Jump to content

The ultimate small speaker - final design peer review thread


lowerFE

Recommended Posts

That looks really good man. Thumbs up on this. Now comes weeks of messing with DSP programming eh?

 

 

Thank you! And yes, lots of DSP tinkering and measurements next. I'm trying to find good spaces to do measurements, and that's proven to be difficult. There is an outdoor parking lot which is great for ground plane measurement, but the asphalt surfaces ruins the measurement accuracy above 500Hz or so. Measuring at an indoor basketball court is fantastic, but they are really reluctant to let me use it, and when they do, it is at most for 1.5 hours, which isn't much time at all. I only have 3 hours worth of booking time lined up, and I need a lot more than that, especially for cardioid experimentations. 

Link to comment
Share on other sites

Nice work!

Ground plane. Get friendly with some pieces of plywood and duct tape. Place them on the pavement and tape the joints. Two sheets will allow you a 2 meter measurement distance all around your box.

 

I might do that, but that'll be the last resort. The issue is that the noise floor is too high, especially when measuring the rear response for cardioid experimentations. The parking lot I'm using has a few cars driving by every minute, so I'm always crossing my fingers hoping no cars come driving by when I'm doing a sweep  :D

 

Or dig a hole in your yard or a field. That's what I have planned when I work on some speakers. Either that or rent a boom lift for a weekend.

 

Now that's extreme! This will work for speaker measurements, but I have found that when I measure in my front yard my house actually acts as a wall and I can see a dip in the bass region caused by the reflection of the house. 

 

Good to hear you're going to do speakers! I think you'll find them to be just a little more complicated than sealed subs  :lol:

Link to comment
Share on other sites

  • 2 weeks later...

I got the speakers yesterday, and I'm grinning uncontrollably as I listen to my speaker. rockout.png

I implemented the crossover and did some rough tuning. Even with no bass (passive radiators tuned too high), the sound quality already greatly exceeded my expectations.

The beryllium version of the Scanspeak Illuminator tweeter is 100% worth the 2.5x higher price. I expected minor improvements over the non beryllium version in my previous speaker, but the difference is drastic and immediately noticeable. The tweeter sounds sweeter and more detailed.

The Wavecor woofers are a significant upgrade over the Tang Band W4-1720 woofer. I'm hearing a ton of lower midrange detail I've never heard before.

The speaker also has much greater dynamics capabilities. Listening to music, especially classical music at high volumes is thrilling.

I can't wait to put this speaker through an extensive measurement and tuning session.

  • Like 1
Link to comment
Share on other sites

That's excellent News!  Can you fix the PR tuning?

 

 

Yes, I just need to add mass to it to drop the tuning. Tang Band has 2 versions of the passive radiator I use, one with 25g Mms and one with 55g Mms. Based on what I'm seeing it looks like I have the 25g version when I thought I got the 55g. I'll just add enough mass until it hits the desired tuning. 

Link to comment
Share on other sites

  • 2 months later...

Long overdue update. Been incredibly busy lately, so here's a look at the progress I've done with this speaker.

I added a preamp board into the speaker. The amplifier needed 2.7V of input sensitivity to reach full output, but my output stage on my Xonar U7 only outputs 1V. This is made from the OPA1642 op amp. The second PCB is for filtering the noisy auxiliary power from the ICEpower amp. The preamp steps up the voltage by 4x. Even though I only need 2.7x, I used 4x to allow spare headroom for the DAC. Now I can finally get full power from the amps and really crank out the SPL.

ejtpmp.jpg

There were many problems along the way. I had a lot of buzzing problems! I had a ground loop buzz problem that took a lot of effort to find the cause, and it was because the RCA connectors were directly mounted to the baffle without an isolator. Even though wood is a terrible conductor, it was enough to cause a very slight hum that was audible since the amp is running full tilt. I had another buzz problem that was extremely peculiar as well, and turns out it was a few screws that were not tightened inside. I had another buzzing problem that was because the rear end of the passive radiator isn't completely flat, so the washers were rubbing each other at high excursions. Another buzz was due to wires slapping the baskets at high volumes. 4 different sources of buzzing, and this has to be the hardest game of whack a mole I've ever played!

I did some more tuning of the speaker. I need a really high quality speaker to be used as a reference for this speaker. So I went to a friend's house to use his ATC SCM150ASL's as a benchmark and reference.

2vcx2mc.jpg

Before the tuning, while the Reference Mini held its own quite well, the ATC sounded noticeably better in every single metric. The speaker sounds much better after the tuning, and sounds much more similar to the ATC except in the bass. The ATC is a bit bass shy, and I like mine with more bass. 

  • Like 1
Link to comment
Share on other sites

Nice. Looks like you are finally getting somewhere. I can understand the whacka-mole game trust me! Tracking down unidentified causes of unwanted noises from a speaker system is always fun. Sometimes it's is obvious like a wire contacting the back of the cone and other times it is something much more difficult to ascertain.

 

Any guess as to what one of these completed weighs?

Link to comment
Share on other sites

Very nice.  I guess that's one way to voice a speaker.  That ATC monitor looks pretty awesome.  :)

I guess I'm lucky.  The only speakers I've had to track down unwanted buzzing or noise problems for weren't my own.  Perhaps some day I'll regret having not wrapped my internal wiring with fluff or something, but so far so good.

Link to comment
Share on other sites

On 2017-08-02 at 2:56 PM, mwmkravchenko said:

Happy to hear you have attained your goals. Now the real fun begins.  Listening to music!

Not yet! Far from it actually. There's a lot of DSP optimizations and tricks I need to add on this. But at least right now I can have some fun listening to music as I continue to work on this. I'll be a little less busy for the next month or so, so it is actually great timing to be enjoying the heck out of these things

23 hours ago, Ricci said:

Nice. Looks like you are finally getting somewhere. I can understand the whacka-mole game trust me! Tracking down unidentified causes of unwanted noises from a speaker system is always fun. Sometimes it's is obvious like a wire contacting the back of the cone and other times it is something much more difficult to ascertain.

 

Any guess as to what one of these completed weighs?

I can't even imagine for you. I've had problems where it only happens at a high enough volume. I'm sure that happens for you too, but for the stuff you work on the volume would be :o

This came in heavier than I expected. It weighs 14.7 pounds each. It's a heavy little sucker, and it always surprises people when they try to lift it. Now I have a silly little idea in the back of my mind to make the next speaker I build as heavy as possible :D

22 hours ago, SME said:

Very nice.  I guess that's one way to voice a speaker.  That ATC monitor looks pretty awesome.  :)

I guess I'm lucky.  The only speakers I've had to track down unwanted buzzing or noise problems for weren't my own.  Perhaps some day I'll regret having not wrapped my internal wiring with fluff or something, but so far so good.

That ATC monitor is indeed amazing. That doesn't substitute measurements though. I have done some measurements, but I don't think they are totally accurate. Soon I will be doing extensive polar measurements on this speaker. Gonna do it really right this time around, but it'll take some time. 

Link to comment
Share on other sites

  • 3 weeks later...

Another update. After learning about measurements for some time, I think I'm getting a hang of how to really properly measure speakers. I don't have anything to show yet, because I ran into problems doing accurate polar measurements. I will show measurements when I have a full set of measurements ready.

A few things I noticed. Turns out bipoles are tricky little things. Maybe this is why there are hardly any bipole speakers. I have to use these vertically. The midrange response is rippled and has a dip around 200Hz in horizontal orientation because the baffle is much wider, and the woofer is off center. This causes various interferences due to the path length causing the rear woofer to not integrate with the front woofer properly. The issue is mitigated in vertical orientation since the baffle is much narrower and the woofers are centered. One unintended bonus from the 200Hz dip in horizontal orientation is that when placed vertically, the 200Hz dip appears in the vertical plane, and it is actually pretty close to the typical ceiling bounce cancellation frequency in a normal height room. So the problem of ceiling bounce is unexpectedly reduced. 

I cleaned up the phase of the speaker and used FIR filters to do the bass boost, baffle step, high pass, and clean up the phase from the 3-band compressor. The bass is significantly tighter now, which isn't surprising because I had 6 major sources of phase shift (8th order HPF, 26dB bass boost, 2 LR4 band splitting for 3 band compressor) and it really screwed with the time domain of the bass.

I did a comparison between doing the DSP processing at 44.1KHz and 176.4KHz. The comparison isn't ideal because it isn't blind, and it wasn't instantaneous because JRiver has to be restarted to change sampling rate. I *think* the 176.4KHz had a little more clarity, but the difference is very small. I went back to 44.1KHz because the delay for 176.4KHz is really long at nearly 2 seconds. I need to find a way to process the bass at a lower sampling rate so I can have 176.4KHz for mid and treble without the long delay. 

 

 

Link to comment
Share on other sites

Oh it'll be done more than right. There will be lots of tweaking, but it won't be the kind of minute tweaking that a lot of companies do. I will be focusing a lot more on the stuff that makes a bigger difference in creating a better subjective listening experience. This means LOTS of various DSP features that has seldomly or never been used with speakers before, but done right. It'll take a lot more than 9 months! 

Link to comment
Share on other sites

Nice!  I hope you are having fun.  I know I get a lot of joy out of "upgrading my speaker" every time I devise a new EQ configuration.  I'm a bit surprised by what I've learned along the way.  I figured I'd be well into building the rest of my gear, but I'm still working on getting the best sound from my mains speakers.  It's worth it.  That "last little bit" of performance is really a lot.

Link to comment
Share on other sites

On 8/24/2017 at 4:49 PM, SME said:

Nice!  I hope you are having fun.  I know I get a lot of joy out of "upgrading my speaker" every time I devise a new EQ configuration.  I'm a bit surprised by what I've learned along the way.  I figured I'd be well into building the rest of my gear, but I'm still working on getting the best sound from my mains speakers.  It's worth it.  That "last little bit" of performance is really a lot.

Oh I've been enjoying this,  but not as much as I would like to because I've  had way too many things to do in the past couple of months. At certain times it feels like working on the speaker is a chore that needs to be done. Granted it was a lot of technicalities like measurements and stuff, but now the real fun begins with all the DSP, features, especially since soon I will have a lot more free time to play and enjoy. I'm going to get Dirac soon for this speaker and it should be fun as that should be a big step forward in sound quality in a room. There's going to be endless amounts of tweaking and  optimization that I will be able to do with this. I think that's the best part, being able to incrementally and continuously improve the speaker and have features or combination of features that exist on no other commercial speakers. 

 When I get a breather I got to catch up on what you have done! 

Link to comment
Share on other sites

  • 3 weeks later...

Little update. I just moved last week for a new job, so I didn't get a chance to really work on the speakers much to deal with moving. But for something fun, I improved the limiter design a bit to get a bit more maximum clean bass output from the speaker before the limiter clamps the output. I've wrote a lot about my speakers so far, but this time here's a video of the Reference Mini's doing ~107dB of bass and 109dB peaks. The listening position is about 8 feet from the speakers. For reference the TV is 40". I promised there will be bass! 

 

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...