Ardent D (aka Kurosawa Jr) Winter Camp

[Scottg]

It's a HUGE scam from taxapayer money. My guess is over 100 BILLION has been paid-out with very little improvement nation-wide.

I live in a market where there are several providers around me for high'ish bandwidth (though only a few areas have gig service), but in my subdivision there is only 1 provider (for anything over 15 megabit) - and they can and will *jack-up prices without warning and you can't do anything about it other than say "cancel service" and then "fight" for a more reasonable price. Additionally, no fiber. My subdivision is that sort of pissing-pond problem without any regulation.

*we went from $60 a month 100 megabit down 10 up to being charged for overage fees totaling like $150 because we'd been "switched" from unlimited without our knowledge. Then we hit the "cancel service" run-around and finally spoke to a state-side re-mediator and now it's the same $60 for 350 megabit down and 40 up unlimited (..though even then we had to carefully state that we had our own modem or they would have tacked-on another $10 a month). Of course meanwhile that very same service provider (Optimum) is offering fiber near us (but not in our subdivision) for $50 a month with a $200 prepaid Visa card and a 1 year term OR 200 megabit for $30 AND the $200 Visa card which is available to us BUT for New Customers ONLY, which we can never be because they are the only viable service provider (unless we go without internet for 6 months).

Price-wise I'm fairly happy with where we are now, but it was a pain in the @ss to get to that point, and it likely will be again with their price-hike shenanigans. (..and for TV we had DirecTV before then with exactly the same problems, and dumped them for streaming with Sling Blue - which is what caused the higher use for that month.)

Moreover the ISP definitely throttles certain websites like Youtube. (Though so far our other (Sling, IMDB, Pluto, Xumo, etc.) streaming for TV's hasn't been that bad.) At some point though I might have to setup my own private "VPN" through something like Linode with providers near my City area.

-yeah, it's a sore spot for me! Sorry about the off-topic rant. ops:

[JonMarsh]

It's a big problem in many parts of the country, and like healthcare, what we get for our money is unimaginably inadequate compared with much of Northern Europe. Italy might be a bit different...

No worries. I fully sympathize and understand. We signed a contract today for a new home which will probably be ready about this time next year, (have already driven over the streets where it will be- streets and utilities are all in, the delay tends to be availability of construction materials and skilled trades workers- but this is out in a pretty rural area on the north side of Treasure valley (Boise area) and I expect I will have SpaceX Starlink in my future, as I rather doubt there's any fiber in this town of 7500 people!

[JonMarsh]

Getting back on topic, now that our flurry of real estate activity is abating, we can get back to more mundane matters- like the Ardent D updates and evaluations.

This is how things are looking now after a number of iterative updates- the easily removed PR's and crossover make this less painful than it would be for some of the other builds that have gone down. Famous last words, I don't think we'll be changing any thing else- I have three levels of build quality and component selection identified for Steve, but as you can see there are a bit more parts than a three way done with straight 2nd order crossovers would require, but I think the results are... interesting!



In a sense, these kind of have three characteristics that pop out at you in different ways, as a result of the drivers and design. I'll just outline tonight, and go into details shortly.

1. First, these are a three way speaker with all dome drivers. And if you've paid attention to the posted polar plots for individual drivers we've taken, this means the off axis response should be quite wide and uniform, and it is. Seems almost too good to be true at 60-70 degrees off axis.
2. Next, let's just say that the more I listen to the AS168, the more I'm impressed with it's neutrality and extension and control. I know Accuton put a lot of effort into engineering speific characteristics into this part, and after working with the AS168, I am really curious to hear the AS190 in action. The tuning in this cabinet with the Dayton PR's has worked exactly as hoped and planned- I'll comment on evaluation cuts as well as the equipment setup. On some cuts the bass may almost sound too neutral- no extra fatness or warmth. But very good pitch definition. But then on other cuts, there is a combination of control and power and pitch definition which makes me hear the cut in a new light compared to past experience- this occurred on James Taylor's Gaia, for example, and Stanley Clarke's import album "The Bass-ic Collection"- for example, the first cut, "School Days", is on this album the only CD version that approaches the dynamics of the original vinyl. But the 2nd cut, "Wild Dog" really surprises on some tom-tom rolls from small to large with the presence and power and sheer musicality on display. It seems truer to me to the likely intent of the producers of the recording- and it's just a heck of a lot of fun, to boot!
3. While the sound field all around the room is quite pleasant and well balanced, when you sit down in a classic listening position, even in this rather live and reverberant room, it's like the imaging quickly locks/snaps into place, though obviously what you get is dependent on the source characteristics. This will deserve some exploration with familiar material, but it is almost startling the way this happens as you move into the listening sweet spot- I say that even in comparison to other designs and builds I've worked on. I suspect it has much to do with the Duelund crossover behavior, which is more fully realized in this build than past efforts. (I.E., it's a bit of a b*tch to get right...)

A lot more to come, including 8x10's with arrows and circles and paragraphs on the back...

[Alaric]

A lot more to come, including 8x10's with arrows and circles and paragraphs on the back...

For an Arlo quote, you, sir, have triggered the ancient tradition. You get a dancenana for one of my all time favorite songs.
:dancenana:

[JonMarsh]

Dates us both, huh?

:banana:

[JonMarsh]

well, I dug up some more CD's from my storage to rip, and one's I couldn't find I ordered on Amazon- heaven help us if the only way to get CD's was to buy locally!

Have started some more listening sessions, though this morning I must attend to more mundane chores... Joan Osborne's greatest hits was a treat last night!





And I've got some of the heavy guns electronics for the music server here to hook up, like the Alpha USB... now let's get those chores out of the way!

[Oskar2905]

Hi. I'm making speakers using fountek fw168k and da25tx. I would like to use same RC ultrasonic filter on da25tx like


in this xover but there's no coil resistance listed on schematic and I don;t have an equipment to measure anything above 20kHz. JonMarsh would you be so kind and share what coil you're using ?

BTW. If anyone wants a quality made and good sounding midwoofer I can recommend this fountek gladly .

[Alaric]

Dates us both, huh?

:banana:

Lil' bit. I just looked at the avatar dates and I think I need a drink (Coffee at this hour). My daughter was almost 6 years old when I wandered in here. She's now 21 and lives in a cabin in the woods and hunts her own food. Where in the hell did all that time go???
Back on topic, while this project exceeds my budget by orders of magnitude, I really enjoy the heck out of the learning curve. So thank you for sharing your knowledge and experience. :thanku:

[JonMarsh]

Lil' bit. I just looked at the avatar dates and I think I need a drink (Coffee at this hour). My daughter was almost 6 years old when I wandered in here. She's now 21 and lives in a cabin in the woods and hunts her own food. Where in the hell did all that time go???
Back on topic, while this project exceeds my budget by orders of magnitude, I really enjoy the heck out of the learning curve. So thank you for sharing your knowledge and experience. :thanku:

I can so relate... look at my join date! Over 20 years now! And when I think about what was going down back then, versus now, it feels like much more.

My daughter was 15 back then, and now she's carrying her 3rd child, another girl. The middle one was a boy, though, he will literally be the next Doc Emmett Alexander.

And Stella (the oldest) is looking forward to our visit this summer so we can watch Star Wars and Back to the Future movies (this was her own unprompted request to her mom, just based on the stuff he overhears. She'll be 4 in August.

[JonMarsh]

Hi. I'm making speakers using fountek fw168k and da25tx. I would like to use same RC ultrasonic filter on da25tx like

in this xover but there's no coil resistance listed on schematic and I don;t have an equipment to measure anything above 20kHz. JonMarsh would you be so kind and share what coil you're using ?

BTW. If anyone wants a quality made and good sounding midwoofer I can recommend this fountek gladly .

I'll get the info for you from the Calliope two way I created for Steve Manning, which uses the same tweeter and technique. I've got real estate meeting with our builder most of the day, so it may not be until late, or late this week (travel for three days starting tomorrow).

The Fountek is a nice looking part, kind of like the FW135 I've tested, but haven't used in a design yet.








The actual production FW135H parts have a grille like the FW168K

[JonMarsh]

Looking at my day, if I don't get this done right now, it won't happen until late in the week!


Here's the crossover schematic: Note, it takes two of the 50uH coils wired in parallel to get the target value, unless you have an LCR meter and can unwind one of them to the correct inductance.


PE part number is 257-798.





Here's how well this measured, including distortion. You can see it essentially completely kills the 2nd Harmonic resonance amplification distortion component.





As you can see, the DA25, implemented properly, is a honey of a tweeter, with the maximum 2nd harmonic distortion products at 90 dB SPL at about -60dB. 3rd harmonic is 6-8 dB lower. That's pretty much world class.

[kevinlin1013]

[ATTACH=CONFIG]29926[/ATTACH]

I have a question. In the SW223 woofer part , it said Ardent DATA but in the Ardent Wavecor there are two 4ohm in serial.
On the other hand , this crossover is two 8 ohm in 3.5 way parallel design. Is my understanding right? Thanks a lot!

[JonMarsh]

Hello Kevin!

The "Wavecor Ardent" and the "Ardent D" are two VERY different systems!

The Wavecor Ardent used two Wavecor SW223 4 ohm drivers in series with our concoction of an LR3 crossover (not something created by Linkwitz and Riley) to get a fairly easy to drive load (note how flat the total impedance curve is- apart from sensitivity, relatively tube amp or non feedback solid state amp (Ayre Acoustics, for example) friendly.

The Ardent D is a very different system experiment, using Accuton Cell woofers in combination with 10" passive radiators with the effective Z axis acoustic origin in the mounting plane, making them very suitable for the Duelund type crossover ("D" designation) in conjunction with tweeter and midrange with similar Z axis acoustic origin.

Note also that the build I did was essentially an acoustic POC (Proof Of Concept) for the planned SMJ TTC cabinet design.



This was supplied by Steve in a kit form to Sven, as documented in a thread on HT Guide.

I'll dig up the last VituixCAD files for that project and post later today- they're on another computer which isn't handy to access at the moment.


I've been considering the possibility of updating the midrange on my original pair of the Ardent D to the BlisSMa T74, but availability has been sporadic- not even listed on the BlieSMa site for a while, and they still aren't, but they are back in stock at Solen in Canada.

[kevinlin1013]

This may be illegal technology to use in this quadrant, bypassing the normal net systems, but the quality should be better...

Note that this is the 3.5 way configuration with "warm" voicing...

I suppose this SW223s are 8 ohm version. (Ardent Wavecor is using 4 ohm version wiring in series)
Is it possible to sure the data of 8ohm version SW223 on baffle? Thank you so much!

[JonMarsh]

No, these are the SW223 4 ohm version, wired in series.

[kevinlin1013]

Thanks for your quick reply!
But from the schmatic , it is a 3.5 way topology (DA25+RS52AN+SW223 not the Wavecor Ardent).

[Evil Twin]

The Project Development continues...

Sometimes what is done... it not truly finished.

In this case, transit damage due to high G maneuvers to avoid stellar debris has resulted in damage to both Morel EM1308 dome midranges. And in a quirk of fate, or intervention of the Force, the vintage storage of RS52AN dome midranges was located when completing the relocation of the last of stored components to the new sector facility.




Now, as an aging Sith Lord, I am certain the the wisdom or prudence of replacing a $200 dome midrange (Morel EM1308) with a $40 dome midrange (RS52AN) would be widely questioned, but it should be noted that the EM1308 was the only feasible form factor replacement for the RS52AN when it was discovered in 2020 that new production parts consistently had a "new" feature which must be considered a fatal flaw- a surround resonance causing a serious disturbance in the frequency response, rendering the drivers useless for an advanced design.




This defect is easily decidable just in the impedance plot, due to the sharp narrow peak at 2kHz.






Yet, unquestionably the RS52AN worked most satisfactorily in the 2008 design of the NeoD CC, with extended smooth response, due to the well designed aluminum dome, and robust handling and protection, due to the grille configuration.





The 7 vintage parts I have are tested and free of the defect evident in all samples obtained in 2020 and 2021.







Note that with the now "standard" approach of breakup mode suppression with a notch filter, the resonance amplification of distortion products is eliminated.

[theSven]

The Project Development continues...

Sometimes what is done... it not truly finished.

In this case, transit damage due to high G maneuvers to avoid stellar debris has resulted in damage to both Morel EM1308 dome midranges. And in a quirk of fate, or intervention of the Force, the vintage storage of RS52AN dome midranges was located when completing the relocation of the last of stored components to the new sector facility.




Now, as an aging Sith Lord, I am certain the the wisdom or prudence of replacing a $200 dome midrange (Morel EM1308) with a $40 dome midrange (RS52AN) would be widely questioned, but it should be noted that the EM1308 was the only feasible form factor replacement for the RS52AN when it was discovered in 2020 that new production parts consistently had a "new" feature which must be considered a fatal flaw- a surround resonance causing a serious disturbance in the frequency response, rendering the drivers useless for an advanced design.




This defect is easily decidable just in the impedance plot, due to the sharp narrow peak at 2kHz.






Yet, unquestionably the RS52AN worked most satisfactorily in the 2008 design of the NeoD CC, with extended smooth response, due to the well designed aluminum dome, and robust handling and protection, due to the grille configuration.





The 7 vintage parts I have are tested and free of the defect evident in all samples obtained in 2020 and 2021.







Note that with the now "standard" approach of breakup mode suppression with a notch filter, the resonance amplification of distortion products is eliminated.

Ouch, that hurts seeing that mid cone smushed! I'm clearly interested in how this progresses. Will await for the next update from the dark lord.

[Evil Twin]

In spite of many distractions and other commitments, some considerable progress was made over the weekend in assessing the current state and preferred directions for design- of course, working with a Duelund concept system simplifiers some considerations, as each driver has a clear cut target for amplitude and phase response, and in principle, one should be able to complete a replacement with a new driver and crossover which fits in appropriately with the original design.

Of course, that is theory, and this is real life...




Some basic details, like a problematic threaded insert which needed to be removed and replaced can complicate the progression of tasks.


However, that basic concept is still valid, and though both midranges have near identical Fs, (335 Hz) the Q differs, and the EM1308 has a problematic resonance lower in it's response curve, while the vintage RS52ANs have the typical hard dome break up mode, in this case, at about 12 kHz, not unreasonable for a 2" dome midrange.

In the end, there was little overlap for the two midrange crossover configurations...




An adequate BOM was developed; due to availability issues, only high grade Litz wire options were available for some component values, so this could be considered a "premium" example. It is somewhat incongruous to see a crossover priced at nearly 10x the cost of the driver, but this is more a reflection of the value proposition of this driver, as the original EM1308 is far more expensive.



Components have been ordered for a test build, and that effort should commence by the end of the week.

[JonMarsh]

Well, this has been the season of distractions and extra work and necessary tasks (like eye exams and ordering new glasses for all), but the test build layout was completed, and the first build for the NIB Vintgage RS52AN constructed.


I know, I know, I shouldn't be too proud of this technological terror, considering how much money is being spent on an optimized crossover for a set of vintage $50 midranges. BUT THEN, think of what the AS168 woofers cost now, and then it doesn't seem so out of line.


The build plan:





The actual build:




Tomorrow I'll be setting up the Mac Studio as the Fuzzmeasure cart computer... because it has been displaced from my primary desktop, as the True Technological Terror is the new M4 Mac Mini Pro+ (the version with the extra CPU and GPU cores).

[JonMarsh]

Houston, we are test firing the crossover with new vintage RS52AN, in preparing for full system liftoff (test and tuning).

A reminder, this is how the RS52ANs from PE looked back in 2020, brand new purchase, and not fully fixable even with substantial surround modifications. I must have gone through six of these trying to figure out a fix... some improvements, but not nearly enough.





Now, here's the vintage RS52AN from way back, with the new crossover:



Nearly no sign at all of the 1800Hz to 2kHz surround resonance issue, as expected, given the clean impedance plot testing this driver.

Next step is to check with tweeter, make sure I have absolute phase on the cabinet wiring with new crossover correct (inverted midrange) and see if there are any obvious issues with the mid to tweeter blend:




The phase blend seems to be on the money, but the mid level appears to be around 3dB high; need to adjust the L pad values and check again...


as I like to say, slow work grinding the sausage takes time!

[JonMarsh]

Well, slow work takes time, especially when adjusting resistors requires removing a passive radiator from the back of the cabinet, pulling the crossover board off the side panel it's velcroed to, and then updating resistors. Was fairly close yesterday, but not quite there- went back to sims, decided to add one more resistor in parallel on the shunt leg of the midrange (it's a nuisance pulling parts out) and, whether it was skill or luck, the measurement for the mid and tweeter combined was right where I want to see it...


And as soon as I reconnected the bass to mid/high input binding post strap and fired some music up, I felt like it clicked- started off with two Holly Cole albums, and now Jacque Lousier Play Bach Air on a G String. Been working on the new LF crossover and LCR boards test setup and connectors, and enjoying the music in the next room. I'll give this another day or two of listening, then post the crossover details, in the unlikely case anyone is interested and has some vintage RS52AN, Peerless DA25TX008, and Accuton AS168's! I can't really say if this is an "upgrade" over the Morel EM1308's, but since mine got smashed up in moving, they are for me, and I didn't have to EQ out a few of the oddities of the 1308's response.

[JonMarsh]

Well, here is a view regarding an update on the enclosure documentation, not just for "closure", but because there is a chance that a set like these may be built by Steve for an HTG member.

This is part of an overall methodology which will include a more complete review of the crossover (some possible tweaks already identified, could be considered optional, but... not for me!)- this will inhabit my living room for a while once the crossover updates are completed. Updated STEP files are ready to share with Steve. And the midranges in this implementation are the set of NIB Vintage RS52AN as described above.

Here, I will share a few views and a photo- the Shapr3D views are both Birds Eye and mouse eye to show how the perspective behavior in CAD drawing can be misleading, in a way, at certain angles! This was mentioned even for the early versions prepared with ViaCAD. It shows up with the upper facet bevels looking longer than the lower ones, when the reverse is actually true.

Note, these images are shown with the woods used in my original build; Maple pin block for the front panel and top panel, strip bamboo with an inner wall of BB ply for the sides and back.

[technodanvan]

Still a classic design, I really like this aesthetic. I might be biased.

[JonMarsh]

We all have our preferences, of course. I do like how this turned out, and how it has a nice visual flow, a very simple finishing process, and no veneering!

of course, there is a price to that, in the choice of materials- such as the strip style bamboo board from Woodworkers source, 1/2" BB ply for internal panel build ups (Woodcraft), and the use of 1.5" maple ply pin block from Piano Showcase for the front panel and top and some internal parts. You don't get any of that stuff at the big box stores, but it's fairly easy to get online!

[JonMarsh]

Been a long busy day since early, many things accomplished, including installing another Vintage Mid in the 2nd enclosure, measuring it by itself, and taking a look at the full crossover design update based on this new data, plus tweaking the woofer and tweeter a bit.

But until I get the tweaks and updated midrange crossover implemented (which also means waiting for parts), I'm just going to share a raw measurement of the RS52AN vintage part in the 2nd cabinet. I didn't get around to doing that measurement in the first cabinet, because I knew the driver was basically OK based in the impedance check- and all the vintage parts have gone through that screening.

So, just a tease... I imagine the first "gold" prototype samples of the RS52AN looked a lot like this... but the 2018 and 2019 parts I got sure didn't.

[JonMarsh]

Here's what the day's work looks like in VituixCAD with the updates... with just a little luck, when I have final measurements after the parts arrive and physical updates are possible, this should be confirmed. This evaluation used the early AS168 test data which was in a sealed enclosure set. The LF alignment has also been updated, moving the box Fb from 27-28 Hz up to 31-32 Hz, increasing the PR coupling and output.





This basic SPL plot shows the voicing intent:





And impedance- an amplifier friendly load.

[technodanvan]

Looks really promising Jon!

It's been a while, could you provide some insight on why the AS168 was chosen given the final tweeter and mid used?

[JonMarsh]

Well, it would be much easier for you to just read the earlier parts of this thread than for me to write it all over again… some relevant post numbers are 5, 31, 33, 37, 39, 80, 227.

But the key point is that unlike the LR3 crossover I came up with (that is not an alignment sanctioned or described by Linkwitz; he was surprised when I explained my reasoning to him for a 3rd order mathematical equivalent at a CA DIY meet), the Z axis acoustical origins of all drivers for a Duelund crossover must be in the same plane. Additionally, having measured the AS190 at length for the Kurosawa system, I really preferred how the characteristics necessary to get the acoustic origin in the mounting plane could also create a nominal baffle step compensation built in, so that large inductors can be avoided in the crossover. The AS168-9-470 and the AS190 both like to live on 12" wide baffles.


AS190



AS169-9-470





AS168-9-470 12" wide vertical test enclosure, in room, not outdoors. 1/48th octave smoothing (I.E., none for all practical purposes.)


One more thing- a polar set 0-60 degrees for the AS168-9-470:





All of the chosen drivers have exceptional polar response throughout the operating range.






As to the DA25TX, I still maintain as ET did then, that it's the best $400 tweeter money can buy, especially at its normal price. Thats based on many comparisons with competing parts from well known name brands, with flatness, distortion, and uniformity off axis being key considerations. And by distortion I mean comparable HD2 and HD3, unlike the parts from some well known builders.

[technodanvan]

Thanks Jon, I'll go back and reread those posts as well.

[JonMarsh]

Just to be sure I’m clear, those woofer measurements are just the raw driver, no crossover filter.

[JonMarsh]

It also occurred to me that a very brief refresher of what a Duelund crossover is based on might be handy. I have Steen's original papers, we should put together an article based on that, but for now I'm just going to post some key pages from a previous presentation showing how the calculations work. This was all done in one of the earlier versions of MathCAD, though I expect it should work fine in current versions with the new owners of the company.

Once the calculations are setup, there are two controlling parameters- a gain coefficient aleph, and the crossover center frequency. Note, if the aleph is set below the square root of 2, it devolves to a two way system.



Here is an example plot for several different values of aleph, showing how the midrange gain and shape changes, and how the spread and slopes of low pass and high pass are altered. Note, the high and low pass ultimately converge on a 4th order slope. I included some level marker lines, because they are useful in understanding how low in level the driver and crossover behavior should be controlled- I find -18 dB to be a minimum effective target. YMMV.





A Duelund crossover has the midrange connected out of phase, and only a single phase roll over, like a two way system. Both the impedance and the impulse response can be clues to whether a system is using this concept; for example, this is characteristic of the Vimberg Mino.




Here is a plot for the phase of Low pass, midrange, and high pass of a single defined crossover filter...




And this shows the critical crossover implementation region for the first system I built using this concept- the NeoD CC. My listening tests as well as measurements showed that hitting the target transfer region within this target "box" is critical for achieving usable results. This is a broader range of crossover control than many speaker systems achieve. Note, at the transfer function center frequency, all three drivers must be in phase and contributing to the overall output level to sum properly.

[JonMarsh]

Well, at this point, getting crossover parts from Solen in Canada seems like swimming upstream against some national trends, but the parts I wanted to tweak up the DA25TX crossover have arrived yesterday. Due to internal AudioWorx Design Rules and procedures, this means building a complete new version, not modifying the original, so that the original is always available in its native state for comparisons.

Yes, I have a lot of crossovers sitting around here documenting past design stages, but I find this comes in very useful for exact checks and comparisons, and sometimes for doubling back in a sense, as is happening for the TPL75 crossover for the Saint-Säens. I'm too used to a certain kind of engineering environment at work, also with my self imposed rules.

Note, this is not an intellectual exercise, but continuing the work for a value customized version of the original build for JimS, who's working with Steve on the cabinet configuration after I sent them some updated 3D CAD files setup for the vintage RS52AN I have on hand, along with DA25TX00-8.

Here are the parts that arrived yesterday, Solen was chosen because PE was out of stock on too many parts. Hmmm....

Solen caps and inductors, with a different brand of non-inductive resisters using a film composite.



And the layout for the tweeter test boards I'll build, replacing the boards currently in my own Ardent D's...

[JonMarsh]

Well, one of the two boards is ready to test, so that's up next before completing the other one. Not expecting any surprises, but these days, who knows?

Keep in mind, this is what I would call a "value build" - no Jantzen Alumen caps, no high end Mundorf resistors, but some fairly solid choices like the Jantzen Superior Z 1uf caps to get exactly 2.0 uF, and the Solen Silver 10uF PP cap. Some choices were guided by what was in stock... especially for inductor values, where I've had problems with my usual vendor PE with availability of some values. L9 20uH is a custom part unwound from a higher value Solen part and verified with one of my BK Precision LCR meters.



Of course, this could have been laid out on a smaller board, but it fits the form factor of the original board and if changes need to be made, there is plenty of room.

[JimS]

Nice progress! I’ve been traveling and not paying too much attention, but now that Jon has outed me I guess I should get around to starting a build thread 😀

[theSven]

A build thread would be great to share the progress of this.

[JonMarsh]

Sven, this IS the original development and build thread, started back in 2019. This latter addition is just an update switching over the midranges to the Vintage 2008 RS52AN, which were in storage, and I was originally trying to use 2019 purchased parts which were found to be somewhat defective, due to surround issues, and thats why it switched to the much more expensive Morel midrange.

These are the original cabinets, built in 2019/2020.

This has been on hold due to focusing on the LF POC 3 update for the Saint-Saëns. Only one of me...

[JonMarsh]

Now, the DA25TX00-8 tweeter used in this project and some others had been announced in early April to be back in stock at Parts Express by May 3-4. Checked today, and not surprising to me, it's not the case after all...

I expect we'll be seeing messages like this for a lot of Peerless and other brands of parts.





And, No, PE, those "Similar Items" aren't remotely similar, other than they are also tweeters of a sort.

[technodanvan]

It was also originally indicated they'd be available in May...now I'm showing August-October.

[JonMarsh]

I have a bad feeling about this, that it may become a familiar story. Glad I have a small stock on hand.

[technodanvan]

Fortunately I also don't need any of them at the moment...I have a large stock of unused drivers at this point and have been working to organize them into expected designs and kits. I easily have several years of projects to do at my current rate.

That said, I did really want to nab some more Peerless drivers while they were available. So much for ordering early.

[JonMarsh]

Status on the update of my original build - I'll recap the tweeter also, though I posted about that not too far above.

Steve has been working on an updated cabinet build for JimS, and here I will show the crossover work for adapting to the Vintage RS52AN and updating the tweeter and woofer crossovers- for the woofer, this is occurring due to availability of more ideal values and tightening up the phase and amplitude conformance.


Doing a second build on the tweeter crossovers with some tweaks proved to be more of a PITA than I expected- mainly due to component availability issues at PE. So a lot of the parts came from Solen in Canada.

A few comments may be in order:

This drawing (created in EazyDraw) shows the relatively relaxed layout, based on using the previously available board space.





Note, for example, the two 1.0 uf caps comprising C7. Do NOT substitute a 2.2 uF cap in place of these for optimum results- it's literally about 10% too much capacitance for the design. Note, the tweeter ultrasonic filter has been converted to a parallel configuration series notch. Easier to get precision than adding a small value resistance in series with L9.

Here is a photo from the top of one of the actual builds- input connector on the left. Resistors are very low inductance parts available from Solen. PE didn't even have the right standard values in stock when I was building these months ago.





I like these Molex connectors because they have no magnetic alloys- just a brass like copper alloy.


Something to be noted- in building these originally, I figured I might be swapping boards in and out now and then for testing or updates, so screws or clamps are not used for mounting the crossover. Instead, the enclosure wall on the right and left side of the upper interior are covered with 3" wide Velcro loop material, and the back of the crossover board with 3" wide Velcro hook material.





Here you can see the HF and MF crossover boards "mounted" at the top of the enclosure on the sides.

[JonMarsh]

The midrange crossover is a fairly straightforward bandpass filter... but that doesn't mean it's small!


This is the original working layout drawing in EazyDraw-





A few tweaks were required in construction and after some measurements- to make the inductor layout more workable, and to adjust levels slightly.





With the MF and HF board tested in concert, the results are very close to expectations- final testing awaits updates to the LF crossover.




This is with 2.83VRMS drive sweep. (monitored on digital storage scope). Half hamming impulse filter with 40 msec window.

[JonMarsh]

The woofer crossover is a snug fit on the lower part of the front panel, also affixed through the use of 3" wide Velcro strips.




AWG12 cable was used to connect to the input from the binding post plate, and Cardas 101 speaker cable from the LF crossover board to each AS168-9-470 driver.

The woofer crossover is getting some small updates since a component availability issue has been resolved- the original build use Jantzen Permite core inductors, but PE carries a very limited stock, and even HiFi Collective didn't have the value I preferred (680 uH), and I ended up using 600uH with some other adjustments.

But, earlier this year, the 680 uH version because available at PE, and I purchased a pair and started looking at adjusting the network design for using those.



This edited early floor plan layout for the LF crossover shows the updates:




Besides updating L11 (gluing the new part in to the crossover board with JB-Weld) C17 is updated from 47uf to 68 uF; (for space reasons I elected to replace the original, as finding room to add a 20uf cap was problematic); R12 is increased to 10 ohms, and R15 slightly increased from 5.1 ohms to 6.2 ohms.

This process is underway; the JB-Weld takes some time to cure up!

[JonMarsh]

Small WIP update this AM, for woofer crossover, should have hardware test results later today.

[technodanvan]

Morning Jon! What wire are you using for connecting crossover components?

[JonMarsh]

Do you mean on the boards, or between the boards? On the boards, AWG 14 copper or the intrinsic component leads. Between boards and drivers, mostly Kimber 12TC, 8TC to tweeter.
The layout above was done in Altium Designer, this was before I started using EazyDraw so much for board layouts.

[JonMarsh]

Well, the LF board components are updated, (L11 and related components) so after lunch I'll be re-installing the LF crossover board, installing stuffing, and passive radiators, and hopefully some time today checking things out.

[JonMarsh]

well, Sunday-Fundays is not how things have been working around here, this was more like a Monday-Gruntday, but at least there is a good sense of satisfaction bouncing around between all the needed tasks.

And!

this included getting the modified LF board re-installed, getting the enclosure "Acousta-Stuffed" (instead of partial fiberglass stuffing), and re-mounting the passive radiators.

But!

This took until well after 4 pm what with multi-tasking some communication activities, and Sunday Union Rules say when started at 6AM, no working past 4 PM, so I only squeaked out a quick impedance measurement check today-





Things are looking as expected, Fb tuning a tiny bit low, but just one weight disk, so tomorrow we'll get some real measurements and see if things are where I hope and expect...

Just for comparison, here's the impedance curve before installing the PRs, open back.

[JonMarsh]

Big change in plans the rest of the week, one of the in-laws got a surprise diagnosis Monday of stage 4 cancer, and we're driving to Seattle tomorrow so my wife can see her older daughter. Probably be gone most of the week.

[theSven]

Condolences on the sad news Jon. Safe travels and I hope your wife can get some quality time with the loved ones.

[JimS]

Wow Jon, so sorry to hear. Best wishes for your family.