My_Ref Fremen Edition - Build thread and tutorial

Excellent values :)



You mean the single transformer you have is a 300VA one?

If so it's plenty to drive both modules.

There's a slight possibility of having some hum but it will work fine.

Obviously the second one will improve things further.

I have a small affordable enclosure which I got from China for my chipamp.com LM3886 amp, it is a nice little enclosure and already has a 300va transformer, but it's too small to fit two FE boards inside. I went through all the options and in the end I bought another small enclosure the same as the first one along with a spare back panel. When I get a second 300va transformer, I can then use both enclosures and transformers to build monoblock FE's :)

 
Remember this....



Well, it turned into this by the time it got to Denver.









The cause was flying transformers. The straps and the wooden wedges broke and the tranis were free to fly as they might.

Jim T. sent back the mess along with this advice: "The best way to check your packaging is to hold it above your head and drop it on the floor." :D I took up the challenge.

I'm still going to do the original super compact design in all metal , but I decided to make wide monoblocks just to do something different. To hold the transformer in place I used a piece of 5/16" NC-18 threaded rod, a chunk of round stock and three 3/4" long 8-32 wide head screws from the bottom. The threads are continuous through the puck and into the floor. All the panels are 1/8" Alu, the spanners are solid and the rest of the screws are 6-32 steel (aluminum screws cost a fortune). Let's see that come apart.:grumpy:









The traditional arrangement for a chassis this size and shape is front to back. I wanted to keep the signal leads as short as possible and this setup did the trick. The heatsink is a piece of leftovers from Heatsinks USA. I did't do an air flow cut-out on the top yet primarily for aesthetics. With all the aluminum pieces bolted together I'm hoping I won't have to.













The LED leads are pieces of tiny copper tube and heatshrink tubing. Signal leads are hard wired, but the placement allows for easy removal if necessary.



In the end, only the LM3886s had to be replaced - the cap that appeared split was only forced out of the pads. A stud that is seated through the entire web of the heatsink is used to fasten the LM3886 with a 6/32 nut. That also eliminates the struggle to fit a screwdriver between the AG caps and over the top of R3 and it's HS.

Lots more work to do in an attempt to get a good brushed aluminum finish (tips and techniques welcomed), but that will wait till after the Post Office gets another chance to do it's "Handle With Care" trick!! :cuss: I haven't done the 6L6 overhead test yet, but chances of survival appear significantly better. :rolleyes:

Both amps sound great and special props go to Dario for developing a super strong board. Not a single crack or pulled pad on either board. :worship:
 
Last edited:
Lots more work to do in an attempt to get a good brushed aluminum finish (tips and techniques welcomed)...

Bob, excellent metal work there - I couldn't dream of doing it myself, but I do have access to machine shops that do milling, drilling, etc. for a fee.

I use 80 grit waterproof silicon carbide Emory paper to manually wet-sand each flat surface to remove scratches, surface imperfections, etc., followed by 400 grit, both in the preferred direction of the brushed finish. After this, there are several (alternate) options:

1) Wash it with water, let it dry, use a paper towel to lightly buff it to remove any surface residue from the drying water droplets, turn the flat surface upwards and spray paint it quickly with acrylic clear coat from an aerosol can. I use a fairly cheap (~USD 3) can which contains about 400 ml (~ 0.1 gallon).

2) If you want a smooth grey finish, wet sand it once again with 1000 or 1200 grit Emory, wash and and spray paint as 1) above.

3) If you want a shiny finish, wet sand it once again with 1000/1200 grit, than use a small amount of white bathing soap (Dove, etc.) to coat the surface, and wet sand it gently until the shine appears and droplets of soapy water form beads on the surface without wetting it. Keep it up until a uniform shiny surface appears. Wash it with water, dry it with a blower and a paper towel (careful about finger prints). Spray paint it with acrylic aerosol paint as above.

4) For a mirror finish, you'll need an even finer grit emory like 3200, and a process similar to step 3) above.

After the acrylic dries (15 - 60 mins), it can be lightly wet-sanded with 1000 grit (to remove any surface dust, fibres, etc.) and additional coats applied as necessary.

After the final coat dries, it can be polished (maybe after a day) with any wax or car polishing/finishing compound. I had good results with a single-component Zymol formulation long ago, but that's not available locally here, so I'm limited to traditional waxes if needed (it's not usually needed, but it depends on your clear coat. I use a brand called Hacsol which dries to a gloss finish, but Duplicolor or similar should be just as fine.)
 
Here are pics that illustrate some of the finishes:

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


The side panels are the finish obtained from process 1 above.
The left and right sections of the front panel are from process 2.
The rear panel is from process 3 (not exactly mirror finish, but close).
 
That looks very nice also Siva, Reminds me of the front panels on the Parasound Halo line with the mixed surfaces. I printed out your processes and will try them later this winter. I started on this technique but have to get better media. The grit on the sandpaper I used isn't strong enough.

Did you have to place vent holes in the top, or does the bulk of the metal produce enough heat spread to eliminate the need?

Marra, That was the only piece I had to farm out. I have a metal bender but it's not strong enough to handle this stuff. The curve is a result of the limitations of the bend process (sharp corners available only with preformed stock) but once I saw it, I liked it as a design element. I'm planning a traditional layout using the same piece, but with a clam shell two piece design. My guy - Ackerman Brothers - can make them for ~ $11 each. If there is any interest from other FE builders, that price can be lowered by bulk - and I can produce kits of either/both styles.

This might be a good time to start an official MyRef FE logo contest. It needs to reference both Mr. P and Dario in the design. I'll throw out a couple ideas in a few days.
 

Attachments

  • 20131120_080227.jpg
    20131120_080227.jpg
    283.2 KB · Views: 259
  • 20131120_080421.jpg
    20131120_080421.jpg
    134 KB · Views: 254
  • 20131120_080747.jpg
    20131120_080747.jpg
    211.7 KB · Views: 244
... I started on this technique but have to get better media. The grit on the sandpaper I used isn't strong enough.

Did you have to place vent holes in the top, or does the bulk of the metal produce enough heat spread to eliminate the need?

That technique is similar, though he has a nice jig to help get straight brush lines (I just do it by hand). He also starts with 80 grit, but stops with coarser paper like 240 - it's largely a matter of how fine you'd like your brush lines. He uses dry sandpapers, I use wet silicon carbide emory paper (which is more efficient at removing material and also gives smoother, flatter, more even abrasion). Aluminium forms a very hard oxide layer (basically sapphire) on the surface, so you'll need a hard material like silicon carbide to cut through it efficiently. The water layer is to prevent it from forming another layer of sapphire immediately. The emory papers I use are grey ones from Carborundum Universal - the adhesive backing is durable (especially on the 80 grit) and holds the grit to the paper for a long time. I've also seen recommendations for the 3M purple-coloured wet emory papers.

I have vent slots on the top panel for convection cooling, though for low-power chipamps and Class-D amps it would be sufficient to have conduction cooling through the case alone.
 
That's not cheating. I enjoy the craft of metal and wood working and have collected a nice selection of tools/equipment. Others may not. With all the great options available on the internet/ebay a different approach is only a matter of preference.

Also, remember there is no longer any adult supervision (read wife) here anymore, so I can get away with things others can't. :D
 
Last edited:
I did have access to a decent commercial engineering workshop up until a couple of years ago but was too busy working on peoples boats in it. When I finished work I just wanted to get home so I didn't get much diy done.

Now I look after the missis full time and don't even have a garden shed at home to work in. The lad's old bedroom comes in handy for drilling stuff etc, lots of hand tools in there now.

Definitely something missing in my life though, i'll be complete again when I get a shed built here :D