I've looked through the photo gallery and have been quite impressed with many of the gorgeous enclosures the members of this forum have constructed.
I've noticed, however, that very few of you have taken venting into consideration. All of those power resistors, high voltage caps, chokes - and in some cases - the valves themselves, are tucked underneath those beautiful boxes, leaving no space for their copious thermal discharge to escape. Sure, some of them have open bottoms, but I also know that you're a clever lot that is most likely aware of the aversion that heated gas has for downward travel.
I've been reading Morgan Jones' second book - Building Valve Amplifiers, and he puts quite a bit of emphasis on venting whenever possible.
Is the idea of keeping parts relatively cool simply in the interest of lengthening their life span, or would cooling also aid in keeping noise down and tolerances tight? Or, lastly, is extensive thermal engineering simply an unnecessary step that is only a consideration of the obsessive?
Fans make noise, so they're pretty much ruled out. Holes in the chassis may not be that ugly if placed strategially.
Adam
I've noticed, however, that very few of you have taken venting into consideration. All of those power resistors, high voltage caps, chokes - and in some cases - the valves themselves, are tucked underneath those beautiful boxes, leaving no space for their copious thermal discharge to escape. Sure, some of them have open bottoms, but I also know that you're a clever lot that is most likely aware of the aversion that heated gas has for downward travel.
I've been reading Morgan Jones' second book - Building Valve Amplifiers, and he puts quite a bit of emphasis on venting whenever possible.
Is the idea of keeping parts relatively cool simply in the interest of lengthening their life span, or would cooling also aid in keeping noise down and tolerances tight? Or, lastly, is extensive thermal engineering simply an unnecessary step that is only a consideration of the obsessive?
Fans make noise, so they're pretty much ruled out. Holes in the chassis may not be that ugly if placed strategially.
Adam
A good thread topic. I have routinely used plate load resistors that get hot, very hot, for example. Air needs to flow freely from below to above if no fan is used.
A very valid point worth repeating now and again. There are diminishing cooling returns for increasing a fan's output flow (CFM). There can be a significant cooling effect from a slow (and quiet) fan.
A very valid point worth repeating now and again. There are diminishing cooling returns for increasing a fan's output flow (CFM). There can be a significant cooling effect from a slow (and quiet) fan.
Using wood as my main material, my latest case has 3 vents on the top, which match 3 holes in the bottom, the heatsink inside is from the front to the back of the case and runs right underneath the line of holes... I have some round plastic vent grills that will go over the holes.
The wood has had 2 coats of Glatex 8, and after many hours of sanding is near glass smooth with no wood texture left...almost ready for the topcoat.
Want to try somethig similar for my Akido.
The wood has had 2 coats of Glatex 8, and after many hours of sanding is near glass smooth with no wood texture left...almost ready for the topcoat.
Want to try somethig similar for my Akido.
Attachments
OK - so it IS an issue that deserves consideration.
As a teenager, I had a part-time job building hot-rod PC's. With high-cycle CPU's and GPU's, cooling was always an issue and often demanded clever airflow designs to keep components stable. I'd even run a cloud of dense CO2 gas (dry ice + water) through the intake with a piece of plexiglass in place of the side cover to determine spots where the air wasn't flowing. Of course, this was probably over the top, but also fun and a great sales gimmick.
I still have oodles of CPU fans in my stash, so I'm sure I use some of them.
Has anyone noticed a SONIC difference when cooling is implemented?
As a teenager, I had a part-time job building hot-rod PC's. With high-cycle CPU's and GPU's, cooling was always an issue and often demanded clever airflow designs to keep components stable. I'd even run a cloud of dense CO2 gas (dry ice + water) through the intake with a piece of plexiglass in place of the side cover to determine spots where the air wasn't flowing. Of course, this was probably over the top, but also fun and a great sales gimmick.
I still have oodles of CPU fans in my stash, so I'm sure I use some of them.
Has anyone noticed a SONIC difference when cooling is implemented?
As a teenager I was building ------------ vacuum tube amplifiers.
Now 40 years later I am building ------- vacuum tube amplifiers.
We sold the Craig 4 function calculator in the stereo shop that I worked in, in 1972. It sold for $250 in 1972 dollars. Since the store was next to the University of Miami (expensive private school) we actually sold a few. At this time (I was 19) I was fixing vacuum tube (and solid state) amplifiers.
In between I built computers, starting in 1975. Due to lack of time, I just bought my first pre built desktop PC. The darn thing didn't work and now the pieces are scattered all over my workbench, and the tube amp is in the closet. I have now wasted a whole week finding a bad memory simm that passed memory tests for 6 straight hours, but caused random crashes and BSOD.
Tube amps or computers, all electronics are sensitive to heat. As the temperature rises the MTBF decreases. In a tube amp the electrolytic caps, and resistors, and the tubes themselves are the most affected components. The hotter they are the shorter they will live. I have not noticed any sonic differences, but resistors (and semiconductors) will be noisier as the temperature rises.
When I was a teenager we were using slide rules.
I remember in Electronics class, the smart kid (definately not me) bought an electronic calculator & everyone was googling over it.
I usually let the amp run for 10 minutes or so & then turn on a small table fan.
Very annoying on stuff like solo Classical guitar, but on most other music only rarely notice the sound of the fan.
I remember in Electronics class, the smart kid (definately not me) bought an electronic calculator & everyone was googling over it.
I usually let the amp run for 10 minutes or so & then turn on a small table fan.
Very annoying on stuff like solo Classical guitar, but on most other music only rarely notice the sound of the fan.
My ideas: use only metal chassis, with a mesh base, punch large holes for tubes and recess them into the chassis. The heat will suck air through the chassis and past the tubes.
Also use thermally over-rated components (e.g. 10W resistors instead of 5W), 105 degree caps instead of 85 etc. site capacitors away from power resistors, and attach all power resistors to the chassis with thermal glue to act as a heatsink.
Also use thermally over-rated components (e.g. 10W resistors instead of 5W), 105 degree caps instead of 85 etc. site capacitors away from power resistors, and attach all power resistors to the chassis with thermal glue to act as a heatsink.
I know the feeling -- I gave my wife a brand new Ford Bronco in 1989 -- she never changed the oil -- dang thing needed an engine rebuild 7 years later with 150,000+ miles on it. Being as it was a Ford, the transmission went only a short time thereafter.anatech said:
Heat shortens the "mean time between failures" but not as predictably as one would expect. According to Tony Kordyban (inventor of the Therminator) the thermal milspec for semi components was established by putting a wet index finger in the air.
Hi jackinnj,
With regard to audio equipment. When I was working as a tech and later running a shop for many years, death due to overheating was fairly common. I regard that as just as stupid as the way my wife retires cars. It is so very preventable in the design stage and certainly the user can help here.
-Chris
Let's not talk about that too much. As a young lad I took great care of my cars. I looked after every service truck I drove. My wife refuses to do maintenance even though she agrees to. The end result is that evey car we've ever owned has died due to lack of maintanence. This is possibly the greatest stress on our marriage. She finally murdered my 93 Buick Roadmaster wagon (my favorite car so far).
With regard to audio equipment. When I was working as a tech and later running a shop for many years, death due to overheating was fairly common. I regard that as just as stupid as the way my wife retires cars. It is so very preventable in the design stage and certainly the user can help here.
-Chris
anatech said:I still have my slide rule around here somewhere ...... I don't think I remember how to use it anymore!
-Chris
I still have my book of 4 and 5-place logarithms -- the one my dad used when he went to Case Institute of Technology before the big one. Still have the ivory slide rule he got in "Occupied Japan". A slide rule gave you an intuitive feel for the way the numbers were going to work out, and you eventually commit to memory the logarithms and trig values.
The discussion of thermal management is very important. I get calls all the time (ever wonder why my fax is hooked up to the office phone?) about an LM3875 or LM4780 design which doesn't work because they used a 20 C/W TO-220 heat sink on the thing.
With respect to tubes -- many decades ago i used to keep my morning donut warmed up by placing it on the top of a Heathkit "barefoot linear" -- this while working 20 meters DX. i am thinking of moving my Audio Research D-115 and SP-8 onto the back den and forego the baseboard heating for the winter semester.
I sit among a crew of IC designers at work. They are quite concerned with thermal management. Thermal stress in IC's causes metal migration, especially as we move to smaller and smaller geometries. Thermal gradients across the die make it difficult to cancel out DC offsets in high gain linear systems.
All we have to worry about is keeping our parts from melting. For most parts the wet finger test is adequate (watch out for that plate cap). Electrolytic capacitors are prone to drying out with heat and even the good ones are only rated for 1000 or 2000 hours of life at the max temp spec. ( 85 or 105 degrees C)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.