Oblate Spheroid Waveguide - Consolidated Construction Thread

Jump to Latest
#123
amiklos said:

Electroforming. The jello pan is very precise but also somewhat thin. The pan is sold with the jello, so really any crappy jello can be used because all we care about is the shape of the jello and the thin pan has that part covered, the jello is just there for mass.
Click to expand...

But the metal skin adds nothing but cost. There aren't grades of poly it all costs the same so good or bad its the same cost.

The skin idea is not far from the Summa, where the fiberglass is fairly thin and only serves as a skin for the back filled polyurethane which in the case of the Summa was an expanding poly and hence some cost reduction was obtained by a lower volume of material. But this was a two step process and two steps are never as low a cost as one step. That's how I was able to DRAMATICALLY reduce the cost of the waveguide by elliminating the skin step and directly molding the waveguide in one step. The molded parts DO NOT have tollerance problems, but the skin process does. The metal skin would elliminate the tollerance problem of the skin approach, but it could never come close to the cost effectiveness of the one step molding process. The current approach casts a waveguide in about 15 minutes, which is fine for the 10". The 12" will take about 30 minutes because of the slower cure material that is required for the greater volume.

Tollerances are an issue only with fiberglass, not with the polyurethane castings. The polyurethane has almsot no shrinkage at all, at least none that I have been able to detect. And precision molds are required for any process.
 
#125
mandrel in process

shots of the bottle mandrel for conceptualization

guides005.jpg


guides009.jpg


top cut off, conductive coating applied... soon to enter Cu process
 
#126
Originally posted by gedlee The molded parts DO NOT have tollerance problems, ...
Click to expand...

Of course they do because the open surface of the poured polyurethane forms the most critical part: the throat.

And there's still some kind of two step process by having to build the mounting plate for the compression driver which adds even more inaccuracy.

Originally posted by gedlee The polyurethane has almsot no shrinkage at all, at least none that I have been able to detect.
Click to expand...

Do you monitor that?

Best, Markus
 
#127
Well, the very little amount of shrinkage is not that big of a problem
The mold would only need to be maybe a quater of a millimeter bigger at the throat, but there are quite easy ways to examine that, and once you know then its not such a big deal
The biggest problem is still that when laminating on the outside of a mold there need to be a fair amount of slippage(if thats the right word?) to be able to depart the laminated waveguide from the mold

Electroforming? sounds interesting
 
#129
markus76 said:


Of course they do because the open surface of the poured polyurethane forms the most critical part: the throat.

And there's still some kind of two step process by having to build the mounting plate for the compression driver which adds even more inaccuracy.

Do you monitor that?

Best, Markus
Click to expand...


Markus

This is not correct. I don't have a means for getting the pour to EXACTLY the right level - expansion varies with temp and humidity - so I have to over pour. This then needs to be sanded down to the right spot. This is not a big problem and I do it differently now than I did on yours (which from what you say was not sanded down enough. The surface needs to be square, but thats easy enough (there is a ridge on the outside of the casting that shows when this is correct) and once it mates to the plate then the depth is correct. Where is the error?

When the poly is cast I can insert the throat template and it mates exactly, this is not the case with the fiberglass which is how I figured out that the fiberglas continues to shrink even after it is removed from the mold. The poly clearly DOES NOT do that.

The mounting plate is the part of the design that I like the most because it allows the mounting surfaces to be very solid and the waveguide walls to be very thick. When drivers are mounted on most horns this surface has to be very thin at the point of greatest stress. I consider my method of mounting to be supperior. My waveguides and method of driver attachement are far more solid than the vast majority (if not all) of what I see. This is important to me. How is this going to be done on a thin metal skin?
 
#130
amiklos said:

So maybe this could help with the larger sizes if they require the skin approach, but at the 10" size it seems like a non-starter.
Click to expand...

Its a non-starter on the 12" too since I've already been succesful casting one of those. The 15" in fiberglass isn't so expensive that its not a workable solution.

Electro-plating requires a factory of complex baths and controls. I've done copper plating myself so I know what it takes. I know about how dangerous the chemicals are and how the baths have to be kept clean and the concetrations and temperatures controlled, not to mention the electrical requirements. Ever have anything plated? The local shops here charge a lot for a very thin layer. I'm sure that NASA has this kind of equipment, but I don't (well I can copper plate a stain glass lamp, but thats about all).
 
#131
nonsense

:Pinoc:

gedlee said:


Its a non-starter on the 12" too since I've already been succesful casting one of those. The 15" in fiberglass isn't so expensive that its not a workable solution.

Electro-plating requires a factory of complex baths and controls. I've done copper plating myself so I know what it takes. I know about how dangerous the chemicals are and how the baths have to be kept clean and the concetrations and temperatures controlled, not to mention the electrical requirements. Ever have anything plated? The local shops here charge a lot for a very thin layer. I'm sure that NASA has this kind of equipment, but I don't (well I can copper plate a stain glass lamp, but thats about all).
Click to expand...

...so now you're an electroplating expert too...i don't think so

you don't know squat about the process.. that's pretty clear. Lame attempts to discredit the method by fear-mongering are sadly transparent... copper sulfate and dilute battery acid... stuff from high school chemistry... but there ARE trade secrets involved... that you don't know about

gimme a break

here's some light reading from work I did 20 years ago...

http://www.patentstorm.us/patents/5032464/description.html

http://www.patentstorm.us/patents/4786390.html

John L.
 
#132
Earl, its ok that you like your method and it works fore you

But I am not paying lots of dollars fore a lump of foam, and obviously others wont either, so we are trying to find another solution, which would be more to our liking

Besides its not such a big deal as you obviously wont sell the waveguides seperately and have even announced that you may completely let down the DIY way of this forum, which have made it all possible in the first place

Some things I dont understand

btw, I supposed this thread was about making a DIY OS waveguide, and not about buying a Gedlee
 
#133
Originally posted by gedlee I have to over pour. This then needs to be sanded down to the right spot.
Click to expand...

So it's even a 3 step process. And this does introduce inaccuracy. It was YOU talking about being as close tolerance as possible, not me.

But let's see in which direction the discussion here takes us. I think nobody here wants to change the production method for your waveguides. Maybe we end up with a cost-effective and more accurate way of producing a functional waveguide. I can see why you don't want this to happen 😉

So far we know that injection molding is too expensive, glass fiber reinforced plastic shrinks (how much?). What about CNCing from a block of wood?

Best, Markus
 
#136
Ed LaFontaine said:

Sounds like I need to use 60 or 90 minute set to avoid heat...and I wonder if a slower set reduces shrinkage?
Click to expand...


Hi Ed, you could make a small test making only the throat part and measure how much it shrinks

But what the heck, let it shrink, it doesnt matter
Shrinkage is so little that you can easily sand it away afterwards
Consider the shape of the throat so that it will have the right shape AFTER sanding or whatever method used
Much worse to have a too big throat

Using slow setting epoxy would most likely reduce shrinkage

Look at the surfboard business, they have the most advanced products
 
#137
tinitus,

I gathered your comments regarding mold preparation from the other thread. Your comments are consistent with the reading I've been doing. Thanks
I like your idea for working with the throat first. I appreciate the value of experience I can gain by doing this several times on a small scale before I attempt a full WG.

From Dr. Earl's comment about glass bubbles, they sound like something I know as "silica fume"... a strength increasing additive for concrete.
 
#139
somebody posted a great link all about casting, shrinkage, fillers, set time etc. where did that go? His supplier list was maybe a little out of date but at least one of them had zero shrinkage epoxy suitable for casting in both white and clear in one and two gallon kits. I have no experience with resin casting but that link was pretty good intro.

edit: it wss in a different thread, here
 
#140
markus76 said:


But let's see in which direction the discussion here takes us. I think nobody here wants to change the production method for your waveguides. Maybe we end up with a cost-effective and more accurate way of producing a functional waveguide. I can see why you don't want this to happen 😉

Best, Markus
Click to expand...


Right on Markus, but lets be fair and give Earl the credit of inspiration 😉
 
You must log in or register to reply here.
Top Bottom