Selah Integrity DIY Speaker Kit Review

[JohnBooty]

I never interacted with Jeff Bagby, but what a massive reputation as a kind and talented soul.

He has the kind of reputation that makes me think, "How will I be remembered?" and prompts me to try and be the best person I can be.

Selah I am told will provide CNC machined front panel.

That's a relief to hear. I think very few people have the skill/resources/inclination/tools to cut out the driver recesses on the baffle themselves.

Still, I always think it's a shame that Madisound and Meniscus don't offer flat-pack kits like PartsExpress.

I realize how bulky those flat-packs are to ship and store. PartsExpress surely enjoys some economies of scale there and is thus able to handle that in ways that smaller companies cannot. But there are a lot of people out there who want to enjoy these kits but aren't able to do the cabinet work.

(A common suggestion is, "ask a local cabinet maker!" but unless you are friends with one, that can be rather expensive AFAIK unless they are really hungry for work - please correct me if I'm wrong)

 

[Stephen]

Those two tracks don't seem as heavy on the low bass as that track tested by Amir. I listened to those YouTube vids you linked and according to Orban Loudness Meter the Rodolfe Burger one is about 4 or 5 dBFS short in the bass (my interpretation) of the Amir one. The other track you linked is about 8dBFS short in the bass. So even though both tracks sound like there's some impressive bass, the point is that the Amir linked video contains very low bass in large amounts that doesn't actually sound loud but is actually being played at a very high level (because our hearing is so poor in the 15-30Hz zone), so Amir's video probably would challenge speakers more. But, I suppose it's arguable why choose a test track where the bass is virtually inaudible because it's too low for our hearing, and your tracks you've chosen are probably more applicable to everyday listening because I think the bass is a bit further up the frequency range in your examples.....so it's probably more applicable to real world bass.

Actually, yes exactly, my point is to hear a difference between speakers. If its not audible, then there is little interest. Still the 'North Star' track transforms my Neumann in hair dryer...

 

[MZKM]

Would rounding over the edges of the baffle help tame some of the dips/peaks in the response?

Likely, mainly the issue around 3000Hz where the of-axis is louder than the on-axis, the 1700Hz issue I would assume would stay.

 

[amirm]

(A common suggestion is, "ask a local cabinet maker!" but unless you are friends with one, that can be rather expensive AFAIK unless they are really hungry for work - please correct me if I'm wrong)

I think these companies should provide g-code for CNC machines and then someone should setup shop to make front baffles for people using such. The rest of the box can be built with rudimentary tools.

 

[hardisj]

I don't usually see correlations between frequency response errors and THD percentage but we have one here at 1.7 kHz.

It is extremely common with cone ege/surround resonance (both pieces moving out of phase with each other). With a 6-7 inch mid/woofer you'll see it around 1-2kHz. Like clockwork. I've seen it with Scanspeak, Dyn Audio, Satori, and Audio Technology, just to name a few. Always 2nd-order.

Here's just one example:
https://www.erinsaudiocorner.com/driveunits/dyn_mw172/

 

[jtwrace]

I think these companies should provide g-code for CNC machines and then someone should setup shop to make front baffles for people using such. The rest of the box can be built with rudimentary tools.

OR send a baffle for 3D scanning and we can reverse engineer the baffle.

 

[hardisj]

It's likely. You've probably seen this but putting it here for others who haven't.

http://www.troelsgravesen.dk/Edge-coating.htm

I hadn't seen that. Pretty neat stuff. Troels' site has so much excellent info. I need to stop by there more often.

 

[Selah Audio]

This is a review and detailed measurements of the Selah Integrity DIY Speaker Kit. @Selah Audio designed this kit in memory of DIY designer, Jeff Bagby, who sadly passed away due to Covid-19. The Kit was kindly built by @Rick Sykora. Meniscus Audio will be selling the kit and a portion of the proceeds goes to family of Jeff Bagby. The kit cost starts at $630 plush shipping for a pair.

This is a three-way design and built out of ultra dense and thick MDF:

View attachment 80689

As you can see, there is an oval passive radiator in the back. It was a pleasure to have a pair of binding posts where I could get my fingers in there.

Measurements that you are about to see were performed using the Klippel Near-field Scanner (NFS). This is a robotic measurement system that analyzes the speaker all around and is able (using advanced mathematics and dual scan) to subtract room reflections (so where I measure it doesn't matter). It also measures the speaker at close distance ("near-field") which sharply reduces the impact of room noise. Both of these factors enable testing in ordinary rooms yet results that can be more accurate than an anechoic chamber. In a nutshell, the measurements show the actual sound coming out of the speaker independent of the room.

I used over 800 measurement point which was enough to compute the sound field of the speaker within 1% error.

Temperature was 75 degrees. Measurement location is at sea level so you compute the pressure.

Measurements are compliant with latest speaker research into what can predict the speaker preference and is standardized in CEA/CTA-2034 ANSI specifications. Likewise listening tests are performed per research that shows mono listening is much more revealing of differences between speakers than stereo or multichannel.

Spinorama Audio Measurements
Acoustic measurements can be grouped in a way that can be perceptually analyzed to determine how good a speaker is and how it can be used in a room. This so called spinorama shows us just about everything we need to know about the speaker with respect to tonality and some flaws:

View attachment 80690

The on-axis response looks very good. The only exceptions are the perturbation around 1.7 kHz and a bit of rising energy in the upper end. The dip at 1.7 kHz remains in off-axis as well:
View attachment 80691

Resulting predicted in room response is what we already know:

View attachment 80692

So very good overall.

Digging into the source of that dip it appears to be the bottom end of the midrange driver:

View attachment 80693

Speaking of mid-range, it seems to be playing louder than it can without distortion:

View attachment 80694

I don't usually see correlations between frequency response errors and THD percentage but we have one here at 1.7 kHz.

View attachment 80695

Impedance is around 5.5 ohm which is higher (better) than typical small speaker, putting less stress on the amplifier:
View attachment 80696

The mid-range provides wide beamwidth in the mid-range:

View attachment 80697

Here is the same as a heatmap:

View attachment 80698

The 3-way design makes vertical placement less critical:

View attachment 80699

And here is the waterfall:
View attachment 80701

Speaker Listening Tests
Having seen the measurements before listening, I expected the sound to be good but it wasn't! It sounded muffled in the mid-range. Fortunately I quickly realized I had the EQ on from last test I was doing so turned that off and all was well with the universe. The experience was very good albeit with a touch too much upper end energy. I put it this way as opposed to saying "bright" as the treble response is not high enough to be annoying. It was close enough that I did not bother to EQ but you may want to.

I toed the speaker out a bit and while that did reduce the highs a bit, I had a preference for the more clear direct sound. Anyway, it is something you can play with.

The Integrity is very inefficient by the way requiring me to turn up the volume to very high levels. Fortunately it handled the power well and could get loud despite a single speaker playing.

My subsonic bass test tracks which I call "speaker killers" earned their designation though causing massive distortion. You better have a filter for such bass notes or not have as powerful of an amp as I have. Fortunately the woofer recovered as if nothing had happened and continued playing.

Note that 99% of my music and even those with heavy bass don't have the subsonic energy that cause the problem above.

Conclusions
I must say, I was very nervous going into this review. The work is for a good cause but so is transparency in speaker reviews. I had my fingers crossed that it would measure and sound well. What a relief it was that it did both.

I am happy to recommend the Selah Integrity DIY design Kit. I hope you all support the effort by purchasing them and getting experience with building your own speaker and doing some good in the process.

-----------
As always, questions, comments, recommendations, etc. are welcome.

Our Italian plums are producing this year again and soon I need to dehydrate them as they won't last long on the shelf. No, I don't need your money for the dehydrator. Who do you think I am? A constant beggar??? No, I need money for the electricity to run the dehydrator. So please donate what you can using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Amir - thanks for taking the time to review the Integrity. There was some risk in sending it in for review but I felt confident that it would measure well. I'm also happy to see that my measurements are in agreement with the Klippel machine. On some previous tests I wasn't sure about how well the low end extension was being captured; however, you nailed it here!

 

[Draki]

Re 1K7 dip/resonance: the description says the box is made of "... thick MDF". How thick?
Is it possible to remove the mid driver and inspect if the back opening is maybe too thick, a straight hole without a chamfering?
This mid is tiny and has a small back clearance (have it here) and may (will) react to a thick non-chamfered back.

 

[Selah Audio]

Re 1K7 dip/resonance: the description says the box is made of "... thick MDF". How thick?
Is it possible to remove the mid driver and inspect if the back opening is maybe too thick, a straight hole without a chamfering?
This mid is tiny and has a small back clearance (have it here) and may (will) react to a thick non-chamfered back.

The rear of the CNC baffle was chamfered. This is an edge / surround resonance.

 

[F1308]

Having seen the measurements before listening, I expected the sound to be good but it wasn't!

After reading just that, I was about to stop my reading, considering the speaker tested was being reported as a bad performer.
Then I saw by chance, while shifting bookmarked pages, that "
I am happy to recommend the Selah Integrity DIY design Kit.".

A man went into church and while opening the entry door he heard the prayer saying out loudly: GOD DOES NOT EXIST...

Then he turned around and quickly left the church.
He was just being told by the very prayer himself that God did not exist... Billions of blue blistering barnacles...!!!!

Had he stayed a bit longer, he would have listened the prayer adding....ACCORDING TO THOSE WHO ARE NOT BELIEVERS; BUT WE, ALL OF US, THE PEOPLE ATTENDING THE MASS THIS BEAUTIFUL MORNING DO KNOW FOR SURE GOD DOES EXIST, GOD DOES EXIST. YES. INDEED.

 

[Selah Audio]

I can't find this kit on http://www.selahaudio.com/kits

Nvm, found it: https://meniscusaudio.com/product/integrity-loudspeaker-kit-tribute-to-jeff-bagby/

Cabinet not part of it,

Meniscus is the sole supplier for the kit since Jeff had them stock the the majority of his designs and they had a good working relationship. I have arranged for CNC cut front and rear panels.

 

[Xulonn]

Great tribute to Jeff, who was the absolute nicest guy you can imagine in the DIY world. The design reflects his ethic; good system design with a mix of high end and budget parts in an easy to build format.

I had heard the name Jeff Bagby over the years, but did not know much about him. Found a nice obituary article for him at the Washington Post. {Probably paywalled, but I could view it with Firefox and the appropriate add-on, or Google Chrome Canary.]

Jeff Bagby endured the grueling dialysis treatments, the years of praying for a new kidney and the eventual transplant with an attitude so unfailingly upbeat, his wife, Christine, joked that she should buy him a Superman shirt.

She ended up finding one online. Sometimes he wore it beneath a button-up shirt like Clark Kent, with the “S” peeking out from below his collar. Over the years, she added a Superman mug and a bendable action figure as he took on cancer with the same resolve — and survived it, too.

“It was just his personality,” said Christine, who married Jeff in May 1987. “He just never let anything get him down.”

But the coronavirus proved to be a health threat he could not overcome. On March 24, the Chrysler engineer, math whiz and devoted family man became the first person to die of the virus in Howard County, Ind. He was 60 years old.

 

[Tks]

I've been wanting to try my hand at some curved pieces. Anyone have any idea how they made the front left and right piece? Is it gentle enough that one can form some plywood with soaking and time or is it just easier to kerf it? Or maybe just some other technique I am not aware of?

@Xulonn maybe you can offer some advice?

 

[Draki]

Amir, how do you measure the individual drivers (the near-field measurements)?
Do you disassemble and access each driver directly (via the appropriate crossover "way") - since the speakers have a single pair pair of input posts, you couldn't do it externally.
Or each driver is measured by close miking while they all play the test signal?
Sorry if this has been asked and answered elsewhere, I couldn't find it.

 

[fordiebianco]

(A common suggestion is, "ask a local cabinet maker!" but unless you are friends with one, that can be rather expensive AFAIK unless they are really hungry for work - please correct me if I'm wrong)

Over here in the UK I have emailed a few dozen cabinet makers and carpenters offering them good money for building (or at least cutting the wood/MDF) me speakers from ready made plans, but have never been successful. Either the emails get ignored or I get a curt 'I don't do small jobs'.

Unfortunately my own DIY skills are non-existent, as previous disastrous adventures into speaker building have shown.

Sigh.

 

[Juhazi]

About this 1,6kHz issue - it seems to be associated with half-roll rubber surround, spherically equidistant from center (axial-symmetrical decomposition). Many dedicated midrange drivers use inverted roll or pleated surround to mitigate this problem. ScanSpeak has recently released oval cones. Because of axial-symmetry, the peak/dip disappears off-axis when opposite side signal start to develop phase difference.

I didn't find explanation for H2 dominance of this phenomenom, but it's most likely simple geometric/mathematical phenomenom.

https://www.hindawi.com/journals/aav/2017/2792376/
https://www.klippel.de/fileadmin/_migrated/content_uploads/KLIPPEL_Cone_Vibration_Poster_01.pdf
https://www.klippel.de/fileadmin/_migrated/content_uploads/KLIPPEL_Sound_Radiation_Poster_01.pdf
http://www.troelsgravesen.dk/Ellipticor-1.htm

 

[Rick Sykora]

I've been wanting to try my hand at some curved pieces. Anyone have any idea how they made the front left and right piece? Is it gentle enough that one can form some plywood with soaking and time or is it just easier to kerf it? Or maybe just some other technique I am not aware of?

The fancy cabinet was done by gluing layers of mdf and then machining with a cnc. I have seen simpler chamfering done with just a table saw. For details on the Integrity, see here: http://techtalk.parts-express.com/f...6265-jeff-bagby-tribute?p=1442304#post1442304

The other talented woodworker that I know built a curved cabinet version of the Selah Essenza. He did this by kerfing layers of this plywood and gluing them together. UPDATE: I recently talked to my woodworker friend and found he did not have to kerf. His speaker has round sides and flat front and back. So just used thin layers of baltic birch plywood and strapping to hold in place while the glue set. He said the work took him about 2 weeks vs. 1 day to do rectangular cabinet

 

[dtaylo1066]

I'm delighted to see this tribute DIY speaker received a positive review. Jeff's contribution to speaker building was huge.

 

[BYRTT]

Although its a sad story and news about destiny of Jeff Bagby, thanks to Amir share another great acoustic review and spindata..

Would rounding over the edges of the baffle help tame some of the dips/peaks in the response?

Likely, mainly the issue around 3000Hz where the of-axis is louder than the on-axis, the 1700Hz issue I would assume would stay.

Round overs is always a good thing but no one never use them and especial good because baffle diffraction interference change with angle and distance, below animation is using one of Jeff Bagby's free good spreadsheets to see how midrange position in Integrity build will change using roundover being on axis at 2 meters distance, we see that the diffraction boost fall off nearly in same 1700Hz area that look problematic in Amir's spindata and distortion graphs but think not its related in that for other test where diffraction boost fall off show no problems other than in directivity, rather the 1700Hz problem is cone ege/surround resonance as hardisj and Selah Audio point out ..

Amir's spindata with other tecnical overlays listed in legend..