Jump to content

BioBrian

In at the deep end - SB 15 inch for sensitive bass

Recommended Posts

DATS plots:

 

23 I 52 (8"), in box, with & without 400 Hz LP filter:

5a0e5be97f312_23I52withwithout400HzLP.thumb.JPG.40c4ba1074e977b5463e7dac42a55573.JPG

 

The filter seems to almost get rid of that wrinkle.

 

The 15 H 52 (5") drivers in box, no filter:

5a0e5c4e873b5_15H52baredriversincabssnip.thumb.JPG.c13556f345b435ee43a5b3486d33b3b0.JPG

 

(Apparently no such problem).

Share this post


Link to post
Share on other sites
On ‎17‎/‎11‎/‎2017 at 1:00 PM, BioBrian said:

 

I haven't chased RAAL yet, but they do seem very popular in DIY. How low can they safely go at reasonably high sound levels?

Can go down to 1.8Khz LR4, l run mine at 2.3KHz with LR2.

 

raal-70-20xr.pdf

Edited by lusk

Share this post


Link to post
Share on other sites

No more sawdust for a while, I hope!

 

This pic shows the XOs. The lower mid one will be internal, after testing. The tweeter one will hopefully get a clear lid one day, to keep the dust out.

The bigger upper mid one should get its own stand-alone box, and/or attach to the back of the box.

 

5a1639f77a292_NewSSATupper3-wayshowingXOs.jpg.668058738ade4e9fbb201a31f26c65b4.jpg

 

The new system runs off one amp, and shows well the much lusted-after balance between driver sensitivities. I've tried about 8 different attenuation resistors for the tweeters, and that part's settling in.

 

5a163ca14b8b2_NewSSBATssetup.jpg.61c1885633861e3c99ceb2067d4a3785.jpg

 

The system (they have to be labelled SSBATs now) is giving much of what I was after - really fruity on a great solo piano recording (eg Volodos' Brahms hi-def) and getting better by the hour on orchestral. Female vocals are just extraordinary. I feel safe that the AT drivers were a good choice - very pleased. The larger mid (still called 5") seems more appropriate than the SBA ones for my music styles and room size. Dynamic range is enormous.

Share this post


Link to post
Share on other sites

Spoke too soon about the sawdust - I made up the (brave words) permanent filter boards for the SB42 (bottom 15" woofer) boxes. These are now fitted, after some rather frustrated grunting over not having a nice little battery drill like everyone else these days.

 

5a1cf69cefb05_NewinternalfilterboardsforSB42.jpg.66062b5d4014e1afbdd9fb69d3568040.jpg

 

Total DCR is 0.12 Ohms through these C-coils.

 

Due to these speakers residing next to a person who is not deaf (working on that), they've had limited burn-in, but they are sounding better each time.

 

Great news is that Clay has started on my Konverter, which should complete the "next-level" that I've been working on for a couple of years now:

 

Win 10/ASUS/LPSU fanless DC-powered computer/server - Ethernet cable - router or switch on LPSU - Idealish Snakeoil headless Server with Uptone JS-2 LPSU - Curious cable - Konverter - XLR cables - Parasound A21 power amp - DIY 4-Way passive speakers, as above. Other inputs: opt from TV; and Oppo BDP-83 Special Edition, for both 2 channel and 7.1 channel (same front speakers).

 

Share this post


Link to post
Share on other sites

I think we just had a GTG!

 

In response to my pleading for more ears to come and judge these speakers, Linc magicked himself, Sam, and Russell down here for a couple of concentrated hours of listening and discussion.

 

It's hard to explain just how fragile I feel about the whole project, and this insecurity was a bit justified and highlighted the day before, when I took the 3-ways down to the house for more serious burn-in and testing.

 

It's a perverse quirk of fate that my measuring system got struck down just before these babies got birthed, but I've had the gift of time to just listen, and not fiddle. I knew there was something not right, but being the builder I had this surety that both speakers were the same, there were no wiring errors, and any fault must be in the phase stuff between lower woofers, etc etc - a DIY blindness. Expectation bias, or its polar opposite? Is there such a thing as too much SNA? A head full of diversions?

 

Anyway, as part of a clean-up, rewire, and reshuffle in preparation for the arrival of the Konverter, I trotted out my Blu-Ray test disc, and went through a whole bunch of routine tests. (AIX Records Audio Calibration Disc and HD Music Sampler - think it came with the OPPO).

 

The problem showed in the phase test between Front Left and Centre speakers: some of the sound seemed in phase, some out, then when it played deliberately out of phase, some of it sounded better, other frequencies worse. So I undid the 4 driver wires and tested with a 1.5 V battery, and found one 8" mid-woofer wired in reverse (black to positive). Too soon after morning espresso on wiring day? I just can't get my head around how that is possible,  but it happened. All back together, and the pair worked like fitting a limited-slip diff: much more solid bass, and general cohesion.

 

Back in the Barn on top of the 15" woofers again, the system suddenly had a clarity and power that'd been missing, but my biased brain had glossed over in ignorance. I'm SO glad that got sorted before the 6 fresh Golden Ears arrived last night!

 

So a lot of that jelly-belly insecurity had vanished, and I also got a big pat on the head from the discovery of "audiophile" tracks on YouTube. These make my system sound too good to be true, as I'm sure they're intended to. The makers of these tracks must be selling small speakers, because the bass, even at low levels, had the room really full and shaking, and there's a lot of instrument separation - must be awfully impressive in demo rooms.

 

Playing 'real' music was the test though, and this I think highlighted the AT drivers - there were quite a few comments on the 5" Audio Technology mid, which I feel is still yet to show its full colours - they've barely done a dozen hours so far, and is getting more free in dynamics each hour.

 

The weather blessed us with no wind or rain, so all listening was done with the Rolladoor wide open - there was also a general consensus that the birds and subtle bush ambience was a bit hard to ignore. With the door open, there's much less build-up of long wavelength stuff slapping around from one end of the building to the other - it was fairly different after the party was over, and I closed down/calmed down over a cooling ale.

 

So, thanks gents, for the excellent comments and suggestions - I have a bit of homework to do. Thanks also for the great company, and confirmation that this pursuit is the only really meaningful thing in life, something to keep us mentally healthy and alert for decades to come ;).

 

I'd really welcome any other comments that might come up after the event - I can't stress too much how valuable your insights are. And all other ears are welcome of course.

 

BTW, the counter-tenor 'behind' the beautiful piano sound is a Pom with a Welsh name:

 

5a2878014497b_Counter-tenorandSoprano.JPG.13d2e6463c884bb2abcafebccd3d2646.JPG

 

And apologies for the savage choice of organ music! Aho's music is challenging, but most of it more listenable than the sample we endured.

 

 

 

Share this post


Link to post
Share on other sites

That was a great quickie! (GTG that is)

You did an awesome job getting all that ready in time for our arrival, considering it was essentially a fresh build.

Are you a glutton for punishment or an audio masochist? There was no need to fret over this as you are amongst friends.

 

The definition from the AT drivers was very impressive, (and so they should be for that price). They played the various genres without complaint or distortion, which is a very good sign.

 

Whilst the mountain scenery was magnificent, I found it challenging trying to pick detail from the music whilst the birds were singing along.

 

I hope this project provides you with many hours of enjoyment and education. Keep up the good work.

 

Thanks for the tasty spread also. (I hadn't had a chance to have breakfast yesterday)

 

Have a great festive season.

Share this post


Link to post
Share on other sites

Brian, very impressive speakers in a beautiful location. By far my favorite listening room. It encompasses all my favorite things, hifi, timber, welders and one hell of a view! Thanks for having me!

Share this post


Link to post
Share on other sites

Looking good Brian, well done, would be certainly a treat to have a listen. Hope to get there one day when every thing is settled/burnt in.

 

Nothing like a battery polarity test to show up what's a miss. I use a 9v battery, it sits on the shelf for when ever things become disconnected and then re connected or starting off with something new.

You aren't the only person to have connected a driver incorrectly :lol: in a multi way system

 

(He who never made a mistake never made much)

 

Share this post


Link to post
Share on other sites

 

 

 

 

(He who never made a mistake never made much)

 

 

Never a truer word has been spoken!

 

Share this post


Link to post
Share on other sites

It's very heartwarming to have the support and guidance from my no-longer-scary-internet friends - you're all welcome to share the views and wildlife (and speakers) anytime. Thanks for sharing your enthusiasm and journeys.

 

The loss of Omnimic for the last few weeks has been very frustrating, resulting in a complete rebuild of my Win 10 system. My son came and did most of it, and while we were about it, did a complete rebuild of the Idealish Server (my "S8") firmware etc, which requires some very impressive Linux fingerwork. Blessed are the neurons of the young! (And educated too, of course).

 

We reinstalled Omnimic and .... .... .... Error 217, blah blah again.          .....         BUT later, while reloading all my stuff back onto the SSD and looking at remnant files from Omnimic, I tried opening the program again, and it worked!

 

Firstly I'll say that the new Snakeoil build is markedly better than the old - I previously found MPD (17) to sound quite flat, 2-dimensional, and preferred JRiver, which sounded like it filled in a whole lot of missing info. This time, using new Snakeoil and new MPD (20-Git version), it sounds better than the previous JR setup - the instruments are noticeably more defined in the sound-stage, and the whole thing is more 3-dimensional and alive. I've been rather saddened by the apparent loss of faith in the "Idealish Server" department, so this is like a welcome transfusion for me.

 

With the Clay's Konverter in the hands of couriers as we speak, I'm glad to have both the above issues sorted in time for all the new influences not to arrive at the same time, and not know what to blame, or thank, for the changes.

 

So I'll post the good news first - I was very pleased with the first Omnimic trace: the upper 3-way "SAT", taken at upper-mid/tweeter axis, distance 1 m, 8 ms gate. It's one of the best I've seen in my 4? years of playing with this program and my projects:

 

5a2f908b9eb04_2017_12_12OMSATLchSumphase.thumb.JPG.3eeb0afafaca539efb076630182e718b.JPG

 

And I don't know if this'll be meaningful to anyone - it shows what happened when I wired the 5" mid-driver backwards, and took 2 readings, each with mike placed on the axes between the relevant drivers, to show the crossover frequencies (520 and 2700 Hz):

 

5a2f91386bfb0_SATLchSumbtwdriverdips8mssnip.thumb.JPG.c79cc54987b947d6cf8512db299525ab.JPG

 

Some of the rest looks less thrilling, or confusing, but I'll spare the questions for now.

Share this post


Link to post
Share on other sites
12 hours ago, BioBrian said:

 

And I don't know if this'll be meaningful to anyone - it shows what happened when I wired the 5" mid-driver backwards, and took 2 readings, each with mike placed on the axes between the relevant drivers, to show the crossover frequencies (520 and 2700 Hz):

 

5a2f91386bfb0_SATLchSumbtwdriverdips8mssnip.thumb.JPG.c79cc54987b947d6cf8512db299525ab.JPG

 

 

Very good, LR2 at each XO?

How did you decide and put together the final xo design? as haven't noticed any mention of using software and seems you just put it together.

Share this post


Link to post
Share on other sites

Hi @lusk, The 3-way XO is the kit from Troels/Jantzen: AT 3-Way Classic, with the top level components. The 8" driver is a more sensitive version, and the tweeter is from the same 'family' of Scanspeak Illuminator ~3004s. The 15" woofer gets my trial-and-error (won't call it "design", as I didn't do it from maths or software, but it got a lot of empirical testing) 4th order filter.

 

I can't copy/paste Troels' work here; his schematic is at   http://www.troelsgravesen.dk/AT-3WC.htm   

 

It's not a conventional LR2 - there's a resistor over the 5" mid series caps, and the mid has LCRs at both ends. The tweeter section also has an LCR. I'm running it with no attenuation resistors for all but the tweeter.

 

I've been poring over the test results from yesterday, and it looks like I need some work on the 8" driver section - there's a dip in the phase curve that I'd like to understand better, and it might be a little too sensitive for the 5", as the XO point was supposed to be more like 400 Hz.

 

But the mid/tweeter section seems to do mostly the right thing:

 

5a30511bb1ed0_SATMid-Tweetercurvesinclphase.thumb.JPG.8f7df06a2d6cbde45260db1cd9b1a8cd.JPG

 

I'd like to address that step at 1 kHz to soften the slight peak there in the summed response (ears also guide me here), but that fine tuning might need to wait in the queue! (The SBA tweeter wouldn't have this problem, with it's very nice roll-off, but might not have a matching dynamic range).

Edited by BioBrian
Mistake in description of XO

Share this post


Link to post
Share on other sites

The SAT 3 way looking like game over then Brian, response curve is very good

Is the measurement done in field with the speaker elevated a little to get it away boundaries, or inside a room ? 

If it was done inside, the test may go better with lower boundary interaction although I guess that's what the 8ms gate is negating.

Might also be worth downloading REW for a second software opinion :) and some fun

https://www.roomeqwizard.com/

 

Enjoy

 

Share this post


Link to post
Share on other sites

Thanks for the detail, the result you have there is very good. I would be happy with what you have come up with. You can spend heaps of time unnecessarily worrying over little response dips.

 

If you want to take it further and haven't tried yet l would suggest using Jeff Bagby's xo designer and his FRD response blender to extract minimum phase plots to use in it.

 

It is excellent free excel spreadsheet that is reasonably easy to use, it may take a bit of time to learn using it but you could tweek and refine your xo. l found creating xo with it and the actual response curves come out spot on.

Edited by lusk

Share this post


Link to post
Share on other sites
21 hours ago, 125dBmonster said:

... response curve is very good

Is the measurement done in field with the speaker elevated a little to get it away boundaries, or inside a room ? 

If it was done inside, the test may go better...

Yes, inside. But 3.2 m ceiling, with 200 mm foam mattress up there. Speaker was twisted about 40 degrees to avoid reflection from lower woofer cone, and I had 2 layers of felt and 150 mm of wool batts on the floor. Nothing within over 2 m out the sides. 8 ms is longer than I can usually get away with, and it could have been more, maybe.

 

A while ago I went around the paddocks near the Barn doing clap tests and listening - trying to find the best quiet spot for outdoor measurements, within reasonable distance from a power point. There's a spot that might suit, when I get the energy. Should be a bit of fun.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


  • Recently Browsing   0 members

    No registered users viewing this page.

×