Pros and cons of a PR Sub

[MightyM]

Not too long ago I completed a pair of Adire 281s for my mains and now I'm looking at building a subwoofer to pair them with to help fill out the lower end. The 281s do well down to about 50Hz, but right around there the bass starts to get a bit "thin".

At this point I'm looking at either a Shiva with an Adire PR-15 in a 95 liter enclosure, or a sealed Tempest in 122 liter enclosure (both based on the Adire designs). I can see what the trade-offs are between the two on paper, but I've never actually heard a sub that uses a PR before, so I'm not sure what I'd be getting myself into with it. I should say that I'm not looking for a sub that'll shake the foundations of my house, so mega high SPLs at low freqs isn't my primary goal. Since I use my 281s for music listening as well, I'm hoping for a sub with a nice, tight sound with a lot of "punch"; something that'll be very musical, but could also be used to back up my 281s on the occasional movie. I'm afraid I might be losing some detail and speed if I went down the Shiva + PR route vs. the Tempest, but I do like the nice, flat response curve of that configuration...

Any thoughts on what the "real-world" sonic tradeoffs on the two designs would be? Or perhaps there are other options I should consider over the Adire designs? I should say that I need to use a small enclosure (24"-ish) for a high WAF.

Thanks!

[ThomasW]

The primary benefit to PRs is being able to make the box physically smaller than a box containing a port.

The problem with PRs is that they have significantly higher Group Delay numbers as compared to other 'vented' designs. Now some people claim that GD is inaudible in low frequencies. That maybe true. However, a higher level of GD does impact the transient response, and therefore the sound quality suffers.

If having a large box isn't a big deal, then I recommend going with a flared port instead of PRs. Adire has some nice low "Q" ported designs for the Tempest, those are highly recommended.

[JonMarsh]

Hello Michael,

If you have a PC with a version of Excel from 2000 onwards, you could download Unibox from the FRD consortium or Christian's own site:

Unibox

Then you can model the designs you're considering, and take a look at the predicted transient response and group delay...


Or, you could save yourself some time, and take some advice from an old luddite, and stay away from PR based designs. PR's do allow you to produce a lot of very low frequency output in a small box, but their step response and group delay leaves much to be desired; this degrades the definition and transient response in the lower frequency registers (cutoff frequency to 1-1/2 octaves higher).

In order of fidelity, here's how sub designs rank:

1) Large IB sub, Qtc 0.4 to 0.5

2) Sealed box sub, with Linkwitz transform

3) Sealed box sub, no Linkwitz transform, Qtc 0.4 to 0.566 range

3) Ported alignment with no peaking and shallow roll off to Fb, (lower Q), ~ -3 to -6 dB at Fb.

4) Conventional ported alignment, maximally flat

5) High order ported sub alignments using EQ- Chebyschev or other alignments

5) PR subs.

Just one opinion, but an experienced one. Thomas and I even went through all the falderol when the Stryke HE-15 was introduced, and lots of folks insisted the Stryke cube with 3 PR's was the best thing since white bread- we built that, and our own AS-15 ported alignment with the HE-15, and compared for ourselves regarding sound quality, extension, etc.

Obvioiusly, there are tradeoffs with different sub designs with regards to cost, amplifier power, driver size, etc. You can't get around Hoffman's Iron law, as they say. If I was wanting deep bass extension with a single relatively inexpensive driver, and low power amp, I'd go for a ported alignment like the EBS alignment for the SHIVA, or like that I proposed for the Ascendant Audio Atlas 12 (an adjustable Q subwoofer driver with XBL2 technology licensed from Adire for low distortion- Adire Tumult has this, Shiva and Tempest don't). Tradeoff is requiring a large enclosure in either case.



You can get in the neighborhood of 110 dB in room at one meter with 300 watts or less.

If moderately high output with greatest fidelity in a small box was my goal, I'd go with a sealed driver(s) with a Linkwitz transform equalizer; with the larger throw 12's or 15's, one or two drivers can muster a lot of output with very good fidelity, tradeoff is higher driver cost and more powerful amplifier. Suitable drivers are the Tumult, the old BPD 03 series, and Ascendant Audio Avalanche series drives, especially the upcoming Avalanche 12. But in this approach requires 750-2,000 watts, depending on drivers and enclosure size.

[MightyM]

Thanks for the detailed response, Jon. Much appreciated.

I've read some of your other posts on the Linkwitz transform topic and had started doing some preliminary scouting on the web to find sub amps with LTs. Any advice on suitable sub amps that won't break the bank? I've been thinking plate amp, but should I open my mind a bit further to other options?

I've started seeing more and more about the Atlas 12, and that driver has started creeping onto my short list. I think I should look closer at this option... After reading through your brain dump, I think I'll scratch the Shiva + PR-15 configuration off my list and check out Ascendant's products as well.

I've got Unibox on my computer at home, but haven't had the time to check it out in too much detail yet. Having a 6 month old in the house tends to cut down on one's free-time quotient.

[ThomasW]

Look at the rather modest cost difference between 12" and 15". IMO it's a better investment to go with a 15", if you have the space for the somewhat larger cabinet.

This will net you added output and a bit lower distortion

[MightyM]

I noticed that myself as I was scouring Ascendant's site a moment ago. At this point I'm thinking the Atlas 15 would fit my needs pretty well.

[ThomasW]

IMO a 15" is a better longterm investment than a 12"

[MightyM]

Any idea what kind of power the Atlas 15 needs? I'm going to have to sacrifice some quality for price, so I'm considering amps in the $350 price range like the Hypex HS200 or one of the PE plate amps. It pains me to have to skimp on amplification :cry:, but that's all I've got to work with right now. Would the 200w HS200 have enough juice to drive the Atlas 15 in a small-ish (8 cu. ft.) sealed box?

[Steve Goff]

In order of fidelity, here's how sub designs rank:

1) Large IB sub, Qtc 0.4 to 0.5

2) Sealed box sub, with Linkwitz transform

3) Sealed box sub, no Linkwitz transform, Qtc 0.4 to 0.566 range

3) Ported alignment with no peaking and shallow roll off to Fb, (lower Q), ~ -3 to -6 dB at Fb.

4) Conventional ported alignment, maximally flat

5) High order ported sub alignments using EQ- Chebyschev or other alignments

5) PR subs.

One point: I think Mark Seaton has claimed, in connection with his company's B-DEAP (Boundary Dependent External Air Path) subwoofer, that his horn-loaded bass system actually has better transient response and group delay than an IB sub. Would you or your evel twin care to comment on this?

[ThomasW]

I discussed this with Jon, he's doubtful of the claim.

And let's keep things in perspective........

Entry level for a pretty good IB is under $500 and that includes an amp.

The current system Mark K is installing is a pair of SPL-B-DEAP32 horns, a couple Crown K2s, EQ, measurement software and Mark K or Chris Collins to do the install. Price of admission approx $12,000.

Now for a measley $6000, I can buy a dozen Avalanche 18"s, a Crown K2, a Behringer BFD, and build fairly nice IB ...:wink:

So guess which one I'd choose ....... :B

Micheal M

Do you really need a plate amp? There a lots of much bigger prosound amps if you can live with a cooling fan.

[Steve Goff]

Thomas,

Of course, I agree completely on the cost trade-off, and wouldn't even think of trying to buy a B-DEAP or sneak it into the house. I was just wondering about the technical claim.

[ThomasW]

sneak it into the house

Be sure to get video of that ..... :T

Jon will comment, he was waiting at the light rail station when I ask him about Mark's statement.

[jdybnis]

2) Sealed box sub, with Linkwitz transform
3) Sealed box sub, no Linkwitz transform, Qtc 0.4 to 0.566 range

That's interesting. Why would there be any difference between the two? If anything I would think the box without the Linkwitz transform would sound better, it would have less power compression.

[JonMarsh]

With no equalizer, the sealed box will have it's own natural roll off characteristic - transient response will still be quite good, but won't go as deep.

With the proper choice of drivers and box design and amplification, an LT sub can be made to go arbitrarily deep and flat, with critically damped transient response. Without a properly setup LT transform, it just won't be quite as accurate- in pressure amplitude response, at least, though it will be close. But that's why you have to rank.

Besides, it's just my opinion- my $0.02 if you will.... :wink:

[MightyM]

Micheal M

Do you really need a plate amp? There a lots of much bigger prosound amps if you can live with a cooling fan.

NEED? Nah, I suppose not. I'm leaning in that direction because I fear fan noise. 8O With the space I'm working with, I don't have anywhere to stash the amp except in (relatively) plain view. Plus, I've been running under the assumption that a pro amp would cost considerably more and be out of my price league. I'd be more than happy to be proven wrong, though! Any advice on good "bang for the buck" sub amps in the sub-$500 range?

[Jack Gilvey]

How about this: throw an Atlas 15 (coils in series for ~4-ohm Re/Qts = .395/500w Pe) into 125 liters lightly stuffed and use a Rythmic Audio A350 with the LT option (about $200 total with LT circuit pre-configured to your specs). According to Unibox, the 350 watts applied in this box matches the excursion quite closely below 30Hz, so it looks like a nice mating. The Rythmic amp also gives you the choice of three Qts values via the flick of a switch.

[ThomasW]

All the low buck prosound amps are fan cooled. The cheapest 'best buy' is the Nady XA900 for $180. Next step up is the Behringer EP-1500 for $300

[MightyM]

All the low buck prosound amps are fan cooled. The cheapest 'best buy' is the Nady XA900 for $180. Next step up is the Behringer EP-1500 for $300

Definitely food for thought. Again, all this expert advice is greatly appreciated. :T

...and use a Rythmic Audio A350 with the LT option...

I spotted the Rythmik amps last week but haven't seen any feedback on them anywhere yet, so I didn't give them too much thought. The price is right, but I have to wonder how they can make an amp packed with that many features so inexpensive? What are they giving up? Not to put down Rythmik _at all_, just my own ponderings...

[Jack Gilvey]

It's a standard sub plate amp, should work as well as just about any other within its power limits.

[Brian Bunge]

I've used it in several subs and it worked very well. Definitely worth the money.

[Andrew Pratt]

Fan noise is a REAL problem with the pro amps (other then convection cooled Crown's). Luckily I have mine in another room but even then I can hear it when all is quiet

[Dennis H]

You can quiet the fan with a resistor in series with the fan. We don't abuse our amps like a rock band does so it will still get plenty of cooling with the fan running at a slower (quieter) speed.

[Jack Gilvey]

If actually considering an Atlas, I'd ask Chad before throwing the output of a pro amp at them.

[JonMarsh]

I have to agree with Jack completely; the Atlas series, as well as the Avalanche, have above average sensitivity for long throw subs, by 2dB or so compared with the competition; the Atlas drivers just don't need over 300 watts in almost any conceivable application to hit their stride and give their all. They'll work quite well with a qualit plate amp (I know, that may be an oxymoron) that can deliver 250 to 350 watts at 8 ohms. More power is unncessary, and possibly dangerours- I recall the stories of some of the early guys getting Adire Tumults, and driving them with oversized amps, basically turning them up and up until they expected them to start sounding strained... well, with XBL2 technology, the distortion remains pretty low up to high drive levels... so the drivers gave out phsycially or thermally before sounding as hard worked as many other sub drivers do.

Just something to think about...

[MightyM]

The collected knowledge of this forum never ceases to amaze me. Thanks a ton for all the info. I'll be sure to report back when I get around to constructing my sub.

[Mark Seaton]

Many perspectives...

Since my name was mentioned in a few comments, and the discussion got a bit more general, I would just want to add a few considerations that may or may not be a factor in various applications.

Hello Michael,
Then you can model the designs you're considering, and take a look at the predicted transient response and group delay...

Or, you could save yourself some time, and take some advice from an old luddite, and stay away from PR based designs. PR's do allow you to produce a lot of very low frequency output in a small box, but their step response and group delay leaves much to be desired; this degrades the definition and transient response in the lower frequency registers (cutoff frequency to 1-1/2 octaves higher).

In general I agree that the group delay is higher with PR subs. This should be qualified though to specify cases of the same volume and tuning point. When comparing the same box volumes for a given driver the "knee" of the response will be higher Q and sharper with a PR vs. a port. The airflow through a port will serve to further damp the shape of the knee as output level is increased. Since group delay correlates directly to the shape of this knee, we see lower group delay numbers for a given box size with a port. I would have no arguements with the statement that it is "easier" to get an undesireable sounding response with a PR enclosure than with a port.

I generally don't consider PRs to be the best "bang for buck" solution, especially when you can build a huge box. If you limit your enclosure size and want very low extension, I would consider a PR'd enclosure to offer benefits over the ported, even possibly in group delay.

The benefit in passive radiators is the ability to tune low in a smaller box. Remembering that response shape is directly tied to group delay, a design yeilding an overdamped response and deeper low frequency extension can in fact have equal or better group delay curves. As an interesting example, I quickly modeled the Avalanche-15 in a 200L ported box, using a pair of 4" flared ports tuned to 17Hz and set input power to 50W into the 4 Ohm load. I was using LspCAD v5 for these comparisons, which has been very accurate for me in predictions of some recent projects. I then created another model with the Avalanche-15 in a passive radiator enclosure using the same input power.

Sure enough, with the same internal volume of 200L and a 17Hz tuning, the Q was higher near tuning, and the roll-off slightly steeper, resulting in increased group delay below say 25Hz. At the same time, the PR solution was more efficient around tuning. If we instead shrink the enclosure down to say 120L and lower the tuning to around 15Hz, we now see that in fact the passive radiator system has somewhat less group delay over the range of 30-13Hz.

All this really serves is to confirm that the shape of the response and roll off has a significant effect on group delay. In this particular case we also see that the passive radiator solution offers very comparable performance in an enclosure 60% the size of the ported cabinet, and that doesn't account for the volume of the ports. Of course any EQ or high pass filters applied will similarly alter the results.

While I have done some custom subs with big ports or ducts, my personal desire for high dynamic linearity have always left me wanting just a little more with all but the largest of boxes and ports. In the end, it all comes down to the parameters and resources you have to work within. In my case, many projects I work on allow me tol have a high resolution measurement system and a full featured EQ to contour things to taste. For most home theaters, my preference in covering the lowest octaves is to use multiple, high displacement sealed boxes or an IB followed by very low tuned passive radiator systems.

Just my take,