Jump to content

The ultimate small speaker - final design peer review thread


lowerFE

Recommended Posts

where do you get the idea from that a PC is limited with respect to what sort of filters it can apply?

 

 

Not filters, but audio algorithms. I'm not sure how to do a equal loudness contour, or a compressor, or a bass harmonic synthesizer on a PC without writing my own implementation of these algorithms. 

Link to comment
Share on other sites

Not filters, but audio algorithms. I'm not sure how to do a equal loudness contour, or a compressor, or a bass harmonic synthesizer on a PC without writing my own implementation of these algorithms. 

JRiver does equal loudness contour, limiting, has a multi-band limiter for bass and supports VST plugins. Voxengo makes multi-channel VST plugins and provides compressors, a bass harmonic synthesizer, and many other VST plugins. They all have 64-bit processing so work well in JRiver's 64-bit DSP engine. 

Link to comment
Share on other sites

Interesting, I originally discounted JRiver thinking it is tied to the music player. But looks like there is a WDM driver to let it act as the DSP for everything in Windows. I still would like to have a system independent of a computer, but JRiver will definitely be useful to reuse my current speaker, for example replacing my parent's bedroom stereo. They have a HTPC in the bedroom already, so it is perfect. The HDMI out will be perfect to use with a receiver to do the 6 channel amplification needed for the speakers in a neat package. Will need to see if JRiver can accept an input from a sound card so it can process the CD player. 

 

Thanks for the JRiver suggestion!

Link to comment
Share on other sites

Actually what am I saying, JRiver can be made to work with everything in Windows and accept an input, just like what I said earlier. Downloaded the trial. It looks like there's some serious DSP functionality there. Why didn't I at least check this out before brushing it off a few years ago??? 

 

OK, time to go back to PC DSP. I have a spare MacBook Pro that is perfect for this task. No need to even spend $500 on an Intel NUC, although I kinda want one just because they look slick. But then, a MacBook setup might look slicker since it looks like the external USB sound card is the only piece of equipment powering the speakers. I'm gonna do this at least until I get a Sharc DSP working, which might take some time. Even then a PC would be useful for experimentation since it takes VST plugins. 

I like where this is going! I was dreading the huge wiring mess from having 4 miniDSP's and the potential for time sync issues. 

 

Only thing is there is exactly one 7.1 USB sound card on the market I can use, the Asus Xonar U7. I wish there were a few more choices. I'm not a DAC guy or anything, but I kinda wished it had an optical input so I can avoid 2 unnecessary DAC and ADC steps for the input I want to use. Apparently it doesn't work well with Windows 10 as well. 

Link to comment
Share on other sites

  • 2 weeks later...
  • 3 weeks later...

I know this is taking a long time, but bare with me. It's going to be glorious when it is done! 

 

I was told of a place called Audio Technology that can make a custom driver for me (and burn a $1500 hole in my wallet). Any recommendations on the T/S parameters? It'll be a variation of this woofer. 

 

http://www.audiotechnology.dk/iz.asp?id=4|a|122|||

 

So far the woofer is going to have a 2" voice coil, 15mm voice coil height, 6mm magnetic gap height, 110mm x 20mm magnet. The T/S parameter is as follows. 

 

Fs              29,7Hz
Vas             30Liter
Qms           4,29
Qes            0,27
Qts             0,26
Re              3,6Ohm
SPL             86,5dB (90 dB @ 2,83 Volt)
 
The stock driver is a better sounding driver than the Wavecor. With these parameters it should keep the same high efficiency as the Wavecor while greatly improving on power handling (2" VC vs 1.25", larger magnet) and a much lower Fs. 
 
Any thoughts or suggestions?
 
On other progresses. The cabinet is almost done, just need to finish them. I switched from miniDSP to JRiver as my DSP for active crossover. JRiver is a pain to work with, but when all the kinks are ironed out, it will be far better than a miniDSP system. Having VST plugins will allow me to go crazy with psychoacoustic optimizations. First thing on the list is MaxxBass. I'll be using that to generate the harmonics for content below 35Hz at all volumes and content below 70Hz or so after the speaker reaches their limits. So as the volume goes up, the speaker will produce real bass all the way to its limits, and if the volume goes up even higher, then harmonics will fill in. This should be fun.
Link to comment
Share on other sites

  • 3 weeks later...
Some pictures of progress

 

Enclosure

 

20170404_233413_zpsauwoqrus.jpg

 

A little closer look at how tight things are inside with the amp and midrange enclosure.

 

20170327_011557_zpsrrbiqtgr.jpg

 

Here's a cutaway look on the midrange enclosure.  

 

20170401_191736_zpsfidhlw9h.jpg

 


 

Here's some work with the speaker grills 

 

20170413_221243_zpsvp4wjcuq.jpg

 

As always going way above and beyond by ultrasonic welding the nylon tabs for the grill!

 

20170430_170815_zpsyrbbvtve.jpg

 

Picture of the speaker with the grills on.

 

20170413_232332_zpscojorppt.jpg

 



 

 

 


Picture of the veneered and stained enclosure. 

 

20170416_225630-1_zps1ddaiwvi.jpg

 

Here's a piece with a 2k urethane high gloss on. The entire enclosure will have this kind of automotive high gloss finish. 

 

20170423_201551_zpsimhdndye.jpg

 

It's in the finishing lines! After that is a ton of work in tuning. 

  • Like 1
Link to comment
Share on other sites

This is looking great.

 

I must say I'm a little skeptical of the planned signal processing you have going on in the bass range... Boosting + subharmonic generation? I know you are packing as much as you can into this tiny enclosure but it sounds like the woofers and amp will be operating at or near their limits most of the time with all of that processing. Heat buildup, long term reliability and thermal shifting / compression may be a concern with a jam packed and completely closed off internal airspace. I guess you won't know until you try it. :)

 

Did you end up with the AT woofers?

 

How about something in a pic to give scale to these?

Link to comment
Share on other sites

This is looking great.

 

I must say I'm a little skeptical of the planned signal processing you have going on in the bass range... Boosting + subharmonic generation? I know you are packing as much as you can into this tiny enclosure but it sounds like the woofers and amp will be operating at or near their limits most of the time with all of that processing. Heat buildup, long term reliability and thermal shifting / compression may be a concern with a jam packed and completely closed off internal airspace. I guess you won't know until you try it. :)

 

Did you end up with the AT woofers?

 

How about something in a pic to give scale to these?

 

There are people out there, that surprisingly don't play in bands and don't crank their speakers to 11 (well, 12 in your case ;) ) for extended periods of time  :o  :D

 

I don't expect this speaker to be operating near its limits most of the time. A pair can play over 100dB at 40Hz, and I'm generally in the 85dB peak range for normal music listening. The harmonics are going to be in the >60Hz range where the pair can play around 110dB. The speaker may be tiny, but it can belt out some SPL.

 

So far I haven't been getting very good results with the harmonic generation. I probably need to play around more. 

 

I did get the AT woofers with those specs. I was concerned about thermal buildup, which is probably the biggest reason to go with the AT's with their 2" voice coil. It'll still be a problem, but hopefully the woofer will last longer and I can crank it louder longer as well.

 

I don't have a picture with something to scale. I'll show you that when I get the speaker and I'll put an iPhone on top of it that'll cover like half the speaker! 

Link to comment
Share on other sites

Have you considered this driver:

 

http://www.diysoundgroup.com/anarchy-woofer.html

 

I design and produce drivers for a living.  And I have used this driver.  It is extremely well engineered and is of equal or better build quality to what you have chosen. 

 

I have no relationship with this driver or the link to the website.  It simply is a very well made driver.  And it sounds great. 

Link to comment
Share on other sites

Have you considered this driver:

 

http://www.diysoundgroup.com/anarchy-woofer.html

 

I design and produce drivers for a living.  And I have used this driver.  It is extremely well engineered and is of equal or better build quality to what you have chosen. 

 

I have no relationship with this driver or the link to the website.  It simply is a very well made driver.  And it sounds great. 

 

 

Thank you for the suggestion. I did consider it, but I can't fit this woofer. The speaker would need to be 1" bigger length and width wise and 2" deeper and that increased the speaker size by over 50%, which was too much. Would need 1000W to make use of all that excursion in such a small speaker as well, which is obviously impractical. 

 

But I am curious, could you go in a little detail why you think this is an extremely well designed woofer?

Link to comment
Share on other sites

Thank you. It took a while to find that special exotic veneer, but 100% worth it.

 

Right now I'm working on getting controlled directivity via cardioid radiation like the Kii Audio Three and this is really frigging hard and I'm also running low on free time  :(

Link to comment
Share on other sites

Thank you for the suggestion. I did consider it, but I can't fit this woofer. The speaker would need to be 1" bigger length and width wise and 2" deeper and that increased the speaker size by over 50%, which was too much. Would need 1000W to make use of all that excursion in such a small speaker as well, which is obviously impractical. 

 

But I am curious, could you go in a little detail why you think this is an extremely well designed woofer?

 

A kilowatt is no big deal in terms of power availability.  Enclosure size is cast in stone it looks like.  You are indeed doing a nice job on your enclosure.  

 

The driver has all the quality of the Scandinavian drivers.  With a cost that is about $18 over landed cost.  The only improvements I could make is a smaller stronger motor and a larger coil.  Add to that a custom surround and spider and you have tapped out the possible improvements.That is in fact what I have done on my 7 inch.  A little more X-max and a little more refinement on the cone surround interface.  But I'll tell you that blindfolded in front of two of them I doubt that I could tell them apart.  Under controlled conditions with a set of each in a two way speaker configuration my guess would be I probably could tell them apart.  Basically because I paid a little more attention to the cone.  And I have a larger cone.  An attempt at getting the most from the least.  I know that Stereo Integrity makes a very high quality shallow 6.5.  And it has comparable build quality.

 

I started out making speakers using the best of the Scandinavian stuff.

 

  http://www.kravchenko-audio.com/blog

 

Old pictures taken with a box camera. ( remember those?)

  • Like 1
Link to comment
Share on other sites

A kilowatt is no big deal in terms of power availability.  Enclosure size is cast in stone it looks like.  You are indeed doing a nice job on your enclosure.  

 

The driver has all the quality of the Scandinavian drivers.  With a cost that is about $18 over landed cost.  The only improvements I could make is a smaller stronger motor and a larger coil.  Add to that a custom surround and spider and you have tapped out the possible improvements.That is in fact what I have done on my 7 inch.  A little more X-max and a little more refinement on the cone surround interface.  But I'll tell you that blindfolded in front of two of them I doubt that I could tell them apart.  Under controlled conditions with a set of each in a two way speaker configuration my guess would be I probably could tell them apart.  Basically because I paid a little more attention to the cone.  And I have a larger cone.  An attempt at getting the most from the least.  I know that Stereo Integrity makes a very high quality shallow 6.5.  And it has comparable build quality.

 

I started out making speakers using the best of the Scandinavian stuff.

 

  http://www.kravchenko-audio.com/blog

 

Old pictures taken with a box camera. ( remember those?)

 

Very interesting. However, I find it sort of hard to believe that these are all the improvements there are to the driver. What about inductance? It's not so low no improvements can be made by any means. 

 

That stereo integrity 6.5" is very interesting. I actually may have a use for a really high quality shallow 6.5" woofer. That makes a design I'm thinking of possible, which is bad, because that means another year of my free time will be gone! :o However, modeling that woofer shows rather poor output in a small sealed enclosure, about 2dB less for the same power than a 6.5" woofer from Audio Technology with the same T/S parameters as the 5.5" I ordered. I'm not sure why, maybe the high Fs?

Link to comment
Share on other sites

Oh man, I'm screwed. This is no longer the ultimate small speaker.

 

While doing research about getting cardioid radiation in a speaker, I realized just how important controlled directivity is for a speaker's performance in room. You can put the best speaker in the world, put it in a room and the room is going to mess up the sound. The key is to reduce room interaction by controlling directivity and more direct sound to the listener. This speaker has a very wide dispersion pattern, which means almost maximum room interaction. I can make the 100-400Hz area to be cardioid, and maybe even to ~1000Hz if I do well, but above that the speaker is omnidirectional. 

 

I am envisioning a design with the primary advantage of obtaining controlled directivity for almost the entire range. It will be cardioid from low midrange to hopefully fairly high up in the midrange band, then a waveguide will control directivity for the tweeter. This speaker should be much less susceptible to the room and sound quite a bit better in a room. While I know it should doable, I have no idea how to do some of the things, like where to get a small waveguide, yet. There goes another year of my free time and a lot of money. 

 

Or maybe I should just check myself into Audioholics Anonymous  :P

Link to comment
Share on other sites

Evidence suggests that it is more important to have consistent dispersion than it is to have narrow controlled directivity.  In other words, you want your off-axis response to look as similar as possible to the on-axis response.  Ideally, they should differ primarily with respect to the amount of downward slope toward the high frequencies.

 

If you have nearly omni-directional dispersion up high, then you probably don't want to try to do a cardioid if it gives you more directivity down low than up high.   That is likely to sound unnatural because almost all real-life sources have less directivity below a certain point.

Link to comment
Share on other sites

Evidence suggests that it is more important to have consistent dispersion than it is to have narrow controlled directivity.  In other words, you want your off-axis response to look as similar as possible to the on-axis response.  Ideally, they should differ primarily with respect to the amount of downward slope toward the high frequencies.

 

If you have nearly omni-directional dispersion up high, then you probably don't want to try to do a cardioid if it gives you more directivity down low than up high.   That is likely to sound unnatural because almost all real-life sources have less directivity below a certain point.

 

Yeah that's a concern I had as well. However, I somehow managed to narrow the dispersion on the midrange and tweeter to a 135 degree pattern for 1.5 octave around the crossover point. If I can get that on this speaker, then it would match the directivity of the rest of the range. I need to measure (and listen) carefully to make a decision. The waveguide will take care of this problem in the next speaker. 

Link to comment
Share on other sites

The size and shape of the baffle is probably influencing dispersion a lot in your design.

 

Waveguides can definitely help a lot, but realize that they need to be proportionately larger to hold their pattern to lower frequencies.  Some things just can't be achieved with a small speaker.

Link to comment
Share on other sites

The size and shape of the baffle is probably influencing dispersion a lot in your design.

 

Waveguides can definitely help a lot, but realize that they need to be proportionately larger to hold their pattern to lower frequencies.  Some things just can't be achieved with a small speaker.

 

From what I've seen, a ~5.5" waveguide can be made to hold its directivity to at least 2000Hz, which should be low enough for my purposes, hopefully.

 

The fun of this project is to try to achieve the seemingly impossible. The goal isn't as unrealistic as you may think though. What I'm trying to do is a radiation pattern no wider than 180 degrees from 200Hz and up through a cardioid configuration. I want this so the speaker is radiating only forward from 200Hz and up so there would be little sensitivity to the distance to the front wall due to baffle step loss. In my opinion this is important because the amount of BSC drastically changes the sound of the speaker since it affects such a wide band for a small speaker. The distance to the front wall is going to affect <200Hz, but a rise <200Hz sounds pleasant, so it's not really a problem.

 

As frequency goes up, the directivity of the midrange is going to slowly rise. There's nothing I can do to narrow it to constant directivity. Then, up to a point, the midrange directivity will match the waveguide, and that's where it will be crossed over to the tweeter. The question then becomes what kind of a pattern do I want on the waveguide? The SEOS waveguide may be the "easiest" to source, but the 90 x 40 pattern may be a just a little too narrow, especially for nearfield use. I may want something like a 120 x 60 waveguide, but right now that's just wishful thinking because it's not like small waveguides are readily available in many models (hint, there are none). 

 

Ideally I'd like a constant directivity of say 120 degree from 200Hz to like 10000Hz, but that may actually be impossible on such a small speaker if reasonable SPL capability is also a goal. 

Link to comment
Share on other sites

You can only get constant directivity using a few methods.  And the directivity control is frequency dependent.  So each few octaves of response need another of controlling the directivity.  Correctly designed horns is the only broad band method that really works.  But that is a difficult task.  A good horn only covers a few octaves of frequency response before it looses it's effective directivity.  And as you go lower in frequency the horns must get larger.  Can be done.  And I have done design work for it.  But it is not a small system.  Think 4 horns where the largest is almost 5 feet square to control directivity down to about 64 hertz.

 

A simple suggestion.  Finish what you are working on.  It will be great.

Link to comment
Share on other sites

You can only get constant directivity using a few methods.  And the directivity control is frequency dependent.  So each few octaves of response need another of controlling the directivity.  Correctly designed horns is the only broad band method that really works.  But that is a difficult task.  A good horn only covers a few octaves of frequency response before it looses it's effective directivity.  And as you go lower in frequency the horns must get larger.  Can be done.  And I have done design work for it.  But it is not a small system.  Think 4 horns where the largest is almost 5 feet square to control directivity down to about 64 hertz.

 

A simple suggestion.  Finish what you are working on.  It will be great.

 

Good suggestion. I think I'll take it!  :D . I'm just planning ahead. Look at the date on the first post of this thread to give you an idea how long it takes. 

 

The size is a real problem with waveguides. The only other solution I'm aware of is through gradient systems and DSPs to manipulate phase, but efficiency takes quite a hit even compared to standard monopole, let alone waveguides. One nice thing is that the waveguide on the tweeter will allow the speaker to play a lot louder before straining. This should be interesting, a small speaker that can also play really loud. 

 

I'm thinking I should take this one slowly to spread out the fun. There has been a number of innovative speakers coming out just in the past 2 years, and I've been learning a lot and taking cues from them. I can probably incorporate more improvements as well. I hope I will have all of the things that make a big improvement to a speaker's sound quality in the next speaker, which are controlled directivity, increased headroom with waveguided tweeter, and better amplifiers (I think I have a way to fit Hypex NCores). I'm not too worried about things that make little differences like better DAC's or amps, or even drivers since I'm already at the top and any driver improvements will likely be small. 

Link to comment
Share on other sites

I have done a fair bit of design work for directional control of tweeters.  I also have done horn design since 1994.  So I nit pick when I hear waveguide and horn.  There is so little difference that I have come up with this as a qualifying factor for one versus the other.  If there is an efficiency gain in the output of the driver in the new mounting then it is a horn.  If there is no gain in efficiency then it is a waveguide.

 

The trick in that definition is that there is always some change in the efficiency of the driver when mounted in a constraining coupling area from smaller to larger.  So waveguide is basically a marketing name that tried to distance the concept from terrible sounding horns.

 

And for sure you will get me to agree that there are a lot of terrible sounding horns.  There are also some really great sounding horns.

 

If I remember correctly Genelec has one of the few smaller format monitors that claims to have a very controlled dispersion pattern.  I have seen the enclosure but not worked through it's design in any detail.  But it uses a combination of time delay that is naturally occurring through the cabinet design and the driver mounting coupled with some DSP correction.

 

DSP is not magic.  If your speakers sounds crappy to begin with you will only make it marginal with the most brute force correctional you can buy.  That I have witnessed from one of the top DSP designers in the world.  It is impressive to hear many of the warts fade.  But they are still there.  DSP cannot correct for poor driver design.  And if there are serious resonance distortions in the drivers cone surround or spider.  Those types of flawes in design will dominate the sound no matter what you do.

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...