that and you can use thinner walled tube, afair the required wall thickness /diameter is a more or less constant ratio for the same rated pressure
so a boiler made as a single big tank would need very thick walls to take 500 psi
-Lasse
Nope. I mis-typed a * where there should be a + in the last line. It should be: area/vol = 1/h + 2/r Note that this is for a cylinder with one end open.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
But, doesn't 1/x change at a rate that depends on the value of x? The rate of change (slope) is -1/x^2, yes?
It doesn't matter. Leave both ends open and you get 2/r. It still means an increase in surface area/volume, hence, John's statement stands.
In fact, a 2-dimensional math problem I once had said to find the shape of the greatest enclosed area for the smallest perimeter. It turns out to be a circle. You can try it out using quadrilaterals (rectangles) for a simple example and you will find that a square is the answer.
This extends to solid figures.
I learned this when I ran into heat transfer of large magnetics I used to design. Some required more cooling ducts as they got larger for a constant watts/volume loss (Cu and Fe together). I investigated and decided that the smallest surface area enclosing a volume is a sphere.
Aside:(Small transformers are designed with a target regulation rather than with heat in mind. Some can even withstand a shorted output without overheating (due to the area/volume ratio). At some point, as size increases, that is not true. I never tried to find the crossover point because I had a job to do and, frankly, I probably don't have the mental tools required.
With more thought on this, I decided that the smaller the warm-blooded creature, the more energy input it needs to maintain a given body temperature (when active) so shrews and humming birds eat a lot. In other words, if small, you give up your internal heat more rapidly because of your surface area. It is also why smaller water droplets evaporate more rapidly than large ones.
I challenge you to find an object which has less surface area/enclosed volume than a sphere.
This is a fun subject.
The temperature gradient in the envelope could have caused cracks at the glass-to-metal seals.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 USA +1 845 480 2058 hobbs at electrooptical dot net http://electrooptical.net
I'm doing some pulsed bar laser drivers, and they need, say, 4000 to 10,000 uF at 35 or 50 volts, depending on the laser stack. So far, only aluminums look feasible, and they get tricky below 0C. I wonder why nobody makes polymers in high CVs. If I used a lot of small polymers, packaging would be messy and the array would be very expensive.
We do use polymers for conventional stuff, like in the power supplies for FPGAs. They are great there, but those are values like 180 uF/6 volts, dinky numbers like that.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators
Applies to a water tube boiler (Babcock, etc), such as are used in power stations and ships, where the pressure is inside the tubes.
A locomotive (fire tube) boiler has the pressure acting inward on the tubes, and outward on the containing vessel. 200-250 PSI was usual with those.
--
"Design is the reverse of analysis"
(R.D. Middlebrook)
Did you actually check to see if there was a reduction in envelope temperature? If not, then the increase in surface area or other heat transfer mechanisms were ineffective in performing this function.
A cylindrical tube screen actually increases temperatures at the glass envelope; it is intended to act as a screen only in front ends and low-noise circuits, without consideration of the side-effect. Some served also as retainers (or vibration dampers), where there was a likelihood of the part falling or vibrating out of its socket, but a better retention mechanism would not interfere with airflow.
If the 'sleeve' doesn't have an integral flat surface, intended to mate with a heat exchanger, it's not intended to act as a heatsink.
The tooled foil stock used in the cuffs I've described and illustrated were dramatically effective in reducing temperature of the glass envelope, and the studies quoted monitored this variable scrupulously.
The heat-exchanger, flat surface types were for use in equipment where volume and airflow was restricted (and vibration also possibly a serious issue), as in airborne or mobile modules. The cap clamps from Aavid, strangely enough, also count strongly on heat transfer to the cap mounting base for their effectiveness.
The cuffs that I describe and demonstrate are better applied in larger volume systems, where low cost tooling can convert raw or anodised aluminum slit roll stock. This application has been in the public domain since it's publication date.
RL
Can't you use them both?
I don't think so. That would have shown up as a white or transparent border around the normally silver colored getter. Something like this: As I vaguely recall, there were some gassy tubes, but nobody bothered to check if they were gassy before we added the tube heat sinks.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
No. What happened was my fearless leader went to lunch with one of our competitors, who casually mentioned that they're installing heat sinks on their mountain top repeaters. Fearless leader immediately went to the local surplus store, and bought up a fairly large supply of such tube heat sinks. My job was to install them. What better authority on the topic than our competitor? Testing? Nada.
The building and antenna farm is on the lower right:
The drive from downtown Stanton CA to the top of Santiago Pk takes about 2-3 hrs, depending on road and vehicle conditions. On this day, both were in dubious condition. To avoid returning at some ridiculous hour, I gave myself about 3 hrs to replace the shields with heat sinks, and retune the radios. Yes, changing shields required retuning. This was on a weekend, so there was little traffic.
After dealing with the shields, I had a few dozen other things to deal with. I thought about measuring the temperature, but with limited time, and being by myself, I elected not to rip the thermometer off the wall and check if there was any measurable difference. Besides, the mercury thermometer didn't have the right temp range.
The surface area of the tube heat sink was the same as that with the ordinary tube shield. However, the heat sinks were painted black, thus improving radiation.
In this case, GE Progress Line radios, the shields were there to keep the radio from leaking and picking up RF through the glass envelopes. Take of the shields and they would not act as a very good repeater due to self interference, intermod, desensitization, oscillatory tendencies, etc. I tried running a radio without shields once, and found that it was tolerable for simplex operation, but useless for duplex (repeater) operation.
The space between the glass envelope was filled with a flat spring, with fingers extending out in both directions. When the shield was inserted over the glass envelope, the fingers flattened and provided about 80% surface contact area on the envelope surface. Spring steel is not the best thermal conductor, but should be adequate to simulate a heat sink.
My guess(tm) is that the "hard" quartz glass envelope of the tube is fairly IR transparent. Any heat that I measure on the surface came from the tube and through the glass as IR. I couldn't find any references that actually claimed that whatever glass used for vacuum tubes was IR transparent. The silver colored tube shield reflected some of this heat back into the tube, which certainly didn't help. The rest was radiated outward and removed by air convection through the gap between the envelope and the shield. The black colored probably radiated the trapped heat more efficiently, but there was no convective air flow between the envelope and the tube shield as the space was occupied by the flat spring. So, more heat remained inside the envelope, resulting in premature failure.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Yep. My guess(tm) was that it was the loss of convective air flow between glass envelope and the flat spring inside the tube shield, that produced most of the heating. Putting a possible IR reflector directly against the glass certainly can't help. Without the shield or heat sink, the IR heats up the neighboring cans, shields, and devices. With the tube shield or heat sink, those devices remain cooler, but the shield gets hotter.
I couldn't find much on what type of "hard" glass was commonly used. Nonex would be my guess to minimize dimensional changes with temperature.
Later Eimac tubes, such as the 3-500z, had graphite plates to help handle higher temperatures.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
The missing link: Note the orange graph showing glass contraction vs temp. Looks like Uranium glass is better than Nonex.
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
On 8/3/2013 5:55 PM, Jeff Liebermann wrote: The capacitor ESR decreases with increasing temperature.
That depends on the type of capacitor. Be careful about generalizations.
Well, dy/dx depends on x in both cases (-1/x^2 and e^x, respectively), but one is proportional, the other is much steeper. The exponent function simply isn't as "tall". :)
Tim
-- Deep Friar: a very philosophical monk. Website: http://seventransistorlabs.com
Electrolytic ESR changes a little as they heat up, but the real action is below
0C. ESR can go up 10:1 between 0 and -40.Polymers don't do that, but their energy density is low.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators
Hard quartz I believe is, but as far as I know, they would've never used such expensive, difficult material -- even the mil spec ones are borosilicate something or other. I have a Bendix 6384, basically a
6L6-turned-brick-shithouse, which claims 300C peak envelope temperature. I think they gave a trade name for the glass or base, but I forget. Easy to look up in either case.Easy way to tell would be to grab an IR thermometer, or camera better yet: if the outline of the plate alone is visible through the glass, and it starts glowing brighter, immediately after applying power (you'll have to preheat the cathodes), then you've definitely got it; I suspect most tubes will glow diffusely as the whole glass envelope, taking much longer to heat up, since there's that extra heat transfer step.
Or, poor man's method: fingers. Does radiant heat increase suddenly while the glass remains cool to the touch, or is it only the glass that gets hot and emits radiation?
Tim
-- Deep Friar: a very philosophical monk. Website: http://seventransistorlabs.com
correct, but the fire tube also started with one pipe and more was added when they realised they got more area and they didn't have to make them as thick
-Lasse
My big red Radiotron book has practically nothing to say about tube thermal design or cooling. Neither does Terman's Radio Engineering 4e. Thermal design was neglected then, and is often neglected now.
-- John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators
That might be because thermal management was considered the mechanical engineers area of expertise. Broadcast engineering books have thermal management calculations, but that involved far more heat than found in an electrolytic or small vacuum tube.
Add on vacuum tube heat sinks seem to be popular with the tube amplifier crowd: "the sound from a component utilizing Cool Dampers can be heard to have more controlled and deeper bass, a cleaner sound stage, crisper and greater dynamic range, and better focus of the entire sound field." Ummm.... right.
"The Cool Damper also reduces the operating temperature of the glass envelope of an electron tube by about 10 %, which of course acts to appreciably extend the lifetime of the tubes." Envelope temp is typically about 250-300 C. At 25-30 C drop is quite a bit and probably worthwhile (if the heat sink is for real).
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.