DIY Walsh driver revisited

glorocks

Member
2007-07-05 5:53 pm
Not much interest, eh? On another forum, someone mentioned the Walsh tweeter that Infinity used for a time.
I worked with Bill Seneca on the original prototypes for this thing- We probably built 50-60 pairs of these guys on my kitchen table.
If there is interest, I'll share how we did it.
J-
 

glorocks

Member
2007-07-05 5:53 pm
Cambe-
Can you attach a couple of close-up photos of your tweeter?
I can tell you what we did with the prototypes, but I don't know what Arnie Newdell did in production.
The base for the tweeter was a Seas 1 1/2 inch dome. They were used in the Dynaco two way bookshelf, and back then, they cost about $9. I doubt of you can get them any more, but I did see a couple of likely candidates at Part Express.
We had to carefully cut the dome off of the voice coil, which was tricky because the magnet attracted the Exacto knife blade. We ruined quite a few.
The cone was made of a layer of 1 mil acetate that was cut using a pattern. There was no particular science involved- we just arbitrarily selected a size, about 6" tall, and the angle. The acetate was then smeared with clear silicone except where the end met for cementing. We used 5 minute epoxy.
Now over the coated acetate was placed a layer of heavy duty aluminum foil that conformed to the shape of the thing, and the laminate was rolled into a cone shape and epoxied. I think we used invisible Scotch tape to hold the seam while the epoxy set.
A wooden dowel was epoxied to the center of the pole piece in the tweeter that was long enough to just reach the end of the cone when it was assembled.
This was used to hold a thickness of urethane foam cut ever so slightly larger the the open end of the cone. This served to terminate the acoustic wave traveling up the cone, and helped give the cone a little support.
A hole was cut in the center of the foam to accomodate the dowel.
The cone's small end was trimmed to fit the voice coil former, and was then epoxied.
Surprisingly, we didn't use a jig to hold the cone- we just set it in there and eyeballed the angle, then waited for the epoxy to set.
The, a small amount of Dacron was placed inside the cone around the dowel, and the foam plug was ever so gently eased into place.
That is basically it. We didn't have the luxury of instrumentation- the whole thing developed by trial and error and extended listening. We were told by a true "golden eared" musician that it had a slight peak at about 2KHz. I think Seneca smeared some white heatsink grease around the surround and that helped fix it.
White heatsink grease got smeared on a lot of things back then to see what would happen.
The Promethean Audio cartridge resulted from injecting the stuff inside the cantilever of a $35 Grado cartridge. It dampened micro resonances that were occurring in there.
I'll look around for a few possible candidates for making these things and get back here in a day or so.
Cheers,
Jay
 

glorocks

Member
2007-07-05 5:53 pm
I did a bit of poking around at Parts Express, and found a dome tweeter that might be a good starting point. Part # DC28F-8. It has three things going for it- it's relatively cheap at $22.00, it can be dismantled, so you can cut the dome out without having the Exacto blade pulled out of your hand, and you can get replacement diaphrams. You'll ruin a few before you get it right.
Now, the holy grail- when we were playing with these titanium foil was "unobtainium".
We wanted very much to try it but couldn't get it.
Now, we have Titanium Goat where you can get it in small quantities without draining your wallet.
I mention this because I am thinking about building a Walsh driver that can extend down to 200 Hz or lower, and hoping that some of you will be encouraged to try this and share your expriences here. German Physiks Walsh (they disavow the term Walsh, but that's exactly what they are building) speakers using titanium have received rave reviews, but at $45K a pair and up, I couldn't afford to audition them much less own them.
So, gang, get out your Exacto knifes and Epoxy and have some fun.
Cheers,
Jay
 
Here is few shots. I will take more if needed.
 

Attachments

  • IMG_0174.jpg
    IMG_0174.jpg
    658.9 KB · Views: 4,414
  • IMG_0176.jpg
    IMG_0176.jpg
    681 KB · Views: 4,185
  • IMG_0178.jpg
    IMG_0178.jpg
    976.1 KB · Views: 4,020

glorocks

Member
2007-07-05 5:53 pm
Cambe-
Please send a few more photos- I would like to see the whole banana as well as the top. Especially the top- we need to see if they did anything inside that may make it diffcult for you to get them apart.
I don't recognize either the "cone" or the driver. The ones we built did not have those diagonal striations, nor did the Seas done tweeters we used have, but as I said, I don't know what Infinity did when they went on to production.
At this point, assuming there is nothing inside you can't see holding them together, I'd say you have a very good shot at remaking these. So far, I see that you can take them apart using an offset screwdriver with a very steady hand. Remember that magnet is going to try to ruin your efort.
Once you get the top plate off, you can use a caliper to dertermine the diameter of the voicecoil. Now you know the VC diameter of the dome tweeter you want to look for for a repair. Check Parts Express, and look for cheap tweeters you can take apart.
If you can find a dome driver that's close, you should be able to seperate your cones from the voicecoils. I would suggest you use a Dremel with a thin cut off wheel to avoid deforming your cones. They smell bad, but in a steady hand, they can be effective without a lot of collateral damage.
If you want to take it to the next level, let me know. If you wish, you can PM me, and we'll talk on the phone.
Cheers,
Jay
 

glorocks

Member
2007-07-05 5:53 pm
Now for those of you that are lurking hoping to see progress. I am currently in th throes of moving, so it will be a while before I can atually get something started, so please stay tuned.
I'm going to make my first prototype using an inexpesive Dayton 3" full range driver. I am not presently at my own 'puter, so I don't have the part number at hand, but I will add it ata the beginning of next week. I will also include a link that describes one person's odyssy.
Those of you that wish to play may find thin aluminum sheets a place to begin. I am going to try titanium, and am investigating carbon fiber laminate material. I would also like to use photopolymer adhesives instead of Epoxy. You may want to investigate these materials on your own.
Years ago, I built a kludge that was made with a scrap offset printing plate glued to the motor of a KEF B-110 mid driver. The thing lasted about 10 minutes before it fell apart, but for 5 or so minutes, I only heard music- NO speaker. It is this memory that got me going again.
J-
 

glorocks

Member
2007-07-05 5:53 pm
Cambe-
I want you to look very closely at the area at the base of the cone. Is it white, sort of like putty? You may want to try to get a tiny bit of it on the end of a toothpick.
I may have misread your photos- that could be white heatsink grease smeared over the surround of the tweeter-
J
 

glorocks

Member
2007-07-05 5:53 pm
Cambe-
Well, they sure don't make damping materials like they used to. My original assesment is obviously wrong- you need to start over. You can certainly do this if you don't mind mind experimenting.
I'll be happy to help, but right now, my shop is in storage, so I can't do a build along and trade photos with you. The most difficult part is getting the dome cut off of the tweeter and fitting your new cone to it. Be prepared to make several, because you're going thave to figure out how to damp out resonaces in the "cone", (that's what the foam inside was doing) and terminate the edge so the acoustic wave does not reflect back down the surface. Trial and error and a lot of listening.
The damping and termination was and is where the "voodoo" lies. You can use just about any thin material that is stiff enough to transmit sound and hold it's shape.
The original Walsh theory stated there had to be a relationship between the velocity of sound in the cone material versus that of sound in air. The idea was when the angle was correct, the acoustic wave leaving the device was parallel to the vertical axis.
That is why the original Ohm "A" has two angles- the cone was made of titanium at the top, and aluminum at the bottom. The angle of the two materials reflected their respective propegation velocities versus that of the air.
In pracitce, however, not so much. witnesseth the Ohm "F", whose cone was made of 3 different materials, (Ti, Al, and stiff paper) all at the same angle. I feel fairly certain that the choice of the angle had more to do with stiffness and radiating area than velocity.
In the case of the tweeter, as I said, the angle selected was pretty much arbitrary.
So. give a month or two to settle in, and I'll start playing again and share my results.
Cheers,
Jay
 
Last edited:

glorocks

Member
2007-07-05 5:53 pm
As promised, here is a link to one fellow's experiments with a full range:

http://www.audiodesignguide.com/full/conus1.html

You may have to paste this address into your browser-
The driver the author has suggested is no longer available. Some of the others mentioned are very expensive- I would not risk destroying one of these as an experiment. There are a number of relatively inexpensive full range drivers available, and they would be a good starting point.
Now, on to the tweeter. I located one very promising candidate on Ebay:
# 160746320172
It has two problems- there is a wire(?) "grid" cover over the dome that must be removed, and the faceplate is square. Both can be remedied with a Dremel.
The plusses are it is cheap, and there seems to be a large number available. I e-mailed the store owner, and he indicated he could supply a thousand units or more, and was willing to apply quantity discounts. Group buy? At $7-8 dollars each, one can afford a mistake or two.
Keep in mind that right now we don't care about the audio quality of the tweeter, since we are just looking for a motor platform that can excite an acoustic wave in the "cone". We are not concerned with the mass of the cone either, we just want it to be able to hold it's shape and vibrate along the surface.
In my next post, I'm going to share my thoughts concerning cone material and possilbe dampening strategies.
Cheers,
Jay
 
Last edited: