Kleos - Dipole Line Array

[Johnloudb]

John, I recommend you consider servo sub(s). Maybe an H, W, M or V frame cab. Or get rad and do a Ripole:
http://lautsprechershop.de/pdf/ripol/ridtahler.pdf

Thanks for the link Hank. I seem to have lost what knowledge I gained in my high school German class. But a picture is worth a thousand words and that sub looks cool. Nice dispersion too.

I'd like to do a servo sub Rythik "ripole," but all that stuff does get a bit pricey so will have to see where I'm at after I finish my speaker.

[Johnloudb]

Okay, more progess ... here is the prototype EQ and gain circuit, and includes the power supply regulators. I still need a couple parts - the op amp and power supply - which will arrive here soon. I had initially planned on using the LME49720 but changed to the LME49722, since it has better performance, but need to used the LME49720 in my prototype circuit since I wired it up for an 8 pin DIP socket and the LME49722 is only a surface mount device. So, if I'm like super lucky this circuit will test out fine and I can get it and the crossover circuit wired up in the box I have waiting for it.

[Johnloudb]

Okay, it works!!! Ha! It did take some - close your ears kids - correcting "WTF!" "SOB!" "how did I manage to do that??!!!" errors. But, all is well that ends well. I did get some ringing from the op amps so I had to put in compensation caps across feedback resistors. So, I need to find a way to fit those on the PCB. So next up, I need to install the circuit boards in the box and hook up to the connectors so I can test it on my speaker. I used an unregulated AC/AC wall wart for 16VAC to power the circuit. That barley works to provide +/- 22 volts and with the extra load from the crossover board it will not have enough voltage for the +/-18 volt regulators to work. So going to order a 20 AC/AC wall wart. It's easy and keeps the transformer away from the circuit. Oh, and the batteries you see are not used. They were for a different project.

[wkhanna]

Wooooo Whoooo!



:banana: arty: :banana:


Congratulations, John!


& you did not even need to put smoke back into all those little parts!! :T

[Johnloudb]

Wooooo Whoooo!



:banana: arty: :banana:


Congratulations, John!


& you did not even need to put smoke back into all those little parts!! :T

Thanks!!! Well miraculously non of my errors caused smoke. Very lucky that way.

I was thinking about using a 15v regulator with the same transformer i am using, instead of the 18v regulator i have now. it would save power and lower noise from the regulator. But then I noticed that you get very slightly better op amp performance with 18v, with lme49722 which I plan on using. With this lme49720 I will probably change to a 15v regulator or stick a diode after the regulator to lower the voltage slightly cause it handles less voltage and needs a +/-17 volt supply. 18v is the absolute maximum it can tolerate without risk of damage.

[Hank]

congrats, John! Hey, I like those free-floating RCA jacks - an exotic tweak to keep 'em away from induced chassis hum and ground loops? :W

[Johnloudb]

congrats, John! Hey, I like those free-floating RCA jacks - an exotic tweak to keep 'em away from induced chassis hum and ground loops? :W

Thanks Hank! Yeah, noise can be a real bear in building audio gear. Getting too much noise after the regulator right now and hope the new higher voltage supply that came yesterday will help the regulator work better. Will check out that today.

[Johnloudb]

Well, I'm back at it, and put it in the box with the boards power connected. In a final check of everything I noticed the DC offset of the low pass filter on the crossover board was 0.5 volts. That's way off and it wasn't like that before. So I think i may have mishandled the board and damage some of the FETs which are ESD sensitive. I'll look into more tomorrow and replace the FETs that are misbehaving.

[Johnloudb]

Okay, measured the current through the FETs and they are pulling like 28mA and should only be 10-12mA. So, they are broken!!!! But at least i know what the problem is and I'll have to fix the two JFET buffers on the lowpass section. So, will get on tonight I hope. And then finish wiring it up in the box.

[JonMarsh]

Hope that goes smoothy... are you working on an ESD matt?

[Hank]

I know of a company that makes scads of ESD protection/detection systems and components - it's on the tip of my tongue.ops:
Seriously - I can't wait to read your measurments and more importantly your comments on the sound (or lack thereof) of those beauties! I've ordered a list of upgrade caps and resistors for the interfaces of my "new" Acoustat Model 3's. Can't resist - MUST upgrade. And, I bought an Odyssey Stratos Plus power amp from a gent in Indianopolis, and Klaus (Odyssey Audio is also in Indy) is a friend of his and he will "tune it" for me before shipping. I called Klaus, and as others have stated, he is quite a help and will set the bias per my house AC voltage - he does that for every amp for customers. Everything I read about Odyssey is customer service and bang for the buck performance are over the top. Also good service: I emailed Indy Audio Labs (bought the rights to Acurus and Aragon from Klipsch, which bought them from Mondial Designs and did NOTHING with them), asking if they had a schematic and bias specs for my old Acurus 200x3. One of the partners sent me what I needed.
Keep up the good work!

[wkhanna]

Dan (PewterTA) & I were especially impressed with the Odyssey room at this years Capital Audiofest.
Their gear was as good as some stuff priced far above theirs.
They had one of best sounding rooms there & for the price, it made most other systems seem like a foolish waste of $.

[Hank]

Dan (PewterTA) & I were especially impressed with the Odyssey room at this years Capital Audiofest.
Their gear was as good as some stuff priced far above theirs.
They had one of best sounding rooms there & for the price, it made most other systems seem like a foolish waste of $.

Here's Klaus being interviewed at Capital Audiofest: http://www.youtube.com/watch?v=nCdBFwG4a7Q

[Johnloudb]

And, I bought an Odyssey Stratos Plus power amp from a gent in Indianopolis, and Klaus (Odyssey Audio is also in Indy) is a friend of his and he will "tune it" for me before shipping.

Great find Hank!!! Now I want one! Anyway, will try to focus and get more work done. Haven't felt good lately ... but going to try to get it together, accomplish some things.

[Johnloudb]

Hope that goes smoothy... are you working on an ESD matt?

No ESD mat, probably a good idea to get one. I got a little careless with the JFETs. They've fallen on the floor and still worked fine, so I didn't pay as much to as I should have to handling them. I have used those disposable wristbands connected to a water pipe in the past. Still got one, so I'll use it. Will look into a mat too for the future.

[JonMarsh]

I know of a company that makes scads of ESD protection/detection systems and components - it's on the tip of my tongue.ops:
Seriously - I can't wait to read your measurments and more importantly your comments on the sound (or lack thereof) of those beauties! I've ordered a list of upgrade caps and resistors for the interfaces of my "new" Acoustat Model 3's. Can't resist - MUST upgrade. And, I bought an Odyssey Stratos Plus power amp from a gent in Indianopolis, and Klaus (Odyssey Audio is also in Indy) is a friend of his and he will "tune it" for me before shipping. I called Klaus, and as others have stated, he is quite a help and will set the bias per my house AC voltage - he does that for every amp for customers. Everything I read about Odyssey is customer service and bang for the buck performance are over the top.


Also good service: I emailed Indy Audio Labs (bought the rights to Acurus and Aragon from Klipsch, which bought them from Mondial Designs and did NOTHING with them), asking if they had a schematic and bias specs for my old Acurus 200x3. One of the partners sent me what I needed.
Keep up the good work!

That's good news- I've been wondering how they've been going with that- I hope they make a go of it with that IP. I didn't know Klaus was there, though. I've checked out their stuff, and like the updates to the industrial design and layout.








If I could afford to be a collector, I'd buy an 8008 and a pair of Iridiums just to have this newest generation in the Aragon story.

[Hank]

Hi Jon, perhaps I confused you. Klaus has not gone to Indy Audio Labs. He is still Mr. Odyssey. Both companies are in Indianapolis. I too, wish IAL success in their revival of the Aragon and Acurus brands. It's such a shame that Klipsch bought the brands from Mondial Design and let them languish for years - very strange biz decision:roll:

[Johnloudb]

Progress! I removed the offending JFETs and put in the new ones, matched. This picture shows the one of the JFET pair after I pulled them out.



This picture below shows the JFETS after I put them in, but not soldered. I had to put a small bit of silicone sealant on the bottom of the JFETs to keep them in place for soldering. This worked out well. I have powered them up and they are pulling about 11mA as they should and I was able to adjust the DC offset to zero. So, I think all is well now. Except for the emotional turmoil I've been going through the past month due to getting ditched by my girlfriend. Well, she found someone else but we are still good friends, somehow, and still get together. I don't really care for the guy she found, and don't really get it. But, I have a friend (great guy) who's ex wife hooked up with a guy who hits her. Oh well ...

So, that partly explains my lack of progress and I'm also working hard at getting exposure desensitization, to overcome my sensitivities to sounds and activities which means getting out as much as possible. Excuses, excuses, I know, but it really is hard work and getting back to completely "normal" is a step many with this condition never get a chance to take. But, I intend to get there and my goal is to get back to school next fall and finish my EE degree at the University of Utah.

[Johnloudb]

Singing "Feelings! Misty water color memories of what we were ... Feelings! Of good times gone bye ... "
:boozer:

Okay, tested the crossover and it works good with a crossover point of 600Hz. So, got to wire it up the volume controls and the inputs and outputs on the box. I got this USB to XLR mic adapter so I can use Arta and my Dayton XLR mic to take some better measurements using my laptop.

[Hank]

Sorry about the ex GF John (nuthin' worse than GF problems), but life (and speaker building) goes on. Man, those surface mount puppies are small! I'm still a through-hole guy and will hold out as long as I can.

[Johnloudb]

Sorry about the ex GF John (nuthin' worse than GF problems), but life (and speaker building) goes on. Man, those surface mount puppies are small! I'm still a through-hole guy and will hold out as long as I can.

Thank you Hank, yep life goes on ...
Those JFETs are very small but after a little practice I don't have too much trouble with them.

Got my mic setup and calibrated with ARTA after a some hassle with getting icicle installed. It has a built in driver but had to go through some hoops and a call to tech support to get Windows 7 to recognize and install it.

[Johnloudb]

Okay, it is done! Plan on hooking it up to my speaker tomorrow. So we'll see if this puppy will make music and probably take some basic FR measurements first.

[Johnloudb]

Okay, it is done! Plan on hooking it up to my speaker tomorrow. So we'll see if this puppy will make music and probably take some basic FR measurements first.

Listened, it played music, and took some measurements that I have no confidence in. Will update as things as things progress.

[Johnloudb]

Took some more measurements today and made some more adjustments to the crossover adjusting out the high frequency shelving filter. Didn't need it cause there appears to be something wrong with the RD50? or the amplifier I using, but I kind of doubt that.

There is about a 12dB attenuation above 2kHz. But I did listen to some music and heard some really nice things! Enjoyed it but all the sound seamed to be coming from the woofer array. This, most likely due to the mentioned dip in HF response. Any thoughts? And I did check out the crossover on the scope and it is perfectly flat in HF response.

[Johnloudb]

oh forgot to mention the graph above is of PN Noise input from ARTA's sound generator, used with it's Spectrum Analyzer. I couldn't get STEPs - the other sweep frequency analyzer that comes with ARTA - to work right with my USB mic preamp device, "Icicle." Will keep trying.

[Johnloudb]

I just took some measurements of both RD 50 drivers. They look pretty similar in response and both drop off some above 2 khz. Could it be the different mic I'm using. But it's a Dayton mic and comes with a calibration file that is pretty flat. So looks like I need to change the shelving filter on the crossover to give some boost above 2khz, instead of attenuation. Anyway, will get things worked out. going to try to get "Steps" to work with my mic preamp or may have to buy yet another sound card.

Right Channel


Left Channel

[Johnloudb]

I experimented some more with arta and steps and found the problem I was having with steps. I didn't expect to find the response trace way up at 140dB! Hmmm ... didn't sound that loud? :scratchhead: Anyway when I rescaled the window I saw the trace up there. Don't get that either since entered the correct mic sensitivity and everything else is the same as with Arta. I guess I'll just have to go though the same calibration routine I did with arta.

Here is 30 off axis, at 2 meters, which makes the response look better up to 5kHz, so will need a tweeter array to fix above that. don't know what's going on with the response around 600-1.5khz. Hopefully, moving the speaker to the long wall, along with room treatment will fix the response in that range.

[Hank]

Hang in there John, you're progressing. You're not crossing over to the RD's as low as BG says they can go, are you?: "The bandwidth of the RD Series drivers is extraordinarily large, and is usable from as low as 150Hz to beyond audibility."

[Johnloudb]

Hang in there John, you're progressing. You're not crossing over to the RD's as low as BG says they can go, are you?: "The bandwidth of the RD Series drivers is extraordinarily large, and is usable from as low as 150Hz to beyond audibility."

I'm crossing over at 600Hz. With the Spectrum Analyzer (Arta) the output below the crossover point is garbage. Don't know why it does that but got the same with the iphone software I was using.

Still appears that steps doesn't work with my soundcard yet. The trace I get with it is just a flat response, and not from the mic.

I can do a swept sine response using Arta's spectrum analyzer and my iphone software to generate the signal and that seemed to work well, but got more questions. Anyway, will keep taking meausrements and post some stuff. I just have no confindence yet in what I've measured ... all new to me. But, I'll experiment some more and post some stuff and see if some measurement guru's (Jon, CJD, and so on) can make sense of it.

John

[Johnloudb]

Well, I think I got some decent measurements now. I'm using my iPhone app, Signal Scope to generate a swept sine signal. I was told that the Log sweep was the one to use. And it looks more reasonable. I used Arta set up for external generator, 1/6 octave.

Here is the linear sweep -


Here is the log sweep -


Next up I'll measure the woofers.

[JonMarsh]

That's pretty believable- how fast is that sweep- is it like a sine chirp, or relatively slow stepping? with a dipole you always have to worry to some extent about what you get reflected from the rear wall and how your setup influences that. For crossover design, if you aren't using software with strong windowing capabilities that can ignore a lot of the stuff after 10 msec or so, I recommend measuring outside for an initial crossover design. Better to get the crossover right and then work on the room issues separately, IMO.

Oh, and Hank, ThomasW and I determined some time ago that the sweet spot on the RD drivers was 600Hz and up; below that, HD and IMD rises too much. That's why Tom's setup uses the Acoustant 1+1 in the lower midrange and upper bass (100-600) (hint, hint).

[Johnloudb]

That's pretty believable- how fast is that sweep- is it like a sine chirp, or relatively slow stepping? with a dipole you always have to worry to some extent about what you get reflected from the rear wall and how your setup influences that. For crossover design, if you aren't using software with strong windowing capabilities that can ignore a lot of the stuff after 10 msec or so, I recommend measuring outside for an initial crossover design. Better to get the crossover right and then work on the room issues separately, IMO.

Hi Jon! It's a pretty fast sweep, sine chirp describes it well. The signal generator in Signal Scope won't let you control the sine sweep period, but in IOScope you can control the sweep period, so I will try that next time.

Measuring outdoors would be nice, but we live only a block away from I-80 and the belt route, so traffic doesn't get much quieter than about 55dB and it's really windy right now with canyon winds blowing all the time and cold! But would like to try that sometime, maybe a 3am measurement when conditions are more favorable.

[Johnloudb]

Here are the measurements I took of the Woofer array. These were taken with the mic in the same 30 degree off axis position as with midrange driver. I think the 150Hz room mode goes away on the other wall. -

1/6 octave smoothing


1/3 ocatave smoothing


No smoothing

[Hank]

Oh, and Hank, ThomasW and I determined some time ago that the sweet spot on the RD drivers was 600Hz and up; below that, HD and IMD rises too much.

I remembered that figure - just wanted to make sure John wasn't crossing over too low.

[Johnloudb]

Took some more measurements of the total response, and I'm going to change the crossover's high frequency shelving filter. Going to instead change it to have adjustable high freq boost above 2khz. the way it looks is just how it sounded with music. The music seem to all be coming the woofers.

Okay, here is the on axis total response:


Here is the 30% off axis response:

[JonMarsh]

You're getting there, John! :T. Hang in there, and have a great Thanksgiving!

[Johnloudb]

You're getting there, John! :T. Hang in there, and have a great Thanksgiving!

Thanks Jon! You have a great Thanksgiving too!
Same to you Hank, Happy Thanksgiving!

[Johnloudb]

Okay, fixed the HF response and all is well! Except one thing which I will mention in just a second. But first, let me say I heard some very nice sounds from my speaker and even my dad was impressed, and he loves nothing more to trash my efforts when given the chance! It has real thump in the bass and plays low. So no need for subs anytime soon. Pulling the speaker out into the room another 6 inches really helped and it got rid some hollowness in the sound.



One problem though, I start to get a lot distortion when I turn it up. Sounds very pleasant and musical ... then gets harsh and bright when I turn up the volume. And I don't play it very loud. This can't be the speaker assuming all the drivers are working correctly. So I'm thinking crossover distortion, or perhaps my iphone (my current music source) doesn't like driving the long cable to the crossover? I kind of doubt that though.

Anyway, will have to investigate that, try a different source and take some distortion measurements of my crossover.

[Johnloudb]

I ordered this distortion analyzer yesterday. It's not a sound card but has it's own built in software to avoid problems associated with Windows operating system. Though it is controlled and viewed on using a computer.

http://www.quantasylum.com/content/Products/QA400.aspx

[Saurav]

Never used a distortion analyzer before. Are you planning to insert this in various parts of your signal chain to see where the distortion is coming from?

[CraigJ]

Hi John,

Very nice build and it's been fun following your journey. What amplifier are your using to power the Kleos?

Cj

[JonMarsh]

I ordered this distortion analyzer yesterday. It's not a sound card but has it's own built in software to avoid problems associated with Windows operating system. Though it is controlled and viewed on using a computer.

http://www.quantasylum.com/content/Products/QA400.aspx

[ATTACH=CONFIG]22702[/ATTACH]

Interesting looking unit, John... will have to see if I can get one and check it out over the holidays- it may be useful doing some checks on the NCORE 400 amps I'm building, and easier to setup and document stuff than my HP8903 analyzer. Looks like a lot of performance for $200.

[wkhanna]

Wow....that analyzer looks V interesting!

Your plot looks pretty flat, too!
Glad they sound good!

I am sure you will the rest of it figured out soon.

[Johnloudb]

Never used a distortion analyzer before. Are you planning to insert this in various parts of your signal chain to see where the distortion is coming from?

I suspect the crossover, so gonna try that first. Hope it's not the speaker cause that would be bad! But I will probably take distortion measurements of the speaker too using ARTA once I check out the crossover.

[Johnloudb]

Hi John,

Very nice build and it's been fun following your journey. What amplifier are your using to power the Kleos?

Cj

Thanks Craig! I'm using a Boulder 250AE which puts out 150 watts into 4 ohms. The woofers are around 6 ohms. The amp has a clipping indicator that is supposed to light as soon as distortion is at .1 percent. At the level I play it, I'm probably only pushing a few watts though.