Ardent Speaker Camp

[JonMarsh]

Holy shit, Jon, those capacitors are gigantic!

The bigger ones are for the midrange feeds, 27 uF each. The smaller are 4.7 uF. Cost is strongly related to total package size for some reason.

[---k---]

Jon,
You need to cut a couple of those in half and post a photo of the cross section. How do we know there isn't just a lot of empty space inside that aluminum tube.

[Face]

What you see there are two 54uf capacitors in series and encapsulated in resin, resulting in a tremendous 27uf cap.

[Hank]

Cost is strongly related to total package size for some reason.

You know, all my suppliers share that misconception: the more material in a component/product, the more they charge for it. tsk, tsk

[JonMarsh]

You know, all my suppliers share that misconception: the more material in a component/product, the more they charge for it. tsk, tsk

I meant that the cost of specific MR caps in the series did not appear to be linearly related to uF, but to the package size, so the best deal is the highest value cap that fits in a given package size that they make.

I hope that is clearer...

[Hank]

I was making the small joke - I should have pointed that out since most of you guys are engineers :W

[JonMarsh]

Yeah, our vision is sort of myopic and we often have time seeing the small joke for that reason...

[numberoneoppa]

The bigger ones are for the midrange feeds, 27 uF each. The smaller are 4.7 uF. Cost is strongly related to total package size for some reason.

Sorry if I missed it. How much did they cost?

[JonMarsh]

Pricing on all Clarity caps is on the Madisound site- that's their US distributor.

[savage25xtreme]

I think my heart just skipped a beat when I saw the pricing on those. Jon, Obsess much? I guess when Jon says its an all out design he means it. ;x( 8O

[JonMarsh]

A bit more tweaking and thinking

Have the LF, mid, and tweeter crossovers pretty well sorted out to my liking by ear now, a bit better integration and less stress on the mid- all for the better.

But...

Been listening to the Modula Xtreme's too much. I guess.

I really find myself liking the sealed flux capacitor bass alignment. A lot. Probably a bit too dry for some, but very articulate and at times the speaker "disappears" more in the bottom end, that is, doesn't call attention to itself as a speaker.

Been reviewing the options, and checking out some other drivers. These days you may have noticed I've become something of a Scanspeak fanboi with their 8" and 10" aluminum cone drivers; though they don't publish accurate T/S parameters, they're off in a way that benefits my requirements.

Same for the 7" illuminators. Claiming Qts in the 0.35 range, actually 0.5 - 0.55. Hmmmmm good!







Underhung long throw voice coil, good HD through the upper midrange, what should be a nice bottom end with the right approach. The aluminum cone version looks pretty nice on paper for this application, and is, strangely, less expensive than the paper!

I've got a few more tests to do, but don't be surprised to hear about the Ardents being converted to a Flux capacitor alignment with SS woofers; main stumbling block is hogging out the driver rebates, and I think I have a handle on doing that.

Just say no to ports? Maybe.

I'll publish the current crossover for the record shortly for anyone interested in the Seas design. At that point, that design variant will be considered finished.

This won't help the price point, of course. It WILL make me feel better, I think, about the results versus effort- but at this point THAT IS ONLY A SURMISE! Proof is in the pudding.

Just consider this notice that things have not ground to a halt with the Ardents by any means- and someone's got to do the weird stuff on this forum. It's a dirty job, but it may just as well be me...

[b_force]

They where tested in German magazine Hobby Hifi with the new Wavecors.
They Wavecors perform equal (and cost less)

If you look overall to tests, the only thing to choose Scan Speak is probably for the better tolerance. But I must admit that I don't know how good the Wavecors are.

ScanSpeak are very very great drivers, but not for your wallet.

[JonMarsh]

And Wavecors seems uninterested in US distribution.

In the 90's I was relatively anti Scanspeak based on specific performance issues, but these days I have been doing more Scanpeak drivers than ever before, simply based on having the performance characteristics I'm seeking, and not finding comparable matches, certainly not at lower prices. The D2608/9130 is a good example in a more price sensitive category, but 26W/8867T holds similar sway for me. Not going to be everyone's cup of tea, but if you want very low LF AND midband distortion, it's tough to beat at any price.

[b_force]

To bad for Wavecor

:P I'm really NOT anti Scanspeak, they are really really great!!!
But I think the high price isn't worth it.
Pure "scientific" speaking off course. The most import thing is linear (or constant or controlled) directivity. I think you can design a equally sounding loudspeaker with Peerless or other driver which also perform very well (linear and non-linear distortion behind threshold of what human ears can detect).

But if you look at the whole picture, there are indeed more things to make a great design. It all depends for what you are looking for and what you can effort off course . Sometimes it's just a great feeling with some drivers (and really nothing wrong with that!!!!!) Build quality is also a very good point.

Here are some measurements of the Wavecors:

http://tweakers.net/ext/f/qXZHAU6laGGzHEPnWr6tSJHg/full.jpg

http://tweakers.net/ext/f/gbFRZb8GrPt8udd8A4R4yZ3t/full.jpg

[Johnloudb]

linear and non-linear distortion behind threshold of what human ears can detect

And that is? Well, let's not start a discussion on that, but I think you'll agree there are a lot of different opinions out there.

[b_force]

And that is? Well, let's not start a discussion on that, but I think you'll agree there are a lot of different opinions out there.

There are lot of articles (AES for example) where they tested it.
My opinion is based on that and from experience from myself and others.
But again, I will not fight or argue personal taste for some drivers, because you can not underestimate that taste (maybe psychological) for what (you think) you hear.

I agree with you that this discussion will lead to nothing

[Jed]

though they don't publish accurate T/S parameters, they're off in a way that benefits my requirements.

Same for the 7" illuminators. Claiming Qts in the 0.35 range, actually 0.5 - 0.55. Hmmmmm good!
.

I seem to recall the T/S parameters for the Illuminators as being closer to actual manufacturer specs based on the results published in Voice Coil Magazine, compared to say the Revelators. Not that that is the end-all-be-all of tests.

[JonMarsh]

I don't know if there's a consistency issue- Zaph got some funny results, too. for the 8 ohm 7".

[b_force]

In Hobby Hifi they tested the 4ohm version.
T/S specs:

Re = 3,2ohm
Le= 0,22mH
Fs = 34Hz
Qms = 3,9
Qes = 0,41
Qts = 0,37
Sd = 154 cm²
Vas = 49 l
B*l = 5,0 N/A

[b_force]

There is a distributor in the US for Wavecor:

http://www.worldaudiodistribution.com/

[chrismercurio]

I don't know if there's a consistency issue- Zaph got some funny results, too. for the 8 ohm 7".

You may want to check with Rick Craig as to the accuracy of the published t/s parameters as he has built a few speakers with the aluminum Illuminator woofer. He pops onto this forum regularly, but don't know that he would read this thread...

Based on his published measurements of the sealed and vented loudspeakers he has made, I would say these SS drivers are meeting published spec. Sealed box woofers are few and far between these days. There is always the bigger box plus series resistance trick though...

[JonMarsh]

Well, I've got them on hand now, and will hopefully test this weekend, time being available.

They're as pretty "in the flesh" as they are in pictures.

[PhilDSP]

I'd be quite interested in how you think they sound in comparison with the Revelators. Visually, the tri-petal on the woofer cone would take some getting used to without a grill or covering over the woofer. But of course sound quality is most important...

[JonMarsh]

Keep in mind I've got the aluminum cone version, not the paper version, which is more directly comparable to the 8531 Revelator. That's because I'm not doing a two way, but a Duelund three way with ~300 Hz crossover on the low end. Even SS says it's "optimized for woofer applications in three way systems". They're also making some funny claims about Xmax, because the numbers don't really add up, even though it's an underhung system, as the gap is 20 mm, and the VC height is 8 mm. I don't get a linear Xmax of 9 mm from that. OTOH, with the size of the gap, the performance may still be pretty fair out to 7-9 mm. Just have to see...

Actually, I think this driver "might" be usable in a two way, in the same way that say, an RS180 is- would take some care in the crossover and a lowish crossover point, LR3 or LR4 at 1500, so I'm going to do some testing possibly in that mode, too.

OTOH, the Vifa NE180W would be a lot easier to work with for something like a two way near field monitor, and with 4 ohm Z, a bit more sensitive, too.

Just some other thoughts... mostly I'm hoping for results that will look good in the Ardents, and allow me to consider going from ported to sealed in some fashion or another, and removing the LF zobel from the design (a single sealed Fs will be at a lower frequency and likely not interfere with the crossover as does the upper impedance peak of the ported version).

[JonMarsh]

While the Ardents have been really neglected from a time viewpoint, they've always been in my thoughts and some occasional design review and what if ongoing, based on things I've learned from the Modula Xtremes.

Before switching and trying out the SS woofers, there's one more update pass on the crossover with the Seas ER18RNX which I hope to squeeze in some time to wring out in testing this weekend (evenings) and decide if it's the "final" version of the crossover for that driver configuration.



Then I'll be modifying the cabinets to handle the Scanspeak drivers and updating the crossovers again. But probably not till the second half of May, as I'm basically on a 7 day a week schedule until then.

[Bear]

Then I'll be modifying the cabinets to handle the Scanspeak drivers [..]

:E :E :T

[johngalt47]

Jon,
Some of the resistors in the crossover have "mOhms" for units. Are these milli-ohms?

[JonMarsh]

Some "resistors" are just parasitic components of capacitors or inductors, for a more accurate simulation, or in the case where initial investigations considered placing a resistor as part of the shaping network. This applies to R2, R2, R4, and R8. The "finalized" schematic drawn in Altium Designer doesn't include these "parts". Sorry for any confusion.

[saevox]

6620

If I wanted to use the 6620 tweeter is there an updated schematic for that? Also are there any cabinet plans available other than what is pieced about in this thread? Thank you Jon for all your help, can not tell you how long I have lurked over this project.

[JonMarsh]

Contact me by email or PM and I can send you the cabinet PDF plans. The tweeter crossover was not altered between the two tweeters (they're that similar), but the level pad was adjusted slightly (shunt resistor changed).

[JonMarsh]

This is getting real old...

By two weeks ago, I had a pretty nice pile of new parts....



Some for the Modula Xtreme, some for the Ardent update, to implement this revision of the crossover:



Sadly, all I did last weekend was work in the Office lab on a PFC DUT measurement board, as the tech in Raleigh couldn't get his work done as scheduled, and I got a box of parts and heatsinks (including the wrong ones) Friday last week, and spent all weekend either machining or soldering to put together this, in time for FAE training this week just finished. (and for next week in Raleigh, leaving this Sunday for their early in the AM).



It's been an ugly couple of months- two straight months without a day off. My manager is in Raleigh, and we're having a chat while I'm there.

At this point, the top of my agenda is two weeks off the end of May.

I planning to attend the Northern CA DIY in Davis on next Saturday, the day after I get back at O-dark-thirty at night. Will be lovely to get up at 5 so I can get to Davis on time.... hey, sorry for the Whine and Cheese party, but this is supposed to be FUN, not WORK!

[Hank]

Jon, you are the poster child for work enslavement. When you retire, the company will have to replace you with two unsuspecting newbies. I shan't feel sorry for myself any more while on early morning and evening phone conferences with China and Taiwan, and covering both in person during very compact, intense one week trips. Nossir, I've got it easy compared to you.

[JonMarsh]

Ola Hank!

Well, the last three months have been a real long march, so to speak, but now I think there's some light at the end of the tunnel- still not sure if it's the proverbial freight train! I've been doing some regular middle of the night conference calls several days a week most of the last three months, but that's also off the docket for a while!

Yesterday I installed a new kitchen faucet and spray unit, plus some upgrades for a bathroom.

And TODAY I expect to get to play with that pile of Ardent crossover parts in the picture above- may even take a trip to Woodcrafters for some phenolic for a jig to retrofit the cabinets for the illuminator drivers! And I've started preparation of a web site primarily dedicated to the Ardent for now.

Funny you should mention the hiring thing- we've got two positions we're hiring for, both to take overs some things I'm doing, and not with that much overlap in skill set- but having trouble still finding adequately qualified candidates. Sometimes good experience, but not the right flavor of good experience, even with MSEE's. Interviewed about a half dozen "screened" candidates, but so far, not close enough for either position.

I've also got some new tech development stuff going on, which may lead to me being able to work on something fun- a very high efficiency SMPS for audio power amp plus a Class D design using our 150V or 200V OptiMOS FET's. Got some sims I'll start putting together next week; the power part is an offshoot of work we've been doing for a very compact onboard EV charger- 3.3 KW. PFC stage is about 98.5 % efficient, for example. Using parts we're introducing later this year- Silicon Carbide Power JFETs. The second half of this year may turn out to be OK.... or NOT! Never count the chickens before they're hatched!

Have a good holiday weekend- I'm off until next Thursday.

~Jon

[---k---]

Wow. Jon is still alive.

[CraigJ]

Wow. Jon is still alive.

...and I hope he at least has time for this:

[---k---]

Wireless internet? I thought he needed to relax and unplug??? :P

[Hdale85]

Can't wait for this website to go live Sometime in the next couple of years I think I'm going to try and tackle these for a 2 channel system after we get moved into a bigger place, at which time I will actually have a garage with space to build something like this!

[CraigJ]

Wireless internet? I thought he needed to relax and unplug??? :P

Ha, I knew someone would mention that. I initially wanted to post this pic, but to me, it looked too much like John had to return his bike back to the dealer because he lacked time to ride it.



How about this one:

[ggetzoff]

Looks like Arizona, which is where I ride

[JonMarsh]

Yup. That AZ, during my ride back from Russleville after flying out there on Dec 20, buying the bike, and doing a five day ride back to home in Northern CA.

When I saw that wireless internet sign, of course, my desire was to use that to post live from the road, so to speak- but like a lot of state services these days, it wasn't working. My iPad 3G kept me connected most places and allowed me to make hotel reservations a day ahead once I'd decided how far it was likely I'd go- the first two nights, in Bryant Arkansas just south of Little Rock (the day I flew out and picked up the bike), and Abilene, TX (where I got my 600 mile service at Max's Honda) were booked ahead of time, the weekend before.




Craig is hitting uncomfortably close to home, though, as I haven't had a day off until last weekend since early February, and haven't been to the DMV yet to get the bike inspected and licensed, so I haven't been riding it the last couple of months. We've had a much colder than normal spring, but that wouldn't be any hinderance to this critter, with the weather protection it has. It's very comfortable at speed in 50 degree weather.

Especially with the new used Corbin seat from the guy in Canada:



Today was scheduled to be living room painting, another wall or two, but I've just decided to say, "chuck it", and get out those shiny new crossover parts and see what kind of trouble I can get into with the Ardents. I've also been working on the website design a few evenings this week instead of unwinding with a Blu-Ray. Gotta get back in the groove.

With any luck I'll have another post this evening, if only to confirm or disavow my planned modifications! Plus I've got to get onto the cabinet mods for handling the Scanspeak Illuminator woofers. Good thing Woodcrafter's is open today... need some jig material!

[dar47]

Yup. That AZ, during my ride back from Russleville after flying out there on Dec 20, buying the bike, and doing a five day ride back to home in Northern CA.

When I saw that wireless internet sign, of course, my desire was to use that to post live from the road, so to speak- but like a lot of state services these days, it wasn't working. My iPad 3G kept me connected most places and allowed me to make hotel reservations a day ahead once I'd decided how far it was likely I'd go- the first two nights, in Bryant Arkansas just south of Little Rock (the day I flew out and picked up the bike), and Abilene, TX (where I got my 600 mile service at Max's Honda) were booked ahead of time, the weekend before.




Craig is hitting uncomfortably close to home, though, as I haven't had a day off until last weekend since early February, and haven't been to the DMV yet to get the bike inspected and licensed, so I haven't been riding it the last couple of months. We've had a much colder than normal spring, but that wouldn't be any hinderance to this critter, with the weather protection it has. It's very comfortable at speed in 50 degree weather.

Especially with the new used Corbin seat from the guy in Canada:



Today was scheduled to be living room painting, another wall or two, but I've just decided to say, "chuck it", and get out those shiny new crossover parts and see what kind of trouble I can get into with the Ardents. I've also been working on the website design a few evenings this week instead of unwinding with a Blu-Ray. Gotta get back in the groove.

With any luck I'll have another post this evening, if only to confirm or disavow my planned modifications! Plus I've got to get onto the cabinet mods for handling the Scanspeak Illuminator woofers. Good thing Woodcrafter's is open today... need some jig material!

Good to see you back!

I'm getting torn between the Ardents and this little naked number.

Speedtriple2 Image not available

Kids are gone now no worries do I get the ride or the Ardents, hum maybe both. Looking forward to the site.

[Hdale85]

Here's a vote for both :B

If I do end up getting a bike, a Ducati is very high on the list!

[---k---]

My wife and I refer to my convertible as the "rejuvenator". It is magical to ride a bike or drive a convertible on a nice day. So, I saw forget the speakers and get out on your bike!

[JonMarsh]

Thanks for the votes! Will have to do that later this summer (if we have one here- rain in June in Northern CA? That's positively bizarre...) All you guys go out and get those bikes and/or drive your convertible, K! It is good for the soul.

In other news, the tweaks have been implemented, failed parts have been replaced (from the oscillation from shorting crossover leads that killed one of the Beryllium tweeters back in early 2010), and bass guitar, voice, and piano are now sounding to my satisfaction- both on axis, off axis, and in the next room! :T

Voicing amplifier just a good ole mid Fi 8008 Aragon BB. Some component variations from what's posted above; midrange level is raised, woofer crossover tweaked a bit further, and LCR EQ for midrange reverted to last components (2 mH inductor) and LCR zobel re-installed for mid (don't tell me LspCAD doesn't tell the whole story... :W ).

Any way, I think I'm done wrestling this version!

Now, about those illuminators...

Film at 11 or there about.

[JonMarsh]

Final ER18RNX Update

A quick summary of what I've been doing- updating and finalizing the crossover design with the ER18RNX.

Why changes? Well, even though my past FR response plots looked pretty good, they just didn't sound right at times to me. Mostly the issues I wasn't happy with were in the low mid/upper bass, and overall integration and balance from 100 Hz to 5-6 kHz.

Using experience from the Modula Xtreme's, I took a closer look at the LF and midrange crossovers AND the crossover components...

The Jantzen C toroid core inductors seemed quite promising for reducing DCR- which they do. But detailed testing with a number of different LCR systems, including an HP4192 analyzer showed some inductance non-linearity issues at LOW power levels, not at high- at high, they're not badly behaved for a cored inductor. But at lower current, the permeability of the core is higher, and the inductance higher than the stated value. And it varies where that sets in depending on the value- same core type used for all parts, not different mu cores. So they got kicked to the curve, and the LF network redesigned with a combination of LspCAD and empirical tests, resulting in new North Creek AWG air core coils with new values for L1: 3.0 mH. Now we're cooking with white gas, as we used to say in Texarkana.

The mid crossover was also scrutinized for behavior at the lower crossover point. Earlier I'd believed I would be fine with a 5 mH shunt inductor on the high pass and adjusting the series cap to get the needed curve, but this left the impedance behavior a bit squirrelly at the crossover region. After some simulation and analysis and measurement, L11 is revised to a somewhat spendy 10 mH AWG 15 coil, but the behavior is better and the measured response for the mid alone is where I want it to be. I also looked at some alternative network variations for the upper mid shunt EQ LCR in LspCAD and by measurement, but ended up sticking with the values determined last year.



Put this all together, and the result doesn't look that much different on paper from the last set of sweeps I ran in 2010, but it does sound different- and much better to my ears. Distortion sweep is lower, too.



Note that this was measured with about a 70 msec in room gate, and WITHOUT the felt damper on the front panel- the bump in the 5-6 kHz area goes away with that, but it's pretty OK sounding even without the felt.

Test amp was an Aragon 8008 X3; speaker cable Cardas Golden Presence.

Here's a pic of the updated schematic. Email or PM me for a nice PDF.




Next step is cabinet mods/update to evaluate the Scanspeak illuminator woofers.