DIY Walsh driver revisited

I might take a crack at this soon. I'm wondering about drivers to sacrifice for a wideband tweeter. The obvious choice would seem to be a compression driver.

In terms of wattage per dollar and voice coil size though, a cheapo woofer might compete. There seem to be two sacrifices with going that way:
- Much lower BL
- Higher impedance in the upper registers

I'm mostly wondering about the BL. Which is more important for a driver like this: total power output, or BL?
 
So far the Pyle PDS541 - Pyle PDS541 1" Titanium Horn Driver 1-3/8"-18 TPI - is looking like the best option if the metric is magnet and voice coil per dollar. 60oz of magnet and 2in of aluminum voice coil for $32 from Parts Express.

A couple other questions have come to mind about compression drivers:

At what frequency would a carbon fiber cone start to become pistonic?
I'm assuming that any compression driver will have very little if any xmax. According to the patent, that's one of the advantages of the Walsh design - less vertical movement for a given volume of air moved, if I understand correctly. I don't think the patent covered the math to determine when the cone would become pistonic though, unless it was buried in one of the bits about "successive finite integration in reverse". Any rough ideas or empirical estimates?

Is a compression driver voice coil sufficiently rigid to bond to a CF cone?
I'm guessing that this depends a lot on material. Given Pyle's reputation as a budget supplier, they probably also are not using any more or higher quality VC material than they can get away with, I would guess. Will a 2" aluminum compression driver VC just bend out of the way when trying to drive a CF cone?
 
Yet another option might be lathing down the edge of the CF cone so it'll fit in the magnet gap, and winding a custom VC on it. There are a couple of barriers I can see to this:
- I have been looking for the better part of a day and have not found ANY stats on wire gauge and number of turns on a compression driver VC.
- It would be a lot of extra work, which I'm trying to avoid in order to keep the project fun and interesting at least for the start.

Maybe I'll put that idea off until later if bonding to the aluminum VC doesn't work.
 
Altie I´ve built a couple of Walsh type speakers and one of things I've learnt is that you should use a driver that is intended for full range. Otherwise you risk ending up with speaker that only plays the lower range in pistonic mode. The impedans curve is good indicator. The challenge is to find one with a voice coil close to 50 mm or 2 inch.
This master thesis have theory on bending wave .
http://publications.lib.chalmers.se/records/fulltext/154618.pdf
 
  • Like
Reactions: 1 user
Sandasnickaren, glad to see you back. I actually just picked up a pair of Peavey 22XT wideband compression drivers that have - you guessed it - 2 inch voice coils!

I'm planning on using the resulting Walsh drivers (or trialling them at least) as wideband tweeters similar to the 500-20KHz labeled response range (https://www.google.com/url?sa=t&rct...80302103.pdf&usg=AOvVaw1-Olod8Gu-i-CVIIoFBozo). I'm sure the addition of a big old cone to them will change the frequency response, but at least the driver itself doesn't seem like it'll have a problem impedance-wise with higher frequencies.

My reason for concern about pistonic behavior is that I would like to keep away from that operating regime. I had forgotten that I already looked into this with an old DML experiment: Carbon fiber + nomex honeycomb sandwich DML panel construction project. I still have those panels around, so maybe my first step will be to try using these compression drivers as exciters for those panels. They have to have substantially stronger magnets than the Daytons I was using before. IIRC from that experiment I ended up with a fundamental bending frequency of something like 100-200Hz. Does that sound right in your experience?

I'll have to give that thesis a look again...
 
*blink* OMG....I've started a novella....*L*

Altie, MHO on BL....I opted for a higher BL over just sheer wattage ratings. Generally, the 2 seem to go hand in hand, although that's not an absolute...

The 'logic' behind that being one is asking the 'motor' of the speaker to perform in a far different fashion than it was designed to perform. Instead of moving a relatively compliant diaphragm in a pistonic motion, we're wanting to compress the top of a truncated and rather stiff cone to create radial waveforms that descend and expand the widening skin of the cone.

As with a conventional driver, damping the waveforms at the greater circumference of the driven cone is highly desired....make that nasty IM stop 'n stay...

Easier to do (somewhat) with the typical cone surround, which for a Walsh is not....pistonic motion for the smaller cones I've opted to use exists, but is being damped to 'encourage' the radial excitation of the cone...

As for 2" dia. vc's, that was the size of the original Ohm Fs' and As', which were full-range drivers driving large base diameter cones of multiple materials. This arrangement, in my (perhaps flawed) opinion and approach, made use of the 'phase shift' from radial to pistonic due to the descent of LF waveforms into a 'region' of the cone where the vertical motion of the vc 'overruled' the radial radiation.

The crossover was physical in that sense....so, based on study of the heights, diameters, materials, and the basic profiles of the cones, I used an 'average' as a jumping off point to fabricate the current v3s', of which y'all have seen previously.

Thoughts on voice coils...
My 1st iteration, a pair that suffered 'vc heat death'; the entire coil of both fused....I was surprised that they had lived as long as they had, given the studied abuse applied...

The intent was to investigate the reported 'fails' of the F's and As'...they either cooked the vc's (db over-enthusiasm, too much too long....that is what fried my "v1's") or collapse of a cone segment (*L* 'transient telescoping': lf @ high db...my guess as to that malfunctions' cause). It's been reported that the early Ohms' 'sounded Great @ 200 watts....and cooked @ 201'.

My response to the former:
Apply a driver with a vc meant to withstand abuse; a Pyle PPA6 'professional woofer' with a 1" Kapton former, vented, 400 w. max., music/PA applications.

Since I don't run anything that can apply that level of wattage, the cone revision is totally different than its' original, and opinions/reviews tended positive marks (and the price was about what I wanted to spend on 4)

My recent cones up till now have been 5 mil aluminum...truth be told yet again, lithographic sheets gifted from a friend long ago back in Houston...
Again, the price was right...;)

I've acquired 2 mil aluminum since....and recently, 2 mil titanium...

Carbon fiber cones are intriguing, attractive on a number of levels...and at this point, would require acquisition of the means to create them Properly.
And the learning curve to make acceptable, accurate, and reproducible ones.

I've always made a minimum of 2 drivers. A single one exhibits the fact that one can walk around it, and the sound and response remain constant.

Duh...it's an omni, and should do such....and does, correctly.

Do a pair and one begins to experience why a Walsh driver, and the original Ohms, were and still are, *ah* 'subject to superlatives'....

Now...make 4....sit in the middle of them...apply program of choice.

Get back to me on what you think of it...;)

Oh, vc to cone attachment....I've left a narrow 'skirt' of the original cone on the circumference of the vc, creating a trough for the new cone to attach to, and a means of keeping your adhesive of choice from finding its' way down into the gap of the vc within the magnet. You will likely get upset if such occurs...
I know I was less than thrilled by that...the driver involved wasn't of any great value, but the time involved to get to that point was....

I don't see any huge difference in having the top terminus of the cone inserting into the vc, other than losing some degree of driven weight and mass of the entire assembly....
....which, IMHO, due to the physics of how and why the cone is being driven in the fashion it is, makes the mass & weight 'perhaps' not as much of an issue as a conventional driver, where that becomes more of a factor...

Just mind the glues' characteristics....you can and should be a Control Freak when it comes to 'sticky sh*t'....*L*
 
....and, backing into the attractive qualities of carbon fiber applied to cones....

High Strength 3D Printing – Markforged

Not that I've got the disposable $ to drop a P.O. tomorrow, But....a CF cone with top & bottom edges suitable for my applications & parameters, some interior resonance reduction structure(s).....

Anyone got a winning Lotto ticket that they're planning to 'party away' anyway? *G*

Yeah, thought so....;)...and can't say I blame anyone for that response....
 
....back from the Shadows again....

Hi, y'all...*S*

Despite, and contrary to (perhaps) popular demand...I persist in my form and format....Unretired retired....which comes off as a form of retread, which we generally experience as tire tread chunks on your fav interstate....

So far '18 is being.....Different....*punch esrever....*clunk*

(Tape on the Revox spins up....Sunny, the Pit/Plott, wanders in....bumps take-up 10" reel.....50ish ft. spins out, spider-webbing pet who freaks and flies....)

*survey (ob)scene*Sigh*shrug*

Well, just another miserable day in Paradise.....

YouTube

...which leads me to Anne Dudley's epic comment...

YouTube

(...of which, her piano solo at the end...which, per one 'audiophile critical source', is considered 'then', and IMHO *cough* still the best 'mic'd and recorded piano' in existence'....the brain and beauty behind the AON...)

One fan responded with this....

YouTube

Great for 'tune-ups'....;)

Back sooner....
 

Attachments

  • 20180520_175800.jpg
    20180520_175800.jpg
    474.9 KB · Views: 407
.....oh, no...I Still don't believe it....

YouTube
or
YouTube

You can: Break your lease ~ Bust a move ~ Push an envelope ~
or just

Push It.

....which may inform as to my idiosyncratic taste in music of late....
Have been, and will always be, a FON to (the) AON...and the talent of Ms. Dudley...

Have been either so busy or just merely dazed at the first half of 2k18 that the 2nd whip'round wouldn't surprise one byte if gravity takes a Fail and Earth goes AWOL. Who dealt this mess, anyway?

...Oh, well..."Let's Play PONG."

(For 50k credit...Next time you do the PhysReHab group exercise @ Your CardioKlasse'....;)....bring the 'tooth' speakers with and hit Play.

I kept threatening to take over their AV rack...carried the appropo acoutremon in muh bag'o shiticks....

Meanwhile...found another 8024 eq, and added a MX882, as I feel the need for more oxygen-free copper in my life....not necessarily on my body, but near..,
 
Small update on my Walsh project

Hej!
I’m back.
It’s been a while since my previous contribution. I’ve done some development of my walsh speaker. There are two versions in the picture, the large ugly one that includes two twelve inch woofers and the latest which is the slim one.
Couple of comments on the slim version.
The driver is the same design as in the earlier version (you will find more details in my earlier post).
The big change is in the construction of the whole speaker. The box is a tapered horn more or less as per B&W nautilus design. The advantage is less or no box colouration of the sound.
The slim design also minimizes diffraction, which used to be horrible in the earlier versions. There is still some diffraction from the magnet.
The whole speaker i assembled on a 20mm stainless steel rod from the feet to the top through the magnet (you can’t see it it’s inside except for the nut thats on the top).
The rubber surround is mounted inside the cone.
I’ve included one rough measurement to give you some ide on the level of performance.

It sounds pretty good!


Thomas
 

Attachments

  • D117FADE-1108-41F0-B462-35F0EC337979.jpg
    D117FADE-1108-41F0-B462-35F0EC337979.jpg
    294 KB · Views: 502
  • D1E92AC0-9B05-4460-9353-5C62052A2C16.jpg
    D1E92AC0-9B05-4460-9353-5C62052A2C16.jpg
    148.2 KB · Views: 503
  • 717873D6-0622-48ED-B148-CD5A4817C764.jpeg
    717873D6-0622-48ED-B148-CD5A4817C764.jpeg
    76.3 KB · Views: 488
Finally....but certainly not a finale....*G* ;)

Gary, Thomas....and all y'all, Hi.

I can appreciate max effort yielding min results, fur chure... "We" (in my business life) have been hog-tied with 3 major projects, two of which suffered major setbacks due to weather (rain, unrelated to Florence which gave us a pass locally; we were in TX, D/FW area anyway...), employee 'issues' (had 4, 2 quit, one had to leave due to losing a housemate) and the supplier of our slides and climbers having a major fire that decimated their entire 70K sq. ft. factory. Fortunately, our orders had been manufactured; the resulting chaos made finding them difficult. And this happened After we were informed that the merger 'they' had planned late '17 and put into motion...at the start of the 'busy season' (early spring)....hadn't taken into account That...their computer systems and networks....and their whole means of control of EVERYTHING....

...didn't mesh....*oops* Our April 1 orders....came in after the July 6 fire....

....mid August for one locally....the other mid Sept....mostly...

Add to that my older brother in SoCal passed away from a heart attack on April 13....Yup, a Friday. Just what everyone had expected from yours truly, but he failed to pay attention and act sooner to forstall that.

Not long after, I took the opportunity to have the pacemaker/defib installed that my herd of cardios wanted me to have onboard. Recent follow-ups of a CT scan of lungs shows no cancer and my new aortic cow valve (now 1 yr. old) and my drug therapies are working Beautifully....

So I'll likely outlive my of my contemporaries....:boggled:....unless y'all pay attention when 'stuff don't work right'....*This is a Clue; Take it*

Meanwhile, next pending project in Lexington, KY is pending if the rain lets up so the GC can deal with the sheet of granite under the entire site. If 'they' deal with it, it'll Perhaps forstall all the 'change orders' from nearly All the subs....and the LLC 'we' are already 6 figures in our contract....

Thomas, I'd noticed 'enclosure issues' early on; I called it 'honk', a 'megaphone quality' if you play the Walsh without one underneath. I opted to either have no enclosure other than a short base on one smaller pair, and infill the interior of the cones with a styrofoam cone that 'infills' that space. The foam cones fill +/- 90% of the interiors, effectively damping the bass (to my ears) at what one could say is their crossover point. A small Sony sub takes over at that point....

These units are about the same size as the German Physiks units they're styled after. Remarkably, they sound very similar to my larger pair, which are mounted on 4" dia. x 33" h. PVC sch40 pipes stuffed tightly with polyfill garnered from pillows that our dogs had opted to 'kill'. No porting; a shorter version done earlier performed better sealed....

The tapered version looks elegant. *S* With our pets, I fear they wouldn't stay upright for long, although they aren't near them often. But it only takes once... Actually had one toppled awhile back; it survived the concrete floor unscathed. My nerves, mostly...esp. after I uprighted it and surprised self that it still worked and showed no damage.

I make stuff Tough. I build playground structures. I've learned that 'overkill' isn't a bad concept...;)

Base diffraction: Since I'm reluctant to dive into cutting the baskets off (since I'm still using them for structure/support) and my next variant has vented voice coils (touch the back of the magnets after high spl for a given time...heat has been an issue for me and cooks the voice coils), I'll pass on that. It does yield an elegant and 'clean' appearance, but I'll pass on elegance for now. *S* I'm opting for a 2 way version, with a smaller cone directly above the larger one.

With the digital Xover, I can 'see' where to (perhaps) 'duck' major diffraction.
The MX882 will allow for precise level adjustments and give me better balance control, and the second 8024 will allow for eq in and rta out with the calibrated mic. Theoretically, with a second mic, I ought to use two channels of the MX to feed into the rta 8024 and see any differences....

I like the looks of your plot! I see correlations in it that my rta 'tells' me. *S*

This following year is going to focus on 'enhanced sophistication' of my pile of equipment and the 'puter and programs it runs. I'm not set up to run plots as yet, and would like to run some 3D waveforms as well.

One Great Stagger Forward: I bought a used wood lathe that will allow me to (finally) turn the cone forms that will (well, ought to...:xfingers:) make creating 2 mil alum & ti cones easier.... Push became Shove when I was quoted 400$ to turn 2 cone forms earlier...but the upcoming KY project reqs. 2 cedar forms I was quoted 500~600 for the pair.

The lathe, including most accessories + an extended base + the bench and steel cabinets beneath = 321$ at a local shop. Picked up some turn tools and misc. + the wood; I'm into it for 500ish.

I've used a lathe before, the forms for either are simple, and (big plus!) the spouse (who took wood shop & is handy with things wood) liked using a lathe..Yay!

Now...all I have to do is build a house on our property and do biz and build speakers.... Eat? Sleep? *shrug* ;)
 
...and another....minor movement on fabrication, but the opportunity to assemble an 'audio only' 'puter whose only functions will be reproduction and measurement.

We (the 'business 'I' *L*) are now booked pretty much into the Fall. That does 'dent' 'free-time'...which is Everything But....), but gotta make Sun while the hay shines.

Oh, my recent 'new tool' acquisition..and Yes, it Does throw chips at you. So....'primal'. *L*
 

Attachments

  • PraziBosch.jpg
    PraziBosch.jpg
    501.9 KB · Views: 323
Hi Jerry/Thomas,

Sorry to hear about your brother Jerry.
Glad you are doing well though.

Thomas your speakers look awesome. I bet they sound good.
On my experiments I eliminated the struts so I could try to hear a difference, I could.

Last year I was very busy doing other things but, now I have free time to study and experiment. I also want to build a matched pair of Walsh drivers for my listening enjoyment.

Peace,
Gary (cochleus)
 
Hi y'all...*S*
Pic #2s' base is interesting, as it looks suspiciously like a solid table leg. *G* ('tip' hardware, compression driver held by the shaft from the top surface....) If you carry that 'design ethic' a bit further, it'd make for a very elegant appearance. *S* I'd add a sleeve to mask the nut above the mic stand base, a simple 'fix'. ;)

(Pardon my personal 'quirk' about appearance of things Walsh. If omnis' in general make a serious return, it'd be (IMHO) a big plus if they'd be elegant yet unobtrusive. Just saying...)

The 'audio only' 'puter build is underway in fits and bits.... Had an older case that's sized more like a large audio power amp that may end up rack mounted as most of my 'serious stuff' is now. Will post a pic when it's nearing completion, but wanted to give an Asus Essence STX2 with the breakout board a suitable home. *S*

Spouse Ev is down with Whatever I do audio-wise in our home plan, but it Must include the SMGa Maggies' in some fashion...

(...and if you think I said "No.."
...you're crazier than I am. *LOL*)

Yeah, I agree with David, Sanda. Even if it's just your impressions of what you've wrought with your work. A YouTube vid perhaps? (I hope to replace the rather lame one I did awhile back when I've something worth posting and the means to improve upon it...).
 
Hi and thanks guys for your kind feedback.
I’ll put in some effort and produce some documentation (maybe including a video) on my project so far.
Jerrry, I will have a look on a compression driver and see if it’s a way forward.
Thomas
Ps. Jerry I do have a sleve to cover the nut. I just forgott to install it when I assembled the speaker��