Speaker repair-need help

Joe: When i was talking to Bill, i was going to mention your name, but i ended up getting sidetracked with the conversation that we were having. Needless to say, i was glad to have found someone that shared my love of this speaker and the specific design & performance attributes that it brings with it. This is especially true given Bill's level of exposure and experience with hundreds upon hundreds of various drivers and loudspeaker designs.

As you mentioned, Bill is very high on these speakers as they are from the factory, even though they could have been done better. He also mentioned to me that they are unrivaled once fully modified in the manner that he discussed. As i've mentioned in some other threads, these speakers, even with the less than optimum assembly from the factory, do some things so well that i will never part with mine. While they do have their shortcomings, Bill seems to think that his modifications fully address those problems. After discussing things with him, i can see how / why these problems / design shortcomings come about and how easy they would be to correct. I now realize how much better they may be after performing such mods to them.

Having said that, i've come up with a few design innovations that take his tweeks a bit further. The nice thing is that they involve the same amount of work, so they wouldn't be any tougher to impliment. Given that i'll probably end up rebuilding these drivers myself, i'm going to incorporate these ideas into the rebuild and see how it goes. My thoughts are that Bill's suggestions should improve performance quite drastically with my revisions simply adding a hair more icing to the cake.

Quite honestly, i've very seriously thought about producing a speaker similar to these, but made in a much more modern design. I've gone over a million different electro-mechanical design revisions in my head and keep coming back to one specific concept. This is all still in my head though, so maybe i'll be able to try out some of these ideas on my F drivers if / when i get around to rebuilding them.

At the same time that all of this is going on, my Brother is currently trying to build a prototype of a driver that he designed. I've tried to get him to work with me on the design of the driver that i have in my head, but he's got his mind on his own projects. While he is interested in the design that i came up with and would like to tinker with it, he wants to build his design first. His design is both easier and more conventional in operation than what i want to do. If he can get it to work as he thinks it will, that technology could be applied to my design too. Having said that, there's not a driver made that operates like what he has in mind, so even it is not exactly a "normal" dynamic driver. This driver should offer a phenomenal amount of control, transient capabilities and excursion capacity without generating near as much reflected EMF or impedance variations. Coupling those qualities with a Walsh based full-range omni-directional driver could prove to be a great thing. Needless to say, this is all a pipe dream at this stage of the game : ) Sean
>

Sean...Why the tease? You have told us that the Walsh driver has all these faults, and you have ways to fix them, but we have not the slightest idea what you are talking about. Quite frustrating to your readers.

El: As i mentioned, i've actually contemplated building and marketing such a design from the ground up. I've been thinking about this for quite some time, but it is probably "just a thought". Either way, i've bitten my tongue in this area on purpose, at least publicly, for this reason. I have worked with a few people privately in terms of performing simple tweaks to their F's, which can noticeably improve the performance of this speaker system.

There is another factor involved here though. There were several different production changes to this driver while it was alive. As such, they vary in performance and what needs to be addressed. Without really looking at the drivers, cabinets and damping materials themselves, it is pretty hard to give vague and generic advice.

Much of this is why Bill at Millersound doesn't want to take on these types of projects anymore. He told me that due to production tolerances and differences, it may take him up to 3 complete assemblies & dis-assemblies to get the driver where he wants it to be. Given that this is a VERY difficult driver to work with in the first place, we are talking about a massive amount of time & labour. This is one of the reasons why he considers the drivers to be "broken from the factory".

For those that are interested, I will say that these speakers benefit massively from bypassing the internal factory wiring. The effects of this are NOT subtle to say the least. Altering the type / density / placement of damping material within the cabinet can also make a big difference, improving the transient response and damping characteristics of the driver. How much material and what to use will vary depending on the individual characteristics of the driver being used.

The Walsh's will also benefit from damping the frame rails and motor structure as they are modulated by the motion and the output of the cone. This is not to mention the diffraction that takes place due to having the frame rails in close proximity to the source of radiation. I've devised a much better way to get around this, but i can't impliment it without building a driver from scratch.

As far as modifications go, the really tough stuff comes down to altering the mechanical design of the driver itself. I can see dozens of ways to improve this, which should be obvious to those familiar with speaker design. Then again, most of these involved gutting the driver and almost starting over from scratch. Starting completely from scratch would actually be best, but at the same time and as you know, the likelihood of success is much higher when starting off with a proven commodity.

The fact that Bill offered very specific suggestions as to what he's found to work best in terms of upgrading / addressing some of those problems really helped me out in that area. The fact that i discussed this with several professional rebuilders / reconers and it left them dazed told me that most non-professionals wouldn't be interested in such info. It gets very involved as the parts used require modification for use, so you can't even resort to "off the shelf" substitution.

As such, i refrained from posting such info. Someone attempting to perform such work without basic levels of experience will most assuredly destroy an irreplaceable driver that i dearly love. This is much the same reason that i had previously refrained from posting specific info about AC systems, as i didn't want to encourage potential hazards with even more serious potential results. The only reason that i did so was at the private request of several different Agon members who are / were dealing with that situation at the time. Sean
>

Sean...Although I am a long time admirer of the FR Walsh driver I am not familiar with details of how it is assembled. However, as an engineer, I offer the following suggestions.

1. Big picture...The existing design appears to be difficult to build, nearly impossible to repair, and requiring a lot of skilled manual tweaking. It seems to me that this design was taken from the lab directly to production, without the essential step of industrial design/producability engineering. Redesign the thing so as to retain the basic concept of the traveling wave down the tapered cone, but with simple more automated fabrication as a primary goal. (What is the patent status of the Walsh driver?)

2. I always wondered about the basket legs getting in the way. Why not use a massive "C" structure instead of a basket?

3. Don't worry about efficiency. Powerful amps are cheap today, and (with digital technology) getting cheaper.

Send me a copy for evaluation :-)

Why not use a massive "C" structure instead of a basket

Shape it as a sphere (no sharp angles) while you're at it. Get rid of some of thems standing waves & feedback.
* Also, pls extend usable FR 120-22kHz, if possible, w/out a wizzer.
* Pls provide a +24T strength magnet (for wont of better matl, alnico is aceptable).
Cheers