DIY pcb

[John Roman]

I've gone off the deep end and am considering making some boards for my crossovers. I've just started looking online. Any ideas on a site to check?
regards, John

[corndog71]

I've found hard-wiring to get the job done just as well.  Although it does make it harder to tweak.  I used a hot glue gun to help hold everything together and to secure it to the cabinet.

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi576.photobucket.com%2Falbums%2Fss207%2Fcorndog642%2Fx-ls%2520upgrades%2FIMG_2157.jpg&hash=a14f434361e427fc26000a0c91de6c2a639945a2)

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fi576.photobucket.com%2Falbums%2Fss207%2Fcorndog642%2Fx-ls%2520upgrades%2Fecafc095.jpg&hash=19d8bf6daaaaac9438cc414f842f53a116e25214)

[Grainger49]

I hadn't thought of this when I looked at this the first time.  You have more power at the speaker than anywhere in your system.  The traces on a board should be wide and unusually thick to pass the power properly.  Otherwise wire is better.

[Maxwell_E]

In my experience, homemade boards are at best temporary. I would bread-board a cross over, tweak it, and find your favorite flavor. If it were me, I would then get some terminal strips like what's in all the kits and screw them down to a chunk of wood and solder all the pieces point-to-point.

The lab I work in has a $15k mill for making circuit boards, and even then they are just made to be prototypes. I would stick to wire and PTP connections.

[John Roman]

After looking around a bit I agree with your conclusion. The experience might be fun but I'm too picky to accept a shoddy result. Thanks for the insight.
John

[Paul Joppa]

There are many places that will make prototype boards for you. But they are expensive unless you buy many identical boards - in practice you'll wind up paying $5 to $10 per square inch, and you'll get several copies - more than a stereo pair.

[rif]

Would these help? Disclaimer: I know very little about crossovers, and nothing about this product. But it sounds like it may be what you are looking for and I've had good experiences with PE.

http://www.parts-express.com/wizards/searchResults.cfm?srchExt=CAT&srchCat=197&CFID=40260858&CFTOKEN=3283877

[electrovice]

How about using turret terminals on a piece of perf board or PC board. The copper side of the PC board could be used as a ground plane.

There are a lot of vendors for turret terminals on eBay.

John

http://shop.ebay.com/i.html?_nkw=turret+terminal&_frs=1&_trksid=p3286.c0.m359

[Mudhiker]

I etched my own boards in High school.  They weren't perfect, but it's not that big of a deal, and a steady and careful hand can make a good product.  Sourcing the chemicals is the toughest part, I doubt they have them at Radio Shack any more.

[InfernoSTi]

A nice piece of wood can make a good crossover platform.  Perhaps maple would be nice?  Here are my crossovers:

(https://forum.bottlehead.com/proxy.php?request=http%3A%2F%2Fwww.hawthorneaudio.com%2Fphotos%2Falbums%2Fuserpics%2F11473%2Fphoto-273.JPG&hash=810152c0aa460866c725554621a102208c28c873)

John

[VoltSecond]

If you don

[John Roman]

Here's how the network looks thus far. I've just been arranging and re arranging components until I come up with something I'm satisfied with. At this point the physical layout has taken somewhat of a backseat to the layout in terms of maintaining better distances between inductors. If this is going to be exposed, even on the back side of the baffle it must look clean and even enhance the overall look of the speaker. Any thoughts, you know like take another prozac, get a life or ??
John

[Paul Joppa]

The inductors are too tightly coupled and will interact. It will never be perfect of course (5 coils and only three orthogonal directions in this universe!) but I'll suggest a few small changes that will help.

I'll number the inductors in your picture this way:

   1
                 2
   3

                 4      5

flip 1 so the hole points up (like 4) to make it perpendicular to 3
rotate 2 so the hole points down, towards 4. That makes it perpendicular to 1 and 2.
rotate 5 so the long axis is left/right.
Move 4 and 5 as low as you can.

This leave 1 and 4 parallel to each other but fairly well separated. 1 and 5 are also parallel but even better separated. And each component is pretty much in the same place as you have documented.

[John Roman]

Thanks Paul,
How's this look?
John

[Paul Joppa]

You got it. Looks good to me.