End Game DIY Loudspeakers

[Draki]

Hi Draki you are right. The db scale is a bit wide, but this is not s SPL measurement it's a frequency respose measurement from the MLP.

I really don't want to sound as patronizing, really, but the FR measurement consists of "Hz" and "dB" coordinates. The ordinate shows the SPL expressed in a dB unit.
Hence it is a SPL measurement over the frequency, regardless of the point in space it was taken at.

 

[Jukka]

I think going back to the title of the thread, you've got to face some uncomfortable truths here. I'm not real hifi cognoscenti btw - just on the back of 40 years of buying it, and a few DIY speaker builds along the way.
Just look at the pictures in this thread. There's a thousand ways and more to get sound waves out of a box. Your first mission is to choose even a rough general idea - 2 way, 3 way, 7, horn, array, passive, DSP.... yada yada yada... Many of those you won't even have heard in something close to your listening setup. If you buy expensive, retail speakers you'll almost certainly get the chance to audition them with your kit, and probably in your listening room too. So you're not *completely* in the dark. If you DIY - yes, you'll save a bit of money (or more accurately buy some time from yourself ), but you have to accept that you're taking an expensive, hopefully educated, but a gamble nonetheless. Because the chance that you can hear the set you're interested in is minimal, even assuming you're going with a published design. The chance of hearing them with your kit in your room is, let's say zero to keep it simple. So on the basis of other peoples' opinions on maybe similar designs, but in different circumstances with different kit, you still have to be prepared to lay down 4 figure sums in kit and raw materials without any real clear idea what'll happen at the end. And if you can embrace that, it's loads of fun I really like making speakers, but I'm not going to fool anyone - especially myself - that I've got anything but the roughest idea how they'll sound once I've finished. If your objective is from the start - "These are the last speakers I will ever own", then you may need to think about that a bit before you start.

That might have been the case 20-40 years ago. Nowadays we know.

 

[DrFG51]

I respectfully disagree

We can produce hundreds of pretty graphs and have a really good idea how something *should* sound, but the exact interaction of the speaker that you've built in the exact position in that exact room it's sat in will only become finally clear when you switch it all on. We do know a lot for sure, but there's an almost limitless set of variables in play.

 

[Jukka]

I respectfully disagree

We can produce hundreds of pretty graphs and have a really good idea how something *should* sound, but the exact interaction of the speaker that you've built in the exact position in that exact room it's sat in will only become finally clear when you switch it all on. We do know a lot for sure, but there's an almost limitless set of variables in play.

Thank you, sir, that's true, but the room is a problem with all speakers, not just DIY. It's a lot easier job to deal with the room, with treatment and/or correction, than find a speaker that fits as-is.

Once the room is dealt with, you are back to the original question and in my opinion, you can get more info about how a speaker sounds through simulation than, well, auditioning it briefly without measurement. This will require prior listening experience with the simulated type of build, though.

There are things that cannot be known without listening, like how a specific driver sounds, but the only surprises I've had with my DIY is how well they match to the simulation (mainly VituixCAD).

 

[D!sco]

What a wonderful shitstorm you've started. OP, if you're still here, I'll parrot what I said in your last thread with some clarification. Lots of DIY speakers will compete with or even beat top of the line hi-fi on a regular basis at a technical level. I quite like the look of my speaker sets, but they're basically bottom-tier finishes on a standard retail product and belong in the defective sale pile. Hiring a professional at this price point is a must. Cabinetry and finishes are tough, but a furniture-maker will do a spectacular job. Auto shops will even finish them for you like a car. Neither of these are cheap, but you'll get a properly custom speaker for your setup. It's what I'd do at some point.

Which speaker is kind of the hard part, and what is it trying to be better than? I'd argue that it would be difficult to compete with an enterprise like Genelec, Neumann, Buchardt or Swan, who manufactures specific drivers for their speakers and the electronics to FIR/DSP control them. Supposedly companies like this use an anechoic chamber to calibrate each speaker after it's finished, at least with their higher end models. That's a really tall order for someone like you or me. You've gotta know somebody. On the other hand, this can be done to a lesser extent in-room. It just has to be done in every room thereafter, which it kind of should anyways.

If I remember the room you were talking about, it was rather large. PA speakers may do best due to their high SPL handling, ruggedness, replace-ability, etc. A variation on Perry Marshall's Bitches Brew could be in order. Any of the top end Scanspeak or SB Acoustics towers ought to do it, if that's more to your liking. I'd recommend going to a few of the DIY shows and asking around. I'm not there yet, personally, but I'd also rather get my hands dirty and take a stab at my own designs first. Ugly glue stains and bad paint jobs and all.

 

[Robh3606]

If you are talking about DIY to compete with you having no previous experience it's a fools errand. Even trying to copy a design is a bad idea without some expertise. How do you plan on verifying you have it right without basic design understanding, so you can troubleshoot, and a measurement system you are actually proficient using for verification??

Rob

 

[TimW]

If I were to pay someone to make an all out custom pair of speakers for me I would go to Joseph Crow. His designs have an emphasis on directivity control and low distortion with a goal of smooth and flat frequency response, but not at the levels DSP can achieve. He has done several big horns for compression drivers and large pro audio woofers among other things, and his CNC woodwork is absolutely beautiful IMO.

 

[Mr. Widget]

...and his CNC woodwork is absolutely beautiful IMO.

Hmmm... considering aesthetics only; if I had to pick between Wilson Audio and these it would be tough for me. As much as I love wood, I prefer a more svelte look.

BTW I really dislike the aesthetics of Wilson Audio.

 

[srrxr71]

Just throwing this out there not exactly DIY but sort of it is.

I’m a big fan of point source.

So I would go Genelec 8351 pair.

Plus Rhythmik midbass units

Plus Rhythmik or other sub x 2-4.

All tuned with a Trinnov unit capable of 6 or more channels.

Or to make it more DIY perhaps do the crossovers in Acourate. If one were inclined to do so.

 

[srrxr71]

What a wonderful shitstorm you've started. OP, if you're still here, I'll parrot what I said in your last thread with some clarification. Lots of DIY speakers will compete with or even beat top of the line hi-fi on a regular basis at a technical level. I quite like the look of my speaker sets, but they're basically bottom-tier finishes on a standard retail product and belong in the defective sale pile. Hiring a professional at this price point is a must. Cabinetry and finishes are tough, but a furniture-maker will do a spectacular job. Auto shops will even finish them for you like a car. Neither of these are cheap, but you'll get a properly custom speaker for your setup. It's what I'd do at some point.

Which speaker is kind of the hard part, and what is it trying to be better than? I'd argue that it would be difficult to compete with an enterprise like Genelec, Neumann, Buchardt or Swan, who manufactures specific drivers for their speakers and the electronics to FIR/DSP control them. Supposedly companies like this use an anechoic chamber to calibrate each speaker after it's finished, at least with their higher end models. That's a really tall order for someone like you or me. You've gotta know somebody. On the other hand, this can be done to a lesser extent in-room. It just has to be done in every room thereafter, which it kind of should anyways.

If I remember the room you were talking about, it was rather large. PA speakers may do best due to their high SPL handling, ruggedness, replace-ability, etc. A variation on Perry Marshall's Bitches Brew could be in order. Any of the top end Scanspeak or SB Acoustics towers ought to do it, if that's more to your liking. I'd recommend going to a few of the DIY shows and asking around. I'm not there yet, personally, but I'd also rather get my hands dirty and take a stab at my own designs first. Ugly glue stains and bad paint jobs and all.

I wish. I have a pair of Genelec w371 which are verifiably more than 3dB off each other in distortion from the front woofer.

I have even posted videos of the 2 at the same level.

Let’s not even get into their finish either. Although that isn’t important to me. The minutes do have a nice powder coating though.

 

[D!sco]

I wish. I have a pair of Genelec w371 which are verifiably more than 3dB off each other in distortion from the front woofer.

I have even posted videos of the 2 at the same level.

Let’s not even get into their finish either. Although that isn’t important to me. The minutes do have a nice powder coating though.

Isn't the whole point of buying from a company like Genelec that you have customer support and can propose an exchange/return?

I don't want to sound like I'm putting down DIY by suggesting commercial, but it's practical on that level in particular. A professional company with their reputation on the line is offering a series of products that can be tested and exchanged at little to no cost before finalizing their decisions. I can't actually picture Neumann or Genelec not working with you on that. It would be awful if their customer service is worse for private vs. enterprise like the radio station I worked at.

 

[srrxr71]

Isn't the whole point of buying from a company like Genelec that you have customer support and can propose an exchange/return?

I don't want to sound like I'm putting down DIY by suggesting commercial, but it's practical on that level in particular. A professional company with their reputation on the line is offering a series of products that can be tested and exchanged at little to no cost before finalizing their decisions. I can't actually picture Neumann or Genelec not working with you on that. It would be awful if their customer service is worse for private vs. enterprise like the radio station I worked at.

Yeah they are but really kind of fighting me at every step. At least that’s my perception. I discovered the issue within 30 days of delivery and they have replaced amps (which took a month).

The replacements show the same issue. Which means it’s the drivers.

I just didn’t like their process of doing the least possible and making me run around.

I made the point that if I were using these to produce content or some mission critical task their service would be woefully inadequate.

I can’t understand maybe after 2 years but within 30 days?

Also do they test anything? We have to test for them it seems.

Very disappointing. I had a lot of faith and respect for Genelec. And my wallet being lighter proves how much.

 

[D!sco]

In my studio experience, they expect you to correct them with DSP anyways to fix for room modes and resonances. In that case, the bass tower would only have problems if they bottomed out or hit maximum SPL in some way. It is pretty disappointing that they're being a pain about it, though. I think an unintended side effect of their growing popularity is having a higher cost of business by running through exchanges. This happened with JBL and Sony customer service, IIRC.

Dumb question, but have you checked the dip switches? I've never seen a W371 or it's plate amp, but sometimes the monitor's bass/treble adjustments pass through depending on how they're connected. The digital system is often a big black box to everyone but the engineers.

 

[srrxr71]

In my studio experience, they expect you to correct them with DSP anyways to fix for room modes and resonances. In that case, the bass tower would only have problems if they bottomed out or hit maximum SPL in some way. It is pretty disappointing that they're being a pain about it, though. I think an unintended side effect of their growing popularity is having a higher cost of business by running through exchanges. This happened with JBL and Sony customer service, IIRC.

Dumb question, but have you checked the dip switches? I've never seen a W371 or it's plate amp, but sometimes the monitor's bass/treble adjustments pass through depending on how they're connected. The digital system is often a big black box to everyone but the engineers.

No there are no dip switches on this.

Also bottomed out is exactly what i’m experiencing.


Anyway today I think and hope I’ve turned a new page with Genelec support. I feel some optimism again.

Could you please explain what you mean by running through exchanges?

 

[D!sco]

No there are no dip switches on this.

Also bottomed out is exactly what i’m experiencing.


Anyway today I think and hope I’ve turned a new page with Genelec support. I feel some optimism again.

Could you please explain what you mean by running through exchanges?

Shipping a speaker back and forth is an operating cost that high margin, low volume companies do to keep clients happy. This exchange is important for your BRAND and your customers trust the BRAND if they're more willing to exchange less than perfect items. This lowers their overall margins and makes BRAND less likely to continue exchanging products until the customer is happy. When volume goes up, customer service goes down. I think sites like ASR may have unintentionally reduced the quality of customer service by increasing the volume of niche studio products to non studios.

 

[srrxr71]

Shipping a speaker back and forth is an operating cost that high margin, low volume companies do to keep clients happy. This exchange is important for your BRAND and your customers trust the BRAND if they're more willing to exchange less than perfect items. This lowers their overall margins and makes BRAND less likely to continue exchanging products until the customer is happy. When volume goes up, customer service goes down. I think sites like ASR may have unintentionally reduced the quality of customer service by increasing the volume of niche studio products to non studios.

Yes I see these are priced for bespoke service included. But they suddenly treat you like a mass market consumer when they start selling many of them.

Basically it’s greed. Getting volume sales on niche high margin products and rather than be thankful about profits they then reduce the service quality to match the volume they are getting rather than increase it knowing they have that extra revenue.

I tell you everything goes well in the world until greed enters the room.

Maybe they need more service staff? It seems they kick you back to level 1 staff as soon as possible. They might be slammed with work?

 

[D!sco]

Yes I see these are priced for bespoke service included. But they suddenly treat you like a mass market consumer when they start selling many of them.

Basically it’s greed. Getting volume sales on niche high margin products and rather than be thankful about profits they then reduce the service quality to match the volume they are getting rather than increase it knowing they have that extra revenue.

I tell you everything goes well in the world until greed enters the room.

Maybe they need more service staff? It seems they kick you back to level 1 staff as soon as possible. They might be slammed with work?

Or they're expecting a fad to pass and don't want to change their business. It's tough scaling down after becoming a larger operation.

 

[srrxr71]

Or they're expecting a fad to pass and don't want to change their business. It's tough scaling down after becoming a larger operation.

Yeah I see. But darn give some love to the guy who had his $18k woofer system for less than a month. Really I wonder about QC too. Could be shipping also. Who knows?

 

[JRS]

There is an element of drama in some of these comments. Certainly before the era of near-perfect DSP and amplification there was a lot of artistry in getting driver and cabinet integration right - and there still is for passive designs and less conventional designs, but for some scenarios it's less scary than it used to be....
<snip>
Not that I'm recommending any of this, but I wanted to make the case that for a 3-way active speaker with DSP based on an existing commercial design you like, it's not as scary as it seems. And the result would be super-satisfying!

Although I haven't tried to emulate a specific design (its nigh impossible to avoid applying well known generic principles that are apt to be common to many SOTA commercial offerings) I agree in principle and there are several factors that when combined, place within reach an approximation of SOTA performance:

• Inexpensive multichannel DSP and measurement equipment (granted its not KLIPPEL) which when properly gated and spliced to LF msmts allows one to dial into a superlative on axis FR. I'd opine that DRC is almost de rigueur for a project of this scope.
• The availability of robust and wideband, low distortion drivers with well characterized off axis performance, that given this ridiculous budget can be tried without a second thought. After all we don't need to limit any choices--unlike a manufacturer who needs to constrain cost and keep the total material and production costs to something I imagine is between 15 and 25% of the final MSRP. Drivers from Audio Technology, Bliesma, Acoustic Elegance, Accuton, Scan Speak among many others including the pro sound market (compression drivers perhaps?) serve up a cornucopia of SOTA choices, and while we might not be able to custom spec drivers, can probably hodge podge a 3 way capable of 118dB without strain, and flat within a dB or two.
• When it comes to cabinetry even here we have some options that few manufacturers can afford--one exception that comes to mind is YG Acoustics that machines their cabinetry from solid aluminum, and there are a couple of companies that make drop dead gorgeous, organically oozing cabinets from composites. Mentioned above was the option of outsourcing the cabinetry cuts to a CNC firm--there are some gob smacking veneers out there--yum!; or alternatively I might be inclined to use composite resins and have molds for the panels made--perhaps exploiting some of the cellular layup materials used for aerospace, and then finish with some gorgeous choice of CF weaves (hiring an autoshop to spray paint them seems like a nice approach given all the gorgeous colors out there. Grills? No effing way.
• In terms of approach--I would opt for making some really bad ass active (Purifi what else?) 3 way stand mounts like the Perlistens so as to keep the weight and size down and augment with subs--personally I like the Acoustic Elegance drivers and have a pair of the 15" woofers that are 94dB efficient and able to soak up a couple of kilowatt peaks with aplomb, but certainly there is no shortage of uber woofers to pump out pant flapping infrasonics. Subwoofers are a relatively easy task in many ways as they will likely be subject to a fair amount of EQ in any event and the principles are well known thanks to Thiel, Small et al. No more than 3 cubes and finish to match the mains.

Of course one would have to rough out a prototype or two to ferret out any unanticipated gremlins due to diffraction and off axis mismatches--I'm thinking that wave guides can be modelled and 3-D printed as needed.

Anyhow thats my take on things--I suspect the drivers might run as much as 6k, cabinetry another 6k. Amplification another 6k. Basically engineering overkill and judicious DSP to plane down any serious lumps or humps.
Of course, if one is not retired, you have to consider all the time lost from work.

 

[audiofooled]

Hello Folks,

I expect many of you have seen or contributed to (thank you!) the below thread I started …

$30K Budget - On the quest for my "end game" speaker

Howdy All, Just found this site few weeks ago, been a lurker since then, trying to learn, my first post here. Been in the audiophile "game" for decades, been through many systems, had some great ones to be sure, but always felt I was missing something. NEVER considered active speakers until I...

www.audiosciencereview.com

During the process of creating my short list, the idea of DIY has entered my thoughts as a potential way to achieve same/similar performance for much lower cost. Specifically I am referring to an existing design that has plans (crossover and cabinet) avaible for free or a reasonable cost. To say, I am not considering my own design as I have neither the ability nor the time for such an endeavor.

So, for all the DIY experts here, what would be your recommendation for an end game DIY loudspeaker that would equal or exceed what we on ASR would consider to be the best available on the retail market?

Thanks!

Maybe this sort of DIY could tickle your imagination:

As anything DIY, there are pros and cons but with great drivers, DSP and room treatment it's possible to come up with something you'd really like. Just a thought, simply because loudspeaker enclosures can be so hard to design, build and finish and quite expensive if you want to hire a professional to do it for you. Especially "end game". High expectations equals high expenses, potential money pit if there are more redesigns required before you reach your goal.