt]

There is only one way to find out!  Why don't you try it with one of you speakers, just swap say the RH for LFE to get an idea.  As you suggest, balancing the pressure level may be fun (most active subs have their own volume control for this reason).

You talked me into it.

I think I need a dual-voice coil 4-ohm driver before I'll know what the volume will be.

I'm going to try one of these (same Fs and Xmax of expensive subs, but with a relatively high sensitivity):
http://www.amazon.com/Infinity-Reference-1262w-1200-watt-High-Performance/dp/B0028AYIXK/ref=sr_1_1?ie=UTF8&s=automotive&qid=1305345401&sr=8-1

Did I mention that you just fell into the "I think I'll build my own" speaker trap?

The red speaker ball you posted inspired me:
http://yabb.jriver.com/interact/index.php?topic=63965

[jmone]

I was trying to talk you OUT of doing the home made thing not INSPIRE you.  To complete the insanity how about adding home made electrostatic speakers to the top?  

[Matt]

Got the sub today.  Frankensteined it into my existing subwoofer cabinet with speaker wires running out the port.  Hooked it up to the power amp.

It screams.  I'm having fun.

And now that I know it works, I just need a nice enclosure.  Maybe jmone's idea of a red bouncy ball?

[jmone]

Any photos yet?

[mojave]

Below is a WinISD Pro simulation of the sub in a 1.5 cubic ft enclosure. You need a small enclosure like this to prevent exceeding the X-Max of the driver. In order to extend the lower frequencies, you can add a Linkwitz Transform. If you look at the transfer function of the L-T you can see that it acts as a low shelf filter. However, since the sub can't play louder thab its specs allow, the L-T actually is like a high shelf filter that is pulling down the top end. The attachment with two SPL plots helps illustrate this.

It would be nice if the Linkwitz Transform and low/high shelf filters were added to the Parametric EQ. You can use the L-T spreadsheet on the page I linked to above to figure the calculations for the L-T.

[Matt]

I haven't built anything yet.  I'm trying to decide if I want front firing or down firing.  The sub is on carpet that's on a cement slab, and I'm leaning towards down firing so I don't need a grill.

Below is a WinISD Pro simulation of the sub in a 1.5 cubic ft enclosure. You need a small enclosure like this to prevent exceeding the X-Max of the driver.

Interesting.  I was planning on a 2.0 cubic ft enclosure, since this was the sweet spot given the parameters of the driver.

In order to extend the lower frequencies, you can add a Linkwitz Transform. If you look at the transfer function of the L-T you can see that it acts as a low shelf filter. However, since the sub can't play louder thab its specs allow, the L-T actually is like a high shelf filter that is pulling down the top end. The attachment with two SPL plots helps illustrate this.

It would be nice if the Linkwitz Transform and low/high shelf filters were added to the Parametric EQ. You can use the L-T spreadsheet on the page I linked to above to figure the calculations for the L-T.

Do you know of any technical implementations?  Like C-code or IIR coefficients?

Thanks.

[mojave]

Interesting.  I was planning on a 2.0 cubic ft enclosure, since this was the sweet spot given the parameters of the driver.

If you aren't playing low stuff it won't make much of a difference. It is only below 32 Hz that the smaller box size helps control excursion. See attachment.

Do you know of any technical implementations?  Like C-code or IIR coefficients?

I have no idea. John, the developer of REW, could probably help you. REW can export the L-R circuit parameters to the miniDSP.

[Matt]

Thanks for the graphs.

I was playing with WinISD, and I'm not sure I understand why you'd shrink a box to control excursion.

For example, to get 99 dB at 20 Hz requires 470W with a smaller box and 300W with the ideal box.  But in both cases, the cone excursion is the same.

In other words, the smaller box does limit excursion, but only because it effectively reduces the decibel levels at low frequencies.  As soon as you use a Linkwitz (or whatever) to flatten the curve, the excursion is right back to the same spot as with a bigger box.  But you need more power because the box isn't as efficient when it's not the ideal size.

Does that make sense, or am I missing something?

[mojave]

I was playing with WinISD, and I'm not sure I understand why you'd shrink a box to control excursion.

For example, to get 99 dB at 20 Hz requires 470W with a smaller box and 300W with the ideal box.  But in both cases, the cone excursion is the same.

Drivers have both thermal and excursion limits. The Infinity 1262W is rated at 300 Watts. More power than that can burn up the voice coils depending on duration. If you are designing around your driver's parameters, you ask yourself, "What size box keeps me within the thermal and excursion limits?" Since 300 Watts is your thermal limit, you enter that into WinISD. Then you shrink your box size until you are within the excursion limits of the lowest frequencies you want to play. This means staying within X-Max at that frequency.

By limiting excursion on the bottom end with a smaller box, you are actually losing efficiency everywhere. This is why a while back I asked you for a limiter function for the parametric EQ. It is better, IMO, to have the larger box for greater efficiency. If you look at my previous graph you see that limiting the maximum output below 30 Hz would help prevent exceeding X-Max. Pro Amps with DSP have the ability to set limiters, but I've never used these type of amps.

WinISD shows you the driver's impedence and the Amplifiers apparent load power. These can be helpful for setting the limits. I have a driver (actually 8 arrived a week ago) that can handle 500 watts, but reaches X-Max at 40 Hz with that much power. I would eventually like to use 500 watts per driver and be able to reach high SPL's above 40 Hz, but not exceed X-Max at frequencies below 40 Hz. The driver only needs 100 watts to reach X-Max at 20 Hz. I want to maintain a flat frequency response until I use more than 100 watts. I might have music that never has material below 40 Hz, so the limiter would not even be needed in those cases. Even with low frequency material, I might not not be playing at an SPL that requires the limiter. This means it would have to take into account peak signal levels and overall volume level. If I use only a 225 watt amp for all 8 drivers, I can stay within X-Max down below 5 Hz. However, I give up 12 dB of capability above 40 Hz (118 dB vs 130 dB  ). I am hoping to eventually have 130 dB capability from all channels.

But you need more power because the box isn't as efficient when it's not the ideal size.

But you either don't have more power (amp limited) or your driver can't handle more power (driver limited) so you settle for less efficiency and greater extension or greater efficency and less extension.

[Matt]

Thanks for the response.

By limiting excursion on the bottom end with a smaller box, you are actually losing efficiency everywhere. This is why a while back I asked you for a limiter function for the parametric EQ. It is better, IMO, to have the larger box for greater efficiency.

So are you saying that it might be best to make the most efficient sized box (~2.4 cubic feet) and deal with limiting in software if it's needed? 

This makes the most sense to me because it's easy to turn the volume down in software and hard to add volume in hardware.

These can be helpful for setting the limits. I have a driver (actually 8 arrived a week ago) that can handle 500 watts, but reaches X-Max at 40 Hz with that much power. I would eventually like to use 500 watts per driver and be able to reach high SPL's above 40 Hz, but not exceed X-Max at frequencies below 40 Hz. The driver only needs 100 watts to reach X-Max at 20 Hz. I want to maintain a flat frequency response until I use more than 100 watts. I might have music that never has material below 40 Hz, so the limiter would not even be needed in those cases. Even with low frequency material, I might not not be playing at an SPL that requires the limiter. This means it would have to take into account peak signal levels and overall volume level. If I use only a 225 watt amp for all 8 drivers, I can stay within X-Max down below 5 Hz. However, I give up 12 dB of capability above 40 Hz (118 dB vs 130 dB  ). I am hoping to eventually have 130 dB capability from all channels.
But you either don't have more power (amp limited) or your driver can't handle more power (driver limited) so you settle for less efficiency and greater extension or greater efficency and less extension.

Wow.  Is this for your house?

One good 12" can make me physically sick after a few minutes with the volume turned up.  Granted it's a fun sick, like you get from riding a roller coaster, but it doesn't leave me wishing I had 7 more subwoofers!

[mojave]

I have a crawl space under my living room and want to make an infinite baffle subwoofer. It takes more drivers to do this, but there isn't much visible in the room. If you don't know what an IB is, it is essentially a huge sealed sub with one side of the drivers in your room and the other side in another room. I plan to build two manifolds that will each hold 4 drivers. These will be mounted under the floor joists and there will be an opening in the floor covered with a metal grill. Here are some pictures that show what I want to do.
 

[Matt]

It would be nice if the Linkwitz Transform was added to the Parametric EQ.

I'm a little worried that I'm falling deeper into a rabbit hole.

But I added Linkwitz support, coming in 16.0.112 and later.

[mojave]

Wow!

Did you end up having to adjust the subwoofer level in MC to have the right output level with your Emotiva amp? I recently purchased an 800 watt (4 ohm) monoblock amp to power my infinite baffle system. It has no gain control so I'll have to do the same thing.

[Matt]

Did you end up having to adjust the subwoofer level in MC to have the right output level with your Emotiva amp? I recently purchased an 800 watt (4 ohm) monoblock amp to power my infinite baffle system. It has no gain control so I'll have to do the same thing.

Yes.  The level on the subwoofer in DSP Studio > Room Correction is about 15 dB higher than the other channels.

And then Parametric Equalizer splits the subwoofer to the RL and RR outputs on the soundcard for bi-amping.

I don't have as much total volume as before since the main speakers are down 15 dB at full volume, but it's still louder than you'd ever want.

You'll have to let us know how you fare with the monoblock amp.