Tube Amp Thermal Design: 2A3 Mono Blocks
ฝัง
- เผยแพร่เมื่อ 29 ก.ย. 2024
- In this video we are going to look into thermal design of tube amps. We will be studying a pair of 2A3 mono amps which IMHO had no thought given to thermal design or the problems an unvented chassis has on the life of the internal parts!
You can send me links and messaged at:
www.skunkiedes...
I have a new personal email: stephe@skunkiedesigns.com
Here is my Patreon, as well as the donation page at my site, if you feel so inclined to help fund some of the future projects for the channel!
www.patreon.co...
www.skunkiedes...
Here is where you can buy T-shirts and other Merch with my fun logo.
skunkie-design...
I leave the bottom of any amp I build,open
I have that too on my 300B amp, you still need some holes on the top for the heat to escape!
Great job Steph. I learned my lesson many years ago by building a nice looking all wood chassis. Wood does not dissipate heat well enough on the top even though I did drill out some ventilation holes. The Jupiter beeswax caps did not last very long. My current amp is a study in ventilation and no trouble of any kind at 5 years and counting.
As much the 300B can sound lush and warm.. sometimes too much..the 2A3 'can' sound exagerately clean, clinical, and boring, at times. Sounding 'dark' too..
If one can build amps.. and live with less watts.. then one must give a try to a 45 SE.
It has the lucid clarity of the 2a3, the charm and enchantment of the 300B while having a surprising decent and controlled low end.. you might be surprised.
And since a 2a3 is 2 x45 tube element in the same glass enveloppe, one 45 has a higher impedance and works well with 5k output transf loads..
It's my best amp, I admit. Old used 45 are still easy to find at low cost.
@Skunkie Judging from your thumbnail photo, I thought you were going to do an actual visual heat graph comparison between a modified vs an unmodified amp. So yeah, I got my hopes up. Still though, it was a great post in regards to what is considered logical amplifier design taking the production of heat into consideration. It is obvious that the maker of the amp was oblivious to that. Also I like the points you brought up about the locations of the mute and on/off switches and how they should have been reversed. I am definitely curious to see the measured results of these amps and also how you think they sound. Another thing I am curious about is if the circuit has some negative feedback someplace as that would explain the more sterile solid state sound. But usually on an amp like this, no feedback should be used. I had thought about a 2A3 design, but having to have an extra power supply for the tubes combined with the very low power had me put off that project indefinitely. It was shortly after I had discovered that a triode strapped EL34 would give you more power and similar sound to a 2A3. Finally, your EL34 design gives even MORE power. Great post! Very informative in regards to good layout, not only for sound, but also to keep heat away from critical components.
Ripple current on filter caps will kill capacitors too. A basic circuit simulation helps size filter caps for ripple current. Lots of filter caps are only rated for two or three thousand hours. Less options for long life, high temp, high ripple current/low ESR caps. Besides Nichicon, Kemet has some good options to kook for. Rubycon makes capacitors for Audio Note, so do not pass over Rubycon either.
I agree with all your thoughts on applying some thermal design aspects and wiring improvements to that thing.
When I use a resistor above 1 watt on a PCB, I drill holes under that resistor and set it a bit off the surface of the PCB by kinking the wires in zig-zag. On one occasion I've put a copper clamp on a 10 watt cath resistor and fixed it to a stave of the chassis to get the heat out of the inside of the chassis.
heat sink cooling fins work best vertically. also letting hot air out requires letting cool air in. your improvements make sense.
If solid state rectifiers are used, then you can reduce a lot of the heat by getting rid of the tube rectifier. If you want to have the tube type supply for the sag/compression, then that cannot happen because single ended amps draw full power at all times, so the tube rectifier is moot.
Perfection! This topic should be discussed more often.
Another option re the exposed resistor would be to insulate the leads and put probe jacks on top of the chassis above the resistor.
Yeah I know people hate the exposed external resistor lol. I think it looks cool :P
If you use rivet nuts to mount the bottom cover, you will get a small gap all the way around the bottom plate which will allow airflow. Then you just need holes in the top of the chassis to act as a chimney. I learned this by accident when installing rivet nuts in the chassis. Gap is barely noticeable, may be small gap, but the total area is fairly large.
Nice
Howdy.
Yeah. Bewilders me that the thermal design is so sloppy by the manufacturer ...
Best Regards.
Thermal design... Sooo important. Component reliability is contingent on operating temperature. Excellent topic for the Monday Monologue. Components should never touch other components. At a bare minimum, there should be clearance around a component that is equal to the body diameter, those two large resistors should be at least 2x apart from each other.
Now for the sad part, really sad. I do not know where this amp was physically assembled but far too many children are used to assemble, sometimes in garages. Seriously! The large company that I used to work for had a strict policy against using child labor for electronics and other assemblies. Sometimes I would accompany the purchasing agents to see the complete production line to ferret out this practice. Sorry, I digress, but the amp was not assembled by trained assemblers. For example, the instillation of the 10W 750 ohm square white resistor violates component placement standards, can't dissipate heat and the leads are soldered way too close to the body of the resistor. Yeah. TLDR. 🙂
This amp was assembled in the USA by a older gentleman who should know this stuff.
A good and insightful video! Thanks for showing your mistakes and lessons learned.
I definitely wouldn't install any resistors on the outside; what can be done is using a square aluminum heatsink and drilling & tapping the holes in it so that you could install it on the opposite side of the chassis right behind the resistor. Use thermal grease, removing the powdercoating/paint may be a good idea too.
Separating the wire bundle and rerouting it on the outside to stay away from the big resistors is definitely a good idea. Go for it.
Another good idea (especially if you fabricate the chassis on a CNC mill) is making a circle of holes around each tube socket. That's how they did it back in the day - and convection helps suck the air from beneath the chassis. I'd opt for a perforated cover on the bottom unless a specific air flow path through the chassis is desired, in that case I'd place ventilation holes where needed.
I didn't build this, I am fixing someone else's mistakes :)
@@SkunkieDesignsElectronics ah, brain fart on my part! I often have this "Oh-I-Messed-Up-So-Bad" moment when looking at my designs from a few years ago.
Great work, Stephe.. I would remove the cable ties near the filament transformer and fold the cables and put new ties to keep it away from the shielded input signal wire, like you’ve shown..
If these were my amps, for sonics and my taste, I would replace the white coffin WW 2A3 cathode R with a couple of Kiwame 5W 1.5k Ohms in parallel “raised” a bit like you’ve shown earlier, and put some holes on the cover directly below them.. Then change all the Cathode Bypass Caps (driver and output) to Elna Silmic II 100uF/100V.. Then the change the coupling caps to 0.22uF/630V Amtrans AMCY PIO aluminum foil in oil, along with the new 330k grid leak.. That should mitigate the less than “organic” tone..
Great Video again! 👍Thanks. The few amps I built, I didn't put any bottom either. Perfect.
Those Dale (wirewound?) Black resistors do tend to heat up like stove elements.. can choosing a higher wattage ie: 20w instead of 10w help to reduce the amount of heat produced? Thanks!
Thanks for the interesting video Stephe. I had heat issues with the first amp I built, all the same issues that you are pointing out. Thanks for sharing your wisdom.
Marc in Ottawa.
Really wacky to build an amp with no thought given to ventilation or component spacing. The 3D space in that chassis is valuable real estate!
What you typically want to do is punch holes around the tubes. The tubes are the hottest items and as the air rises off of them it will create low pressure underneath, drawing a lot of air through the chassis. Holes also need to be on the bottom chassis close to the tubes to suck cool air in from the outside into the low pressure zone.
Lovely 🌹🌹🌹 idea _ what about adding a quiet fan in addition to your mods. That amp should have holes all over the place everywhere !
Another great video I really haven't thought about before, Right On! I look forward to your test video.
Hi there - great review of thermals - where do you source those 1 inch vent screens?
ventmastersstore.com/collections/small-round-vents/open-screen-vents-tab-style
Also they are on Amazon
2a3 I have read is low distortion tube...so not the best for 2nd harmonic etc all that heat will change the values of the components as well affecting performance, cant believe there were no ventilation holes etc..great job sorting it out steph...
50 years ago I was a 2 way radio tech. most were tube sets and they really ran HOT. proper cooling is the correct way to do it.
It seems to be otherwise a well-made amp. To overlook the heat issue seems rather out of keeping with the rest of the quality/ The wiring comments are smart too
Yeah, I was surprised at this, especially putting that hot cathode resistor physically touching the bypass cap!
YES ...AIR MOVEMENT FOR ANYONE WHO IS HOT, AND HOT AMPS !!! 👍🫠