The ultimate DIY? A Stax SRM-T2!

[nattonrice]

Call it

[deepak]

Birgir are you betraying us on Team Silver Gear (says the guy with all black components )

[mypasswordis]

Black dissipates heat better... (and looks better)

[luvdunhill]

Black dissipates heat better... (and looks better)

nah, silver dissipates heat better (i.e. raw aluminum)

[mypasswordis]

Link? [s-cars] FW: Anodizing effects on Aluminum thermal heat dissipation Q

[justin]

?

black is a better heatsink

if it's textured, even better

[mypasswordis]

Yeah true, increasing surface area also helps.

[luvdunhill]

Um yeah? Yeah, I wasn't thinking... Powder coating better than anodizing, really? I didn't know that one...

[mypasswordis]

I'm pretty sure it isn't. You have to gain the texture from somewhere else..

[justin]

bead blasted and then black anodized is probably the best

[Dusty Chalk]

I think having heat sinks on your heat sinks is best, a la the cyborg former-life-as-a-dog things in Snowcrash.

[justin]

yo dawg we heard you like heatsinks so we put heatsinks in your heatsinks so you can sink while you heat

[Craig Sawyers]

nah, silver dissipates heat better (i.e. raw aluminum)

I though - nah, that can't be right - the difference must be significant. I remembered Leslie's Cube from schooldays and foggily recalled a much greater effect. But then I found and scanned "An experimental inquiry into the nature and propagation of heat", By Sir John Leslie, 1804 (Readable on Google books) and noticed that the basic experiment only measures radiant effects - not convective ones. So the experiment is entirely relevant for space based thermal radiators (which would have astonished Leslie), but not ones in air.

But then I found the comment relating to normal heatsinks:

"Under natural convection conditions, the performance of a heatsink with a black surface will be 6% to 8% better than that with a plain or bright surface. However, this differential disappears under forced air conditions." Significantly less than I had intuitively expected - but consistent with the primary cooling effect being convection.

The above comment was on Basic Theory , which seems to be a distillation of heatsink info from several sources.

[kevin gilmore]

birgir's knob is done

pictures later today when i get home.

[luvdunhill]

Specifically the heater windings to the EL34's (-500V), and two of the HT transformer windings are referenced to -560V or -500V

Good point Craig. I realized the two 6.3/4A windings and mentioned these, but didn't realize any of the HT transformer windings qualified. Which would those be again? I was told that adjacent windings without extra insulation can hold off 500VAC, so I suppose I need to know which of the others that this isn't sufficient need the extra insulation.

[Craig Sawyers]

Good point Craig. I realized the two 6.3/4A windings and mentioned these, but didn't realize any of the HT transformer windings qualified. Which would those be again? I was told that adjacent windings without extra insulation can hold off 500VAC, so I suppose I need to know which of the others that this isn't sufficient need the extra insulation.

It is the winding connected to the output marked "60V power supply" - that adds -60V to -500 to make -560, and the one marked "300V power supply, that adds 300V to the -560V supply to make -260V.

And yes - winding wire is supposed to hold off 500V - but I wouldn't like to rely on it. It might be OK at T=0, but after repeated thermal cycles I would not have confidence that it would hold up long term. It is one of the reasons that transformers with an interwinding screen are often unreliable - the devil is in the detail.

[luvdunhill]

ok, I see the 75VAC winding goes through the TL783 then is reference to the -500VDC winding to make -560VDC. Check. Then that same -560VDC is added to one of the 285 VAC windings to make -260 VDC (so 300VDC) supply and the other 285 VAC winding is used to make to make a 250 VDC supply? Ok..

[Beefy]

ok, I see the 75VAC winding goes through the TL783 then is reference to the -500VDC winding to make -560VDC. Check. Then that same -560VDC is added to one of the 285 VAC windings to make -260 VDC (so 300VDC) supply and the other 285 VAC winding is used to make to make a 250 VDC supply? Ok..

*Beefy's head explodes*

[Craig Sawyers]

Finished the power umbilicals today (getting everything prepped for the arrival of casework in due course). Forgot how fiddly it was to wire up Amphenol mil spec connectors.

I guess it took an hour or two to cut the wires precisely to length, twist the heater wires (to reduce hum radiation), pull though mesh sleeve and adhesive lined heatshrink the ends. Then cutting up neoprene sleeving to the correct length (around half an inch). All told maybe 4-6 hours for the pair.

After soldering one end I did a pin to pin check to eliminate wiring errors (one found and corrected), and a leakage check to ensure that there were no faults in the ubmilical or the connector. Used an AVO mega-ohm meter, which has a switchable voltage up to 1kV and measures up to 10Tera-ohms (10^13 ohms). Well the leakage of my test setup was 10^11 ohms, so all tests were relative to that (to get much further the whole set up has to be on a teflon sheet). Every pin was checked to every pin, and it was all well over 10^11 ohms at 1kV for every wire to every wire - which is an excellent result. Very happy with that.

I made them up at 4 feet at KG's suggestion - that allows the units *not* to be stacked, so helping heat build up.

[kevin gilmore]

birgir's knob

http://gilmore.chem.northwestern.edu/birgirknob.jpg

It is not polished at all, strictly micromachined.

Even the fingerprints show up... Amazing stuff.

4.5 hours of actual machining time. 100 turns per .0001 inch.

2 passes front and side.

[Nebby]

Simply beautiful.

[Dusty Chalk]

That's a huge picture of a knob.

[kevin gilmore]

That's a huge picture of a knob.

What better way to show the surface finish. Macro lens

Paging Dr. Wood... Paging Dr. Wood...

[kevin gilmore]

ok, I see the 75VAC winding goes through the TL783 then is reference to the -500VDC winding to make -560VDC. Check. Then that same -560VDC is added to one of the 285 VAC windings to make -260 VDC (so 300VDC) supply and the other 285 VAC winding is used to make to make a 250 VDC supply? Ok..

Yep, bizarre, but that is the way it works.

[spritzer]

birgir's knob

http://gilmore.chem.northwestern.edu/birgirknob.jpg

It is not polished at all, strictly micromachined.

Even the fingerprints show up... Amazing stuff.

4.5 hours of actual machining time. 100 turns per .0001 inch.

2 passes front and side.

Most excellent!!! Since the chassis is all black one has to have a bit of silver to make it look better.