Calliope CC Mk2 Design Study

[JonMarsh]

Exciting times! Steve is considering promoting (?) me to Senior Speaker Designer for Home Theater Products at SMJ! That's of course, contingent on whether I come through on these new HT oriented projects! The only downside seems to be that the exotic two channel audio stuff will all be handled by Evil Twin, Lord and Master of Dark Force Acoustics Labs.

Of course, he may keep me on a tight leash for a while, doing things like overriding tweeter choices, in favor of more consistent esthetics. But as National Sales and Marketing Director, he's gotta make the hard decisions, right?

Now, this project is the Mk2, which basically means that a design was well along the way (around the time of the Ardent D), then due to some considerable supply chain issues and increasing performance requirements plus more aggressive targets for cabinet size, etc, well, not just something had to give, but basically the old design was torn up and we started over.

So this thread will explore more complete driver measurements and a prototype build and evolution/evaluation of that test article.

This is a sort of conventional three way, but it will borrow crossover concepts from the mid priced Ardent, not the Neo D CC, though the driver layout will be quite similar, but with a TTC type cabinet ultimately.


At this point, the driver line up is as follows:

• Woofers: Esoteric ES180TiA, 2 each 8 ohm
• Midrange: Faital 5FE120, 8 ohm
• Tweeter: BlieSMa T25A-6

Key design targets:

• Compact - expected LF volume for 2-7" woofers is ~15-18L.
• Low intrinsic driver distortion- lower than typical HT products
• High SPL capability within the intended bandwidth - a function of motor linearity and voice coil/cone power handling, and operational bandwidth of the drivers based on crossover frequencies, and power as specified typically by AES standard.
• Fits in the box!!

Pretty straightforward stuff, just some careful work on driver selection and evaluation and system design using known/proven techniques.

[JonMarsh]

A test cabinet for the dual ES180TiA is near completion...


Why the ES180TiA? Well, I have good test data from a while back on the predecessor baby brother, the ES140Ti, which really could be called the ES150Ti, based on the actual size, and the linearity through 1kHz and distortion behavior were very good. A lot of this I attribute to the 3" VC Neo motor with a demodulator ring that all four versions share in common. For the TiA variations of both drivers, they tweaked the cone design to improve the linearity of SPL response above 1kHz, to make two way systems easier to design. However, this isn't a two way, and either version would likely work fine.










This morning I cut out the test baffle and back bevels, and glued a brace; assembly should be finished later today.

[theSven]

Exciting times! Steve is considering promoting (?) me to Senior Speaker Designer for Home Theater Products at SMJ! That's of course, contingent on whether I come through on these new HT oriented projects! The only downside seems to be that the exotic two channel audio stuff will all be handled by Evil Twin, Lord and Master of Dark Force Acoustics Labs.

Of course, he may keep me on a tight leash for a while, doing things like overriding tweeter choices, in favor of more consistent esthetics. But as National Sales and Marketing Director, he's gotta make the hard decisions, right?

Now, this project is the Mk2, which basically means that a design was well along the way (around the time of the Ardent D), then due to some considerable supply chain issues and increasing performance requirements plus more aggressive targets for cabinet size, etc, well, not just something had to give, but basically the old design was torn up and we started over.

So this thread will explore more complete driver measurements and a prototype build and evolution/evaluation of that test article.

This is a sort of conventional three way, but it will borrow crossover concepts from the mid priced Ardent, not the Neo D CC, though the driver layout will be quite similar, but with a TTC type cabinet ultimately.


At this point, the driver line up is as follows:

• Woofers: Esoteric ES180TiA, 2 each 8 ohm
• Midrange: Faital 5FE120, 8 ohm
• Tweeter: BlieSMa T25A-6

Key design targets:

• Compact - expected LF volume for 2-7" woofers is ~15-18L.
• Low intrinsic driver distortion- lower than typical HT products
• High SPL capability within the intended bandwidth - a function of motor linearity and voice coil/cone power handling, and operational bandwidth of the drivers based on crossover frequencies, and power as specified typically by AES standard.
• Fits in the box!!

Pretty straightforward stuff, just some careful work on driver selection and evaluation and system design using known/proven techniques.

You had me at Exiting Times! I'm curious how the voicing with the Bliesma T25-A-6 would blend with the Peerless that is in the Calliope MT? Would one notice a difference or it all blends together like one harmonic family?

[JonMarsh]

I'm a principled believer that if the design is consistently executed, (this means in regards to axial and power response, a function of off axis behavior) and the intrinsic behavior/quality of the drivers is similar, you should wind up with a very similar gestalt.

The DA25 with it's various crossover concepts is bent to my desires, so to speak, and winds up being quite smooth and free of upper breakup mode resonance amplification of distortion products. This is due to the series notch filter.

The BlieSMa T25A has a pronounced ultrasonic breakup mode also, though at a somewhat higher frequency- the cure is basically the same though the specific details would differ.

Now, if you want to try the high priced spread, there is the T25B-6, which I have a couple of and which Solen stocks. It seems to have a near optimum alloy and dome shape, and like the Scanspeak 6640 used in the Wavecor Ardent, it hardly has a significant breakup mode resonance as regards the SPL plot or the impedance plot.




This roughly doubles the price of the tweeter, but there's no arguing with the results. OTOH, it's a lot more money than a notch filter. It's a bit flatter overall too.


The dome size and profile tends to set the stage for the polar response. The DA25 with it's very flat clean baffle plate is likely class leading in that regard. The BlieSMa T25 series is fairly close. The problem with the DA25 (which I have about a half dozen new in box pairs) is that it's just too big for a CC application in a reasonable size baffle. We didn't consider it previously for the same reason.


I plan to try building a set of these for myself, too, but as an experiment, with the GRS PT2522c. With the extended top end response and polar behavior of the 5FE120, I think it may be feasible with a 2800 to 3000Hz crossover.

This is the 45 degree off axis response, which due to it's construction similar to the Neo3 PDR, with only the center set of slots contributing most of the output, the horizontal source width is quite narrow and the off axis high frequency level is superb.

[CraigJ]

I plan to try building a set of these for myself, too, but as an experiment, with the GRS PT2522c. With the extended top end response and polar behavior of the 5FE120, I think it may be feasible with a 2800 to 3000Hz crossover.

This is the 45 degree off axis response, which due to it's construction similar to the Neo3 PDR, with only the center set of slots contributing most of the output, the horizontal source width is quite narrow and the off axis high frequency level is superb.

Any chance this has something to do with the exciting PS Audio FR30 clone, or the upcoming FR20?

[JonMarsh]

Now, if I didn't think otherwise, I'd suspect you have some of Evil Twin's Imperial informants checking up on what's going on in this neighborhood, as well as in Steve's....

Steve has given all the high falutin two channel work to Evil Twin, but I have it on reliable authority from overheard conversations and intercepted emails that several variants of that project are under consideration... ET is not the only one with spies. Rumor has it that besides a version with four RSS210HF-4, there is a dual woofer PuriFi version (awaiting delivery of 8 samples from PuriFi- you know, they basically take July and August off over there), and for the thrifty or impecunious types, a dual RSS210HF. Reportedly the new GRS mounting kits go some ways to helping to deal with diffraction issues with this tweeter, which can be something of a stinker with a bare front mounting. The measurement above is reportedly with a heavy felt diffraction control panel for both the midrange and tweeter. I think they have one more spin on the rear enclosure design to get to the desired point... but you know how that can go. It's not as touchy as building high power rockets, but it's a lot trickier than a typical cone driver.

[theSven]

I'm a principled believer that if the design is consistently executed, (this means in regards to axial and power response, a function of off axis behavior) and the intrinsic behavior/quality of the drivers is similar, you should wind up with a very similar gestalt.

The DA25 with it's various crossover concepts is bent to my desires, so to speak, and winds up being quite smooth and free of upper breakup mode resonance amplification of distortion products. This is due to the series notch filter.

The BlieSMa T25A has a pronounced ultrasonic breakup mode also, though at a somewhat higher frequency- the cure is basically the same though the specific details would differ.

Now, if you want to try the high priced spread, there is the T25B-6, which I have a couple of and which Solen stocks. It seems to have a near optimum alloy and dome shape, and like the Scanspeak 6640 used in the Wavecor Ardent, it hardly has a significant breakup mode resonance as regards the SPL plot or the impedance plot.




This roughly doubles the price of the tweeter, but there's no arguing with the results. OTOH, it's a lot more money than a notch filter. It's a bit flatter overall too.


The dome size and profile tends to set the stage for the polar response. The DA25 with it's very flat clean baffle plate is likely class leading in that regard. The BlieSMa T25 series is fairly close. The problem with the DA25 (which I have about a half dozen new in box pairs) is that it's just too big for a CC application in a reasonable size baffle. We didn't consider it previously for the same reason.
jun

I plan to try building a set of these for myself, too, but as an experiment, with the GRS PT2522c. With the extended top end response and polar behavior of the 5FE120, I think it may be feasible with a 2800 to 3000Hz crossover.

This is the 45 degree off axis response, which due to it's construction similar to the Neo3 PDR, with only the center set of slots contributing most of the output, the horizontal source width is quite narrow and the off axis high frequency level is superb.

Thanks for the insight. You would think I would remember how much space the DA25 takes up from my last two builds, yet I'm still a total junior here in that regard. That's part of the experience getting to learn and discuss this and hope someday it all makes sense in my head 😂.

[CraigJ]

Jon,

Thanks for the response and apologies for the slight thread drift. Looking forward to Steve and Evil’s development of the Fr clones and sincerely hope it becomes public to this impecunious young Padawan.

[JonMarsh]

Yeah, there is more than a little excitement about the FR Klones, when you think about it, it's a pretty strange story with my start down that path and then Steve letting me know what Chris and the crew at PS Audio had done. I see nervous looking droids scurrying around in the work shop, when the Dark Lord is not completely satisfied with the behavior of the latest attempt at optimizing the midrange enclosure systems, but I get the impression that it's getting pretty close, and the concept and understanding seem viable.

[JonMarsh]

The next step in the Calliope CC Mk2 is getting a basic POC test enclosure built- and of course you first have to have a plan! In spite of a lot of other chores and distractions today, that's getting pretty close, just need to take care of the midrange enclosure and internal bracing, which will probably be combined, like the Wavecor Ardent.

This is at the level of what I would call a 3D modeling sketch, and some of the design choices were driven by what materials are on hand- such as bamboo board, and some extra pin block maple ply...

Yes, it may be that the world doesn't really need another three way center channel, but then, duck folks, because it's coming anyway!

And interestingly, there was another 6RS140 "open box" at PE available, so I ordered that today... technically, I'll have enough for two MTM mains and two rear channel speakers.

[Paul K.]

Jon, I wanted to send you a private message but your mailbox is full. I'm currently trying to decide which of the ES woofers I want to use in a new TMWW project and was wondering if you would share T/S measurements you presumably made for the ES140TiA and ES180TiA?
Paul Kittinger

[JonMarsh]

I'm on a "Blitz Trip" on the road (in CA today) but when I get back I can certainly do that.

Note, the ES140's I have are all the original version, for which the cone has been modified in the TiA version to smooth out the top end. Probably no change to T/S, but ya never know.... The ES180's are all the newer part.

Stay tuned...

[JonMarsh]

BTW, Sven upped my PM allotment to about a gazillion messages, so shouldn't be any problem with a full PM inbox again.

[Paul K.]

Jon, since the ES140s you measured and used were the first version and I would be using the current version, I don't really need your T/S measurements as long as those and the published values were pretty much in agreement such that a box designed based on the published measurements would still be correct if your measurements would result in the same, or virtually the same, performance. Also, if I proceed with the project I'm currently considering I will use the ES140s and not the ES180s, so I don't need your measurements at all for the ES180s. If you decide to send me your measurements for the ES140 "just because", I'd then also would want to see the published values if you have them, like on a data sheet.
Thanks a bunch,
Paul

I'm on a "Blitz Trip" on the road (in CA today) but when I get back I can certainly do that.

Note, the ES140's I have are all the original version, for which the cone has been modified in the TiA version to smooth out the top end. Probably no change to T/S, but ya never know.... The ES180's are all the newer part.

Stay tuned...

[Paul K.]

BTW, Sven upped my PM allotment to about a gazillion messages, so shouldn't be any problem with a full PM inbox again.

Well I think you PM allotment of about a gazillion was waaaaay short because I tried to send you a PM yesterday and your box is still full. Anyway I would greatly appreciate your input on your measured T/S values for the original ES140s as compared to the published values. If the two sets of T/S values were essentially identical, or if they weren't but the differences would be essentially self-correcting such that an enclosure designed based on the published values would still result in essentially the same expected performance with the different T/S values of the actual drivers.
Paul

[JonMarsh]

Well, our big guy here told me it was increased 5 fold. Been really tied up with time critical domestic infrastructure, but I think I can get to this tomorrow or Monday morning. Sorry for the delay, but heck, I haven't even had time to hook up the TV since we moved in! (But that tells you how important TV is to me, too!). And me finishing building the Gazebo is really important to my wife right now...

🤣

[Paul K.]

Well, our big guy here told me it was increased 5 fold. Been really tied up with time critical domestic infrastructure, but I think I can get to this tomorrow or Monday morning. Sorry for the delay, but heck, I haven't even had time to hook up the TV since we moved in! (But that tells you how important TV is to me, too!). And me finishing building the Gazebo is really important to my wife right now...

🤣

Okay, I understand you are very busy and you definitely want to keep your wife happy. Just wanted to make sure I hadn't been forgotten.
Paul

[JonMarsh]

I'll follow up soon on the PM thing- yesterday was a massive "domestic" work load, but this AM I will finish assembling the known volume test cabinet for testing the ES180TiA.

Man, retirement is just too much work!

[JonMarsh]

Test cabinet is put together and the epoxy is setting... still a lot of "domestic" stuff going on, but should get back to this sometime this afternoon or evening.

[JonMarsh]

OK, time for some data now.

I prefer the known volume method for T&S data extraction, it removes uncertainty about the added mass, and I don't keep clay like that around anyway, nor a scale to measure it.


Free air impedance:






Closed box impedance, driver inverted, 15L (driver volume does not subtract from enclosure volume)






Calculated parameters:


MANUFACTURER: Dayton Audio MODEL: ES180TiA
Fs. 35.16 Hz
Re 7.40 Ω
Sd 132.7000 cm²
Qms 7.72
Qes 0.52
Qts 0.48
Cms 0.7247 mm/N
Mms 28.2808 g
Rms 0.81 Ωm
Bl 9.47
VAS 17.8190 L
Zmin 7.70 Ω
L1kHz 0.48 mH
L10kHz 0.30 mH

[JonMarsh]

I swear this year should be called the year of molasses, because of how long it seems to take to get peripheral things done to get to the next desired phase of work! It's my sense that Steve is wrestling with this, too.

The Gazebo is 95% erected, and I'm hoping to get the big boxes of patio furniture out of the garage shortly, so that I can setup for series routing and cabinet building before the month is out. I've mastered creating piece part 3D drawings in Shapr3D, and I'm eager to make some carefully contained sawdust!

I've got materials to build two POC cabinets for the CC; one will be with the T25A-6 or T25B-6 (both on hand) and the other may have some version of modified PT2522c or PT-2522. I've built up a few parts with different techniques, and currently have 4 using the PT-2522 that I'm fairly happy with and for which the results are pretty consistent. For the moment I'm not ready to share details, until I feel like it's really locked down. Well, other than with Steve, my resident skeptic! I think Steve is actually. from Missouri (the "show me" state), though he claims England. Can't say I've ever heard a Missouri accent out of him, though.

Using data from an assortment of cabinet, none the same as the planned design, it's been straight forward to develop a POC crossover approach (stolen from the Wavecor Ardent, basically), while including modeling the driver acoustic offsets in the Z axis (necessary for the so called LR3 technique I use to actually work).

This one is for the modified PT2522c. It's not hard to work with, but the modified PT2522 is flatter and easier to work with, and the assembly process and materials are more repeatable.





The LF data was not taken especially carefully, but just to match the Z axis distance and level of the mid and tweeter. Crossover adjustments would be needed due to baffle loading differences on the target design. This is just a fast Proof Of Concept, but the simulated response is in a nice envelope of +/- 1.5 dB from 500Hz to the top end, indicating to me that even this configuration with an unconventional driver merits further investigation.


Oh, and we had a "dry" thunderstorm yesterday afternoon, with sustained winds at 45-50 MPH and gusts to 70; you should have seen what some of the neighbor's new young trees were doing, but the new gazebo came through fine. The chosen location is well sheltered from the prevailing winds here, and having gotten the missing parts from China, the remaining assembly was 95% finished yesterday morning.

Now, about that garage working space I need...

[Steve Manning]

I swear this year should be called the year of molasses, because of how long it seems to take to get peripheral things done to get to the next desired phase of work! It's my sense that Steve is wrestling with this, too.

Well, other than with Steve, my resident skeptic! I think Steve is actually. from Missouri (the "show me" state), though he claims England. Can't say I've ever heard a Missouri accent out of him, though.

Steve can certainly relate.

The resident skeptic was contacted by ET ..... I was told no free rides for Mr Marsh, make him work for it! 😎

[JonMarsh]

Steve can certainly relate.

The resident skeptic was contacted by ET ..... I was told no free rides for Mr Marsh, make him work for it! 😎

🤣

Yup, that's how this summer has been going!

[JonMarsh]

There has been a lot of, let's say, family drama the last 30 days, and my daughter is in the hospital, possibly another 2-3 weeks, and I may need to go to CA to help out with childcare.

I still need to build a test box for known volume method that will work with the ES140Ti. The specs overall for the two drivers (ES140Ti and ES140TiA are considerably different, especially considering that all four of the ES drivers reportedly have the same motor assembly, that is, magnet assembly and voice coil.

I had an enclosure on hand that would work with the ES810TiA, so I ran that test, so you could make a general comparison of how close they might test compared with specs. As you can see, the tested correlation is somewhat mediocre. Is it a problem with the Clio setup and methodology? or with the correlation between data sheet values and production parts?

My preferred methodology is to use either measured T/S parameters, or if I can't do that conveniently, factory data (confidence varies with vendor) and then build a basic LF test cabinet to model my target design, and test it. Of course, the most basic check is the impedance curve and free air resonance, as well as correlation with the factory supplied Fs curve. That isn't always a given.

I do not have any ES140TiA, only the older version, and considering the differences in published T/S parameters, I would say the "reliable" thing would be to test the new one, not an old one- but at this time I do not have any plans to purchase samples of the new one, as the smallest CC concept is now relatively low priority.

This simple chart highlights the differences - I can't even find an ES1480Ti data sheet, PE has taken it down. But these numbers come from my early Unibox modeling.




These numbers, in comparison, don't really make sense to me. Especially not the sensitivity numbers versus cone mass. Obviously, the VC are different, given the change in RDC.

Another project soaking up time the last few weeks was preparing architectural landscape drawings for the contractor, who should start work this Friday on the side and backyards.

I never dreamed retirement would be this much work...

[Absinthe]

Sorry to hear about your "family drama." I hope your daughter recovers without complications and things begin to return to normal (whatever that is these days?)

[theSven]

Sorry to hear about your "family drama." I hope your daughter recovers without complications and things begin to return to normal (whatever that is these days?)

I second that. Hope for a fast healthy recovery.

That is interesting that the specs are so different between the speakers. It's like a box of chocolates I suppose. You never know what you're going to get...

[JonMarsh]

Well, Paul K received the ES140Ti DS and forwarded it to me... for, me, it just raises more questions, on two grounds- the T/S parameter list, and the spec'd SPL response. Mind you, in regards to the latter, I've measured several samples of this part, and participated as a judge for a Design contest using this driver and a specific tweeter at a DIY Audio California meet. My measurements of the ES140Ti and those of the supplied design files for that contest way back when agree quite well, and they don't look anything like the PDF Paul just received from PE.


So, here for reference is the ES140Ti DS Parameter chart, with the ones that didn't match what I'd entered and calculated in Unibox back when these first came out:








And a comparison of the data sheet SPL






With typical data I measured and very similar to what another person measured for the design contest:

[Paul K.]

Jon, I don't know what to say about the discrepancies you've observed for the ES140. I did notice in your Post Spaced Out at least one oddity regarding the T/S values shown for the two ES160 variations; Sd is shown for both drivers with a value of 98.5 cm2, when it ought to be about 130 cm2.
Paul

[JonMarsh]

I caught that later, too. The chart above should say ES140Ti and ES140TiA, not ES180. I'll update that later today. We have the contractor here working today on the landscaping. The data is all for the 140 versions.

Were you able to save the attachment when I sent the data sheet back to you?

Given you're using the ES140TiA, all of this is somewhat a moot point...

This is the parameter set I used for the ES180TiA; entered values on left, calculated values on right.

[Paul K.]

Jon, I haven't received anything from you.
Paul

[JonMarsh]

Note the below image capture from my Apple Mail App...




The address was just copying the link from your incoming mail. Apple Mail normally displays the first page of a PDF on screen in the mail.

Maybe, check your junk mail folder?

[JonMarsh]

Steve Manning and I send lots of mails back and forth. Same with quite a few other friends and former colleagues. Also, having PE accounts, we get stuff form them and have no problem downloading attachments. Maybe there’s an issue with your mail client?

[Paul K.]

Steve Manning and I send lots of mails back and forth. Same with quite a few other friends and former colleagues. Also, having PE accounts, we get stuff form them and have no problem downloading attachments. Maybe there’s an issue with your mail client?

I checked and it did not show up in my Junk or Trash folders, nor did it show up in the Spam or Trash folders in the server for my Spectrum/Roadrunner email account. I have saved the original email from the source of the data sheet and I had already printed out a hard copy. Anyway I appreciate your efforts and help Jon.
Paul

[JonMarsh]

I sent a fresh email to you again, Steve on the address line, and double checked the address from your original mail- but I'd used it to add your email address to my address book.

Let's see if you get it this time, after your mail client update.

[JonMarsh]

Steve has received the email.