2 ms·

Tubes are still sometimes used in specialised high power applications – typically broadcast TV, or for very high power RF amps for typically either military pur

Tubes are still sometimes used in specialised high power applications – typically broadcast TV, or for very high power RF amps for typically either military purposes or MRI scanners (my domain!) where you really want the ability to blat tens of kW of RF at a person for a very, very short period of time. They break. They get hot. They contain high voltages that can kill people. They're basically the definition of a nonlinear device. Yet for some reason, they're the best in their niche – if you want a tube amp that can work up to 112 kW, Eimac will sell you one [1]. Some of them come with wonderful descriptions like "The tube does not require a socket as it is designed to bolt directly to the chassis by means of the grid flange. Cathode and heater connections are made by bolting directly to the amplifier circuitry." Solid state (GaN / IGBT) are getting there but there are still new machines with tubes in being produced today. Of course, they have to (literally) warm up and are very annoying to work with – and can kill you – but, hey, valves aren't totally dead yet… [1] https://www.cpii.com/product.cfm/1/22/174 https://www.cpii.com/product.cfm/1/22/174

analog31 4y ago
Those things are works of art. Of course, 112 kW into 50 Ohms is something like 2400 Volts, so you're still going to get killed if it's solid state.
datpiff 4y ago
> They're basically the definition of a nonlinear device I don't think this is right. Triodes and pentodes have very linear operating regions. More so than a lot of solid-state devices. We make modern circuits more linear by using lots of active devices with negative feedback.
- azalemeth 4y ago
  This is completely true and you're right – smaller triodes and pentodes do have a very linear operating region (Wikipedia has a nice example [1]). I believe, perhaps erroneously, that the larger, high power variants however decidedly have their "foibles" however – an example higher power curve is here [2]. There are plenty of linear operating regions. The trouble is largely that they're not going to have the same gradient! I realise it doesn't matter that much when you've designed the amp, but in practice I have wondered about e.g. 275 µs delay times (at ~50 MHz) that may well be due to amplifier imperfect phase responses away from the frequency they were designed to work. [1] https://en.wikipedia.org/wiki/File:TriodeECC83Characteristic1.png https://en.wikipedia.org/wiki/File:TriodeECC83Characteristic... [2] https://imgur.com/a/D622n2E https://imgur.com/a/D622n2E

analog31 4y ago

Those things are works of art. Of course, 112 kW into 50 Ohms is something like 2400 Volts, so you're still going to get killed if it's solid state.

datpiff 4y ago

> They're basically the definition of a nonlinear device I don't think this is right. Triodes and pentodes have very linear operating regions. More so than a lot of solid-state devices. We make modern circuits more linear by using lots of active devices with negative feedback.

azalemeth 4y ago
This is completely true and you're right – smaller triodes and pentodes do have a very linear operating region (Wikipedia has a nice example [1]). I believe, perhaps erroneously, that the larger, high power variants however decidedly have their "foibles" however – an example higher power curve is here [2]. There are plenty of linear operating regions. The trouble is largely that they're not going to have the same gradient! I realise it doesn't matter that much when you've designed the amp, but in practice I have wondered about e.g. 275 µs delay times (at ~50 MHz) that may well be due to amplifier imperfect phase responses away from the frequency they were designed to work. [1] https://en.wikipedia.org/wiki/File:TriodeECC83Characteristic1.png https://en.wikipedia.org/wiki/File:TriodeECC83Characteristic... [2] https://imgur.com/a/D622n2E https://imgur.com/a/D622n2E