'Perceive v2.0' Construction Diary

[Audiophilenoob]

Originally posted by ShinOBIWAN



BTW Your getting robbed on those W26's over in the states. With quantity pricing I can get 6 x for £1000.

I'm missing where we're getting ripped

we get 6 here for $1500... or around 850 pounds

[ShinOBIWAN]

Originally posted by tktran But do you need 4 x 15, or 8x 10 or 12"? IMHO, when you select your woofer I think you should work backwards with what you need, then go off and do some modellling to various woofers to make a shortlist of what may hit your target.

I'd consider SPL requirements from seated listening position, then work backwards to find out what you need for SPL at 1m. Then I'd consider what I want for Vb and Qtc, and if I wanted to use existing power amplifiers I'd consider the available power on tap. The best device for the REQUIRED application is by far the best way to approach this.

Good advice and although I don't show my work here, I spend lots of time investigating the theoretics of certain layouts. So that gives a good estimation of figures if not sound.

I guess part of the high end appeal is trying things others haven't used, and it irks me that sometime the motivation behind driver choices are not based on engineering or design decisions, but more on what we perceive to be the best.

As is the case with the Tumult, great driver but it can certainly be bettered with additional thinking. As I've already stated I like the XBL2 sound but its not the end of the search.

Just wait and suggestions will keep rolling in, but only you know what you need. [/B]

I'm not in a position to buy yet anyway so I'm in no rush.

[ShinOBIWAN]

Originally posted by Audiophilenoob



I'm missing where we're getting ripped

we get 6 here for $1500... or around 850 pounds

I really should look at exchange rates before I put my foot in my mouth next time.

Damn the dollar is weak against the pound. It almost makes more sense to import if there wasn't duty and big delivery charges.

[thylantyr]

I haven't followed this thread, but have you defined the spacing
goals?

What is your max volume for the sub project? One or multiple
boxes? How big is the room? Why do you want 10hz?

How much is the DIY amp costing you vs. pro audio amps solutions?

Have you audiitioned any PR subs?

[thylantyr]

I seriously doubt the QSC would offer an improvement over the P101.

Sounds like a cool challenge. The 200w P101 vs. the 4000w QSC
or do you want to challenge the 9000w QSC?

[ShinOBIWAN]

Originally posted by thylantyr
I seriously doubt the QSC would offer an improvement over the P101.

Sounds like a cool challenge. The 200w P101 vs. the 4000w QSC
or do you want to challenge the 9000w QSC?

Ugh, not another one who judges amps on power output alone surely!

Doesn't SQ matter here? The QSC is OK but its hardly a Bryston or Krell is it? I've heard a QSC 2540 and Behringer EP2500 both were OK but not brilliant, I prefered the QSC though. My comments are here:

http://www.diyaudio.com/forums/showt...150#post613150

You think that the XLS or Volt drivers I'm looking at needs that much power? What about bridging P101's?

Sorry but it sounds needless to me. Fine if you've got subs rated to 1600w, I agree more power is needed than the P101 provides. However 200w into 8ohm or 380w into 4ohm is more than enough for the XLS, Volt etc.

Stuff like the Diamond are DVC so power could be bumped up to around 800w into 4ohm simply by using a dedicated amp for each coil. Or with bridging I'd be looking at 1600w into 4ohm.

[ShinOBIWAN]

Originally posted by thylantyr I haven't followed this thread, but have you defined the spacing goals?

First thing I did.

What is your max volume for the sub project? One or multiple
boxes? How big is the room? Why do you want 10hz?

I'd like around 100ltr NET max. Two boxes positioned next to the mains.

Room is 28ft x 22ft x 9ft/12ft(the roof slopes from 12ft to 9ft)

flat 15hz is overkill for music but for HT I'd say essential for the full effect and that's why. This is an AV setup.

How much is the DIY amp costing you vs. pro audio amps solutions?

The DIY amps are much cheaper to build since I've got half the bits already for the another 6 x P101's and I've got 6 x already built.
I've also got all the components for the Krell's, I'm just waiting on the PCB's.
The AKSA 55's are around $430AUS for a complete ready to go stereo pair.

Have you audiitioned any PR subs?

I have a Mackie HRS120 sub which features a PR as well as Servo feedback.

[thylantyr]

Ugh, not another one who judges amps on power output alone surely!

All differences aside, to drive subwoofers you need power.
Lets assume that someone has a magical SQ tube amp
rated at 10 watts. This amp will not even bring a beer to the
party when driving subwoofers whereas the QSC will provide
what the woofers need -> power.

I would not run the PL series full range as it's an H bridge output stage. I would run class AB outputs full range regardless if
QSC makes the amp or Krell. The modern amplifier is not
designed to perform bad whether it's pro audio or home audio.
You'd be surprised to know that there are many audiophile
type of pro audio people who are picky about SQ too. Don't
stereotype that industry as designing bad amplifiers. Audio
amplification design is easy and a mature industry.

I've heard a QSC 2540 and Behringer EP2500 both were OK but not brilliant, I prefered the QSC though. My comments are here:

I have a 2450 and I'd sell my Adcom home amp with no hesitation
to get another QSC. It's a great amp for the money. I would
prefer to use something bigger to drive a subwoofer array, the
4kw or 5kw amp, the PL series optimium. Also look at Crown.

I also know what the difference is between EP2500 and QSC
2450. The EP is over-rated in specs, less power in real life due to
the smaller transformer. But it cost 1/2 the price.

You think that the XLS or Volt drivers I'm looking at needs that much power? What about bridging P101's?

You need to operate one P101 per woofer. Lets assume you
need ten XLS woofers. You need ten P101 amplifiers to ensure
each woofer gets 200w. If you want more power, then you
have to bridge the P101's, so you now have 20 amplifiers to
deal with. How much will this cost vs. a big monster pro amp?
You don't have to limit yourself to just one proamp either,
get two or three -- haha.

I'm looking at the P101 design right now. This is a simple design,
not very robust with two transistors per rail. I took a similar
design over 15 years ago and modded the amp with more
output transistors, eight per rail and placed them on a massive
heatsink with a 1/4" aluminum base plate, T0-3 transistors.

I think I had 200,000uF of capacitance per rail and used 2KVA
transformer. I had to mount the predriver transistor on the heatsink because they ran hotter and I reduced amplifier gain
because it was to hissy for my liking. I installed a 5532 opamp
on the front end to feed the amplifier a large voltage to drive it to
full power to offset the reduced gain. The net result was a less
hissy amp, you could place your ear on the tweeter and not hear
hiss. /sweet.

I drove 1/2 ohm per channel playing test tones with this amp on
two house circuits. The point is, that P101 has mod potential
because it also uses lateral mosfets. You can parallel a few more
transistors on each rail without matching transistors like you do
with vertical mosfet... so you don't have to buy a ton of transistors to find matches.

Something more powerful than P101 is the Holton amplifier,
AV800, actually the modded AV800 1KW.
http://www.aussieamplifiers.com/av800.htm

10 transistor per rail for the 1kw module.

This is an advanced project as the documentation had errors,
I think I trapped all the errors and drew up my own set of
docs, but I haven't verified it. I notice Holton is changing his
business model, no longer offering plans on the website. Lucky
me I have them - harr harr harr.

For subwoofer only, it makes you wonder if it's even worth
making a class AB amp when there is 'digital' amplifier options
out there.

I worked on a PCB layout project taking the AV800-1KW design
and adding extra 'stuff'. I was my pet project for a while and it
has 14 transistors per rail on all on PCB.

Problem is. My layout is 4 layers, a big board and expensive to
fab. It would cost me $2k just to get a batch of PCB's made.
Another $2k-$3k to get a big chunk of awesome heatsink extrusion, the really good stuff. Factor in machining a chassis
to make the result look 'store bought' and the price tag rises
alot.

... then when I price a QSC 4kw amp for $1000, then why do I
want to mess around with DIY amps ? Prices of pro audio have
dropped, quality improved, and better designs can be had for
cheap. DIY amps are more of a novelty now, for the thrill of
making it not because there is huge savings like DIY loudspeakers.

If you can't afford an array of Tumult and the big QSC or Crown
amp then you have to settle for less and not have the ultimate.
Everyone wants to have the uber setup but it cost money unless
you steal.

The only way to keep your project cost down is to do what you
said. A stock P101 per woofer, or per coil, or bridged. It's
a good solution but you'll have alot of amplifier clutter when
all you may need is one or two big monster pro amps and they
are more feature rich than a bare bones P101. Clipping LED's,
protection circuits, etc.

[thylantyr]

Perhaps I don't understand, you want to the subwoofer
box to be only 3.5 cubic feet of volume per sub tower?

Two subwoofer towers, each next to the mains looks nice
so that is the strategy vs. using one bigger box elsewhere in the house?

[ShinOBIWAN]

Originally posted by thylantyr
Ugh, not another one who judges amps on power output alone surely!

All differences aside, to drive subwoofers you need power.
Lets assume that someone has a magical SQ tube amp
rated at 10 watts. This amp will not even bring a beer to the
party when driving subwoofers whereas the QSC will provide
what the woofers need -> power.

I would not run the PL series full range as it's an H bridge output stage. I would run class AB outputs full range regardless if
QSC makes the amp or Krell. The modern amplifier is not
designed to perform bad whether it's pro audio or home audio.
You'd be surprised to know that there are many audiophile
type of pro audio people who are picky about SQ too. Don't
stereotype that industry as designing bad amplifiers. Audio
amplification design is easy and a mature industry.

I've heard a QSC 2540 and Behringer EP2500 both were OK but not brilliant, I prefered the QSC though. My comments are here:

I have a 2450 and I'd sell my Adcom home amp with no hesitation
to get another QSC. It's a great amp for the money. I would
prefer to use something bigger to drive a subwoofer array, the
4kw or 5kw amp, the PL series optimium. Also look at Crown.

I also know what the difference is between EP2500 and QSC
2450. The EP is over-rated in specs, less power in real life due to
the smaller transformer. But it cost 1/2 the price.

You think that the XLS or Volt drivers I'm looking at needs that much power? What about bridging P101's?

You need to operate one P101 per woofer. Lets assume you
need ten XLS woofers. You need ten P101 amplifiers to ensure
each woofer gets 200w. If you want more power, then you
have to bridge the P101's, so you now have 20 amplifiers to
deal with. How much will this cost vs. a big monster pro amp?
You don't have to limit yourself to just one proamp either,
get two or three -- haha.

I'm looking at the P101 design right now. This is a simple design,
not very robust with two transistors per rail. I took a similar
design over 15 years ago and modded the amp with more
output transistors, eight per rail and placed them on a massive
heatsink with a 1/4" aluminum base plate, T0-3 transistors.

I think I had 200,000uF of capacitance per rail and used 2KVA
transformer. I had to mount the predriver transistor on the heatsink because they ran hotter and I reduced amplifier gain
because it was to hissy for my liking. I installed a 5532 opamp
on the front end to feed the amplifier a large voltage to drive it to
full power to offset the reduced gain. The net result was a less
hissy amp, you could place your ear on the tweeter and not hear
hiss. /sweet.

I drove 1/2 ohm per channel playing test tones with this amp on
two house circuits. The point is, that P101 has mod potential
because it also uses lateral mosfets. You can parallel a few more
transistors on each rail without matching transistors like you do
with vertical mosfet... so you don't have to buy a ton of transistors to find matches.

Something more powerful than P101 is the Holton amplifier,
AV800, actually the modded AV800 1KW.
http://www.aussieamplifiers.com/av800.htm

10 transistor per rail for the 1kw module.

This is an advanced project as the documentation had errors,
I think I trapped all the errors and drew up my own set of
docs, but I haven't verified it. I notice Holton is changing his
business model, no longer offering plans on the website. Lucky
me I have them - harr harr harr.

For subwoofer only, it makes you wonder if it's even worth
making a class AB amp when there is 'digital' amplifier options
out there.

I worked on a PCB layout project taking the AV800-1KW design
and adding extra 'stuff'. I was my pet project for a while and it
has 14 transistors per rail on all on PCB.

Problem is. My layout is 4 layers, a big board and expensive to
fab. It would cost me $2k just to get a batch of PCB's made.
Another $2k-$3k to get a big chunk of awesome heatsink extrusion, the really good stuff. Factor in machining a chassis
to make the result look 'store bought' and the price tag rises
alot.

... then when I price a QSC 4kw amp for $1000, then why do I
want to mess around with DIY amps ? Prices of pro audio have
dropped, quality improved, and better designs can be had for
cheap. DIY amps are more of a novelty now, for the thrill of
making it not because there is huge savings like DIY loudspeakers.

If you can't afford an array of Tumult and the big QSC or Crown
amp then you have to settle for less and not have the ultimate.
Everyone wants to have the uber setup but it cost money unless
you steal.

The only way to keep your project cost down is to do what you
said. A stock P101 per woofer, or per coil, or bridged. It's
a good solution but you'll have alot of amplifier clutter when
all you may need is one or two big monster pro amps and they
are more feature rich than a bare bones P101. Clipping LED's,
protection circuits, etc.

All valid points but I can't help feeling that much of it doesn't apply to me.

I haven't even decided on drivers yet but you almost assume massive power requirements that in-effecient subs such as Tumult require to get them to even f@rt.

You also assume that building DIY amps is expensive. Fact is I already have 6 x P101's and just need the traffo, PCB's and PSU caps for another 6. I work that out at around £200.

Casework is not needed since I plan to install the amps into the cabinet for the subs.

Project cost are important but since I've spent over $11,000 already including the PC XO, I'll do whatever else it takes. With the P101 I know they sound great, probably wasted just for subs if that's possible. I get to try these before I even think of buying a commercial amps simply because they are already there and are very compentant. The problem with the QSC or any amp is I'll need rackspace also, with the panels or PCB's I can mount to the subs without dissection.

There's a few other options too:

http://www.bkelec.com/Modules/bsbpv500.htm

http://www.bkelec.com/Modules/mf1000.htm