SteveH,
For cheap heatsinks your best plan is to find a local surplus store. I have two electrical hobbies, building stereo gear and Tesla coils. The later mandates shoping in "junk stores" and you would be amazed what you can find.
I have a collection of heatsinks in my shed which I have accumulated buying "other" surplus material and the cost is "dirt cheap".
All the heatsinks for my Son-Of-Zen (internal and top) were surplus & < $20 each. Likewise the sinks for my Aleph were $30 each. And these are BIG sinks !!
Hunt around. Get your hands dirty and you might be surprised what you can find.
Regards
Mark
PS: Grey ..... please use automotive coolant, not water, aluminium and H20 don't last long ;-)
For cheap heatsinks your best plan is to find a local surplus store. I have two electrical hobbies, building stereo gear and Tesla coils. The later mandates shoping in "junk stores" and you would be amazed what you can find.
I have a collection of heatsinks in my shed which I have accumulated buying "other" surplus material and the cost is "dirt cheap".
All the heatsinks for my Son-Of-Zen (internal and top) were surplus & < $20 each. Likewise the sinks for my Aleph were $30 each. And these are BIG sinks !!
Hunt around. Get your hands dirty and you might be surprised what you can find.
Regards
Mark
PS: Grey ..... please use automotive coolant, not water, aluminium and H20 don't last long ;-)
Mark,
I'd considered the antifreeze idea, but decided against it (for the time being) for two reasons:
1) Aluminum and water aren't *that* reactive--consider as an example how many aluminum pots and pans get left in the sink to soak overnight...or even for days. Car engines are usually an alloy--aluminum and magnesium, I believe, possibly with other things (someone help me out, here, I haven't kept up with current auto technology in quite some time). Add to that the fact that car engines operate at *much* higher temperatures than even class A amps and chemical reaction rates double at, what, every 18 degrees or so? At the 100 degree plus/minus 20 level, things shouldn't react all that quickly, if at all. Yes, I expect a bit of aluminum hydroxide, but not all that much.
2) Antifreeze is toxic, and I'll be using my brewing pump for the time being. Don't want no nasties in my beer!
And if I get to the point where the water in my cooling system is that hot...I did something wrong. Time to turn the whole shebang off and reconsider.
Grey
I'd considered the antifreeze idea, but decided against it (for the time being) for two reasons:
1) Aluminum and water aren't *that* reactive--consider as an example how many aluminum pots and pans get left in the sink to soak overnight...or even for days. Car engines are usually an alloy--aluminum and magnesium, I believe, possibly with other things (someone help me out, here, I haven't kept up with current auto technology in quite some time). Add to that the fact that car engines operate at *much* higher temperatures than even class A amps and chemical reaction rates double at, what, every 18 degrees or so? At the 100 degree plus/minus 20 level, things shouldn't react all that quickly, if at all. Yes, I expect a bit of aluminum hydroxide, but not all that much.
2) Antifreeze is toxic, and I'll be using my brewing pump for the time being. Don't want no nasties in my beer!
And if I get to the point where the water in my cooling system is that hot...I did something wrong. Time to turn the whole shebang off and reconsider.
Grey
Mark,
A formal invitation, should you ever visit your antipode...what'll ya have, mate? Beer? Wine? Scotch?
I reckon it goes without saying that the Scotch I'm talking about ain't that blended bellywash!
Grey
P.S.: And the sound system here ain't too bad...fer homebrew, that is! (It ain't that bad on an absolute scale, for that matter.)
A formal invitation, should you ever visit your antipode...what'll ya have, mate? Beer? Wine? Scotch?
I reckon it goes without saying that the Scotch I'm talking about ain't that blended bellywash!
Grey
P.S.: And the sound system here ain't too bad...fer homebrew, that is! (It ain't that bad on an absolute scale, for that matter.)
Being an avid surplus buyer I would suggest the only reason to fill a hole in a sink is for "cosmetic" reasons and this may be quite reasonable.
If you are intending to fill it for "functional" reasons, eg so you can mount a device ..... forget it !!
You will get unreliable heat transfer and I could never recommend it ....... if the sink won't do "as is" from the functional viewpoint, then just keep hunting, something will always turn-up ;-)
mark
PS: Start the evening with a few beers, enjoy a nice bottle of red (or 2) then finish with a good malt. Ain't nothin' wrong with "home brew"!
If you are intending to fill it for "functional" reasons, eg so you can mount a device ..... forget it !!
You will get unreliable heat transfer and I could never recommend it ....... if the sink won't do "as is" from the functional viewpoint, then just keep hunting, something will always turn-up ;-)
mark
PS: Start the evening with a few beers, enjoy a nice bottle of red (or 2) then finish with a good malt. Ain't nothin' wrong with "home brew"!
mefinnis said:
Thanks
Reason for this filling is both cosmetic and functional. 1/5 of the area of the transistors is covering the hole. I thought tin has pretty good heat transfer propperties. What about filling with aluminum?
argo
PS. Beer makes me too sleepy but I do enjoy red wine. Never tried any Australian made wine though. Any recommendations?
Hi - interesting stuff - A few findings of my own with heatsinking...
I reckon what makes the Zen design harder to cope with is the small device count and big power per device => therefore we need a more effective heatsink than a design using more devices running lower powers (e.g. Pass A40) which can be spread about over more (smaller easily obtained) heatsinks.
I'm doing Zen Revisited - each MOSFET needs to dump 128w. Exicon 250W TO3 devices from http://www.profusion.co.uk (internal junc/case is 0.5C/W), which with temp. derate makes a max *safe* junc temp of around 120C.
I'm using TO3 plastic (not Berilium Oxide) washers from http://www.bergquist.com, SILPAD K10 or SILPAD 2000 series, both 0.2C/W confirmed by my measurements, used dry. Non-toxic and cheap. That 0.3ish advantage over other materials saves a lot of heatsink. Devices are mounted on to 12.5mm thick aluminium plate - works a treat to prevent the initial hot spot. This is bolted to a system of alternate 6*25mm Al spacer bars and more 3mm plates (the fins)- all gunged up with thermal grease (beware - not too much) and clamped together with 3 runs of 6mm studding to/thru the 12.5mm plate.
A system like this - fins 250mm high, semi circlular (diam 320mm) in plan gives me (natural convection) 0.17C/W, and with fan 0.11 C/W. All mating surfaces were wet rubbed with 400 grade wet/dry using a very flat metal sanding block to ensure as flat as possible (= good thermal joint). No machining except a £40 pillar drill, just started with 'flat' stock. The grease and clamping does the rest. It does seem to work. Fan is stripped down 30cm desk fan, run ever so slowly so it's silent - and it really is
) - at around 1/2 voltage. Seems to be a good approach generally for lo/no noise fans. Being a big sucker it still blows enough for this usage.
Oh yeah, the fins were painted with water based acrylic blackboard black, not too thick though - it'll insulate! Goes on ok if it's all wire wooled and detergent washed first.
Results: So at 22C ambient I've case temps around 56C (junc temps 56+(0.5*128) = 120, not too bad. Remember NP's comment on MOSFET service life - low temp = long life.
Get the metal stockist to do the cutting! Surprising how 'hard' Al becomes at 12.5mm with domestic methods!
So there's my solution - not living in the land of so much cheap surplus gear. It's custom built too). Guess it comes down to how much chasing around after esoteric bits we're prepared to do - better stop - feel this is getting too philosophical!
On water cooling, unless I forgot my school physics completely, I came to the conclusion that the amount of water used (if not recycling) is quite substantial over a period - exercise for the reader!!
(I asked about Al welding a sink like this - the guy said no-go as the heatsink effect was too great - ironic really).
Now where's that beer - cheers all ...
Pete
I reckon what makes the Zen design harder to cope with is the small device count and big power per device => therefore we need a more effective heatsink than a design using more devices running lower powers (e.g. Pass A40) which can be spread about over more (smaller easily obtained) heatsinks.
I'm doing Zen Revisited - each MOSFET needs to dump 128w. Exicon 250W TO3 devices from http://www.profusion.co.uk (internal junc/case is 0.5C/W), which with temp. derate makes a max *safe* junc temp of around 120C.
I'm using TO3 plastic (not Berilium Oxide) washers from http://www.bergquist.com, SILPAD K10 or SILPAD 2000 series, both 0.2C/W confirmed by my measurements, used dry. Non-toxic and cheap. That 0.3ish advantage over other materials saves a lot of heatsink. Devices are mounted on to 12.5mm thick aluminium plate - works a treat to prevent the initial hot spot. This is bolted to a system of alternate 6*25mm Al spacer bars and more 3mm plates (the fins)- all gunged up with thermal grease (beware - not too much) and clamped together with 3 runs of 6mm studding to/thru the 12.5mm plate.
A system like this - fins 250mm high, semi circlular (diam 320mm) in plan gives me (natural convection) 0.17C/W, and with fan 0.11 C/W. All mating surfaces were wet rubbed with 400 grade wet/dry using a very flat metal sanding block to ensure as flat as possible (= good thermal joint). No machining except a £40 pillar drill, just started with 'flat' stock. The grease and clamping does the rest. It does seem to work. Fan is stripped down 30cm desk fan, run ever so slowly so it's silent - and it really is
) - at around 1/2 voltage. Seems to be a good approach generally for lo/no noise fans. Being a big sucker it still blows enough for this usage.Oh yeah, the fins were painted with water based acrylic blackboard black, not too thick though - it'll insulate! Goes on ok if it's all wire wooled and detergent washed first.
Results: So at 22C ambient I've case temps around 56C (junc temps 56+(0.5*128) = 120, not too bad. Remember NP's comment on MOSFET service life - low temp = long life.
Get the metal stockist to do the cutting! Surprising how 'hard' Al becomes at 12.5mm with domestic methods!
So there's my solution - not living in the land of so much cheap surplus gear. It's custom built too
). Guess it comes down to how much chasing around after esoteric bits we're prepared to do - better stop - feel this is getting too philosophical!On water cooling, unless I forgot my school physics completely, I came to the conclusion that the amount of water used (if not recycling) is quite substantial over a period - exercise for the reader!!
(I asked about Al welding a sink like this - the guy said no-go as the heatsink effect was too great - ironic really).
Now where's that beer - cheers all ...
Pete
Pete,
I can't speak for anyone else, but the system I'm working on will be closed-loop, thusly:
Heatsink (aluminum 2x4")->barbed brass NPT fittings w/clamps to hold the->braided hi-temp vinyl tubing->old heat exchanger from heat pump air handler->(at least for the time being) my brewing pump from my RIMS system->(optional reservoir to act as a capacitor to smooth out 'ripples' in water flow if it's noisy--this I do not expect to need)->line back to the heatsink...etc. etc. etc.
I've already etched and tested a front-end board for my rendition of an Aleph 2. (My layout's different from Mark's (mefinnis) Aleph 4.) Everybody cross your fingers...I will be trying to get the output boards etched today, then do a little more testing to make sure I didn't miss anything. I hope to begin final construction on the heatsink end of things late today or tomorrow. As promised previously, I will report details here, win, lose, or draw as soon as I've got something solid.
(Mark--Nelson says R19 is the ticket...)
More to come...
Grey
I can't speak for anyone else, but the system I'm working on will be closed-loop, thusly:
Heatsink (aluminum 2x4")->barbed brass NPT fittings w/clamps to hold the->braided hi-temp vinyl tubing->old heat exchanger from heat pump air handler->(at least for the time being) my brewing pump from my RIMS system->(optional reservoir to act as a capacitor to smooth out 'ripples' in water flow if it's noisy--this I do not expect to need)->line back to the heatsink...etc. etc. etc.
I've already etched and tested a front-end board for my rendition of an Aleph 2. (My layout's different from Mark's (mefinnis) Aleph 4.) Everybody cross your fingers...I will be trying to get the output boards etched today, then do a little more testing to make sure I didn't miss anything. I hope to begin final construction on the heatsink end of things late today or tomorrow. As promised previously, I will report details here, win, lose, or draw as soon as I've got something solid.
(Mark--Nelson says R19 is the ticket...)
More to come...
Grey
Computer heatsinks
I purchased 27 PII heatsinks recently. They were quite nice, and the fan came off easily. If I would use them with fans, I would put them back to back in a long line and pass air through.
A very good thing about these particular units is that they are cerrated, black anodized etc. Should be quite effective even at low air-speed.
I purchased 27 PII heatsinks recently. They were quite nice, and the fan came off easily. If I would use them with fans, I would put them back to back in a long line and pass air through.
A very good thing about these particular units is that they are cerrated, black anodized etc. Should be quite effective even at low air-speed.
Homebrew CNC Mill
I constructed a homebrew 3-axis CNC milling machine over the course of a few months and it is now operational. It cuts to about 1/64" of precision when I set it up carefully and I have end mills for it ranging from 1/8"-1" diameter off of a 3400 RPM 1/2 HP motor, and the bed is 4ft. by 4ft.
The first thing I did with it was to build an amplifier heatsink. I actually used very simple cuts to cut 1/4" alum 12" H by 30" D panels, and then to cut 1/8" deep by 1/8" wide, 12" into them every 3/8" of an inch. Then i cut 12"H x 3" W 1/8" alum panels and used Wakefield thermal bond to create a bonded fin heat sink. Needless to say it is very large, heavy, and impressive.
I would recommend this to anyone who has the time to do it.
I also have milled 1" thick aluminimum plates and made heat sinks out of them by removing aluminimum to create fins, they are very very solid..
I constructed a homebrew 3-axis CNC milling machine over the course of a few months and it is now operational. It cuts to about 1/64" of precision when I set it up carefully and I have end mills for it ranging from 1/8"-1" diameter off of a 3400 RPM 1/2 HP motor, and the bed is 4ft. by 4ft.
The first thing I did with it was to build an amplifier heatsink. I actually used very simple cuts to cut 1/4" alum 12" H by 30" D panels, and then to cut 1/8" deep by 1/8" wide, 12" into them every 3/8" of an inch. Then i cut 12"H x 3" W 1/8" alum panels and used Wakefield thermal bond to create a bonded fin heat sink. Needless to say it is very large, heavy, and impressive.
I would recommend this to anyone who has the time to do it.
I also have milled 1" thick aluminimum plates and made heat sinks out of them by removing aluminimum to create fins, they are very very solid..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.