The Naked Scientists

The Naked Scientists Forum

Author Topic: Why did Vacuum Tube Technology tolerate heavy loading on the finals of an amp?  (Read 9252 times)

Offline tommya300

  • Hero Member
  • *****
  • Posts: 655
    • View Profile
Is there some reason that the old technology can take more of a Licken and Kept on ticken.
Old Guitar amplifiers that used vacuum tube technology have been a prize to a musician, as it was once explained to me. They were more reliable as explained to me and the details escape me.
« Last Edit: 12/07/2010 00:25:54 by tommya300 »


 

Offline Soul Surfer

  • Neilep Level Member
  • ******
  • Posts: 3345
  • keep banging the rocks together
    • View Profile
    • ian kimber's web workspace
Your question is not understandable and your note does not do anything to clarify this.  If you want an answer please give a more detailed description of your question and avoid unclear colloquial statements please.
 

Offline syhprum

  • Neilep Level Member
  • ******
  • Posts: 3821
  • Thanked: 19 times
    • View Profile
Vacuum tube technology is vastly less robust and reliable than properly designed transistor based technology in all normal cases.
I can only think that the musician in question encounted a poorly designed amplifier that was damaged by short circiting the output which of course is a matter of poor design.
it has been suggested that vacuum tubes might more reliable than semiconductors in an high radiation environment but I do not know of any current applications.
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Valves are perhaps more naturally current limiting than bipolar transistors, so that might account for the observation, but as Syhprum says, a good design will take that into account.
 

Offline Soul Surfer

  • Neilep Level Member
  • ******
  • Posts: 3345
  • keep banging the rocks together
    • View Profile
    • ian kimber's web workspace
Thanks for the improved explanation.  Valves are naturally current limiting because of the emission limits on their cathodes they are also connected to loads via transformers which are also power limiting devices so short circuiting them does not usually blow up the amplifier.  Valves are also physically larger hotter and the power densities are much lower and no special heat management techniques are needed. Transistors however are current operated devices that are connected almost directly to the low impedance load that loudspeakers usually present and unless there is a protective circuit in the design to limit the drive to the output stage when overload is detected very large currents can flow.  Transistors are also much smaller and operate at higher power densities and damaged if junction temperatures get above 150deg C so they need heat sinks and sometimes fans to keep them cool.  Brief overloads can easily damage them.
« Last Edit: 12/07/2010 09:23:09 by Soul Surfer »
 

Offline syhprum

  • Neilep Level Member
  • ******
  • Posts: 3821
  • Thanked: 19 times
    • View Profile
Conversely valve amplifiers do not like working into an open circuit, excessive voltages are developed that may damage both the valves and the output transformers.
Some vacuum tube devices have survived into the transistor age, the Hubble telescope system employs a vacuum type image intensifier, and large radar systems employ klystron transmitter tubes while until recently at least travelling wave SHF amplifiers have been used in satellites.
« Last Edit: 12/07/2010 11:23:13 by syhprum »
 

Offline tommya300

  • Hero Member
  • *****
  • Posts: 655
    • View Profile
Thanks for the improved explanation.  Valves are naturally current limiting because of the emission limits on their cathodes they are also connected to loads via transformers which are also power limiting devices so short circuiting them does not usually blow up the amplifier.  Valves are also physically larger hotter and the power densities are much lower and no special heat management techniques are needed. Transistors however are current operated devices that are connected almost directly to the low impedance load that loudspeakers usually present and unless there is a protective circuit in the design to limit the drive to the output stage when overload is detected very large currents can flow.  Transistors are also much smaller and operate at higher power densities and damaged if junction temperatures get above 150deg C so they need heat sinks and sometimes fans to keep them cool.  Brief overloads can easily damage them.

Thank You for pointing it out! I was trying initially, to fit it in to the title space.
Everyones description jogs something I recognize! The vacuum tube technology was coming to a close and the desired electronics was focused on that. Everyone used these early amplifiers the best money can buy at the time.
The amps were so strong that the speakers would max out and begin to crack in sound which was like losing their load values.
This would eventually hurt the circuits only if the system remained driven for several minutes. Early semiconductor electronics in this state would burn out in seconds of losing the load. Thank you all this refreshed some of my memory is was so long ago it is like a bit foggy.
 
« Last Edit: 12/07/2010 14:08:18 by tommya300 »
 

Offline syhprum

  • Neilep Level Member
  • ******
  • Posts: 3821
  • Thanked: 19 times
    • View Profile
"Early semiconductor electronics in this state would burn out in seconds of losing the load".
This statement really puzzles me, they never do nor never have cared about losing the the load (ie running into an open circuit ) this was one one of the things I really liked about transistor amplifiers when they first came in.
 

Offline LeeE

  • Neilep Level Member
  • ******
  • Posts: 3382
    • View Profile
    • Spatial
Vacuum tube amplifiers tend to produce predominantly even ordered harmonic distortion, rather than the predominantly odd ordered harmonic distortion produced by solid-state amplifiers and in listening tests most people seem to find even ordered harmonic distortion preferable, and even beneficial to the music, compared with odd ordered harmonic distortion, which most people find positively objectionable.

This is especially noticeable with electric guitar amplifiers, where some degree of distortion is more or less essential; an electric guitar played through a very 'clean' amplifier tends to sound rather lifeless and dull when compared with the same guitar played through a vacuum tube amplifier, even when the vacuum tube amplifier is set to produce little audible distortion.

When high levels of distortion are actually wanted, there's really no comparison between the qualities of the two types of distortion produced: the 'natural' odd-ordered distortion produced by solid-state amplifiers sounds wasp-like and buzzy, whereas the even-ordered distortion naturally produced by vacuum tube amplifiers tends to sound thicker and more smooth or 'creamy'.
 

Offline tommya300

  • Hero Member
  • *****
  • Posts: 655
    • View Profile
Vacuum tube amplifiers tend to produce predominantly even ordered harmonic distortion, rather than the predominantly odd ordered harmonic distortion produced by solid-state amplifiers and in listening tests most people seem to find even ordered harmonic distortion preferable, and even beneficial to the music, compared with odd ordered harmonic distortion, which most people find positively objectionable.

This is especially noticeable with electric guitar amplifiers, where some degree of distortion is more or less essential; an electric guitar played through a very 'clean' amplifier tends to sound rather lifeless and dull when compared with the same guitar played through a vacuum tube amplifier, even when the vacuum tube amplifier is set to produce little audible distortion.

When high levels of distortion are actually wanted, there's really no comparison between the qualities of the two types of distortion produced: the 'natural' odd-ordered distortion produced by solid-state amplifiers sounds wasp-like and buzzy, whereas the even-ordered distortion naturally produced by vacuum tube amplifiers tends to sound thicker and more smooth or 'creamy'.

So to a trained ear (e.g. a musician) would rather desire the vintage vacuum tube amplifiers?
« Last Edit: 13/07/2010 01:49:48 by tommya300 »
 

Offline tommya300

  • Hero Member
  • *****
  • Posts: 655
    • View Profile
"Early semiconductor electronics in this state would burn out in seconds of losing the load".
This statement really puzzles me, they never do nor never have cared about losing the the load (ie running into an open circuit ) this was one one of the things I really liked about transistor amplifiers when they first came in.

If I may explain what I think the part of the amplifier I was refering to... is the driver and the final stage, that has a push and pull design. Correct me please if I am at a miss. I am refering to the early transistor amps. I think more safe guards and newer circuit designs may have reduced the sudden  accidental distructions.
 Two identical electrically balanced circuit with a driver and a power transistor, that are sort of back to back. I think one side takes care of the positive going wave the other takes care of the negative going wave. Power dissipation is through heat and sound waves through the speakers.
Short circuit these outputs will it heat the drivers and takeout the finals at the same time?
Open circuit the power has to go somewhere, if there is no load  would it also heat up and allow fry these stages. Not to mention sometimes feedback.
Am I close or an I a bit disillusioned in this area? From time to time this happens?
« Last Edit: 13/07/2010 01:47:46 by tommya300 »
 

Offline LeeE

  • Neilep Level Member
  • ******
  • Posts: 3382
    • View Profile
    • Spatial
So to a trained ear (e.g. a musician) would rather desire the vintage vacuum tube amplifiers?

Well no, not quite.  They'd probably prefer an amplifier that produced the least distortion.

Also, you seem to have fallen into the trap of assuming that all vacuum tube amplifiers are old i.e. 'vintage', and ended up drawing the erroneous conclusion that all vacuum tube amplifiers are therefore more primitive and inferior in comparison to modern solid-state amplifiers.

There are quite a few very modern vacuum tube amplifiers around but because the thermionic valves are much more expensive to produce, not only because of their mechanical complexity but also because of the reduction in the scale of manufacturing, modern vacuum tube amplifiers tend to be considerably more expensive than an equivalent solid-state amplifier.
 

Offline tommya300

  • Hero Member
  • *****
  • Posts: 655
    • View Profile
So to a trained ear (e.g. a musician) would rather desire the vintage vacuum tube amplifiers?

Well no, not quite.  They'd probably prefer an amplifier that produced the least distortion.

Also, you seem to have fallen into the trap of assuming that all vacuum tube amplifiers are old i.e. 'vintage', and ended up drawing the erroneous conclusion that all vacuum tube amplifiers are therefore more primitive and inferior in comparison to modern solid-state amplifiers.

There are quite a few very modern vacuum tube amplifiers around but because the thermionic valves are much more expensive to produce, not only because of their mechanical complexity but also because of the reduction in the scale of manufacturing, modern vacuum tube amplifiers tend to be considerably more expensive than an equivalent solid-state amplifier.

I believe you are most correct about that trap. Just because I go into radio shack and other consumer electronic sales retailers are not the do all end all. That is fore sure obvious now.
Thank you all for the eye opener.
.
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
It was not unusual for early transistor amps to have a single rail power supply and use a capacitor to block DC from the loudspeaker. This was OK, but it was not a good idea to plug speakers into an amp that was powered on, even if it was not actually outputing any signal.

The problem with this arrangement is that it results in a rather large current flow through the positive side final transistor while the capacitor charges up. If the design had insufficient current limiting resistors and there was a momentary short circuit between the output and ground, it could easily take out that transistor.

Later designs tended to use dual power rails which allowed the capacitor to be eliminated. In this case, "hot plugging" is a lot safer, at least when the amp is "quiet".

There are some circuits on this link that show the differences.

http://sound.westhost.com/project12.htm
« Last Edit: 14/07/2010 07:17:57 by Geezer »
 

Offline chris

  • Neilep Level Member
  • ******
  • Posts: 5339
  • Thanked: 65 times
  • The Naked Scientist
    • View Profile
    • The Naked Scientists
In the "old days" valves were all the rage, at least in amplifier terms; but then they were pretty much all we had, so they had to be.

With the advent of the transistor came the trend of "going solid state"; this offered the clear advantages of improved lifetime, reliability, cost (including running cost) and more faithful signal / sound rendition.

But therein lay the problem, because although digital devices did maintain the integrity of the original signal, the distortion and sound colouration originally added by the valves was missing, meaning that digitally rendered music or speech could be left sounding a little thin and lifeless as a result.

Part of the reason for this is that unlike digital circuitry, which will just "clip" noisily whenever the sound level "peaks", valves are much more tolerant of high signal and will just saturate like a car running into wet sand. This means the excess energy in the signal just gets progressively soaked up, rather than producing unappealing artefacts.

Consequently, in recent years, valves have come back into vogue; usually working as a combination of valve and digital technology. The TLAudio preamp I use for my broadcasting is valve driven and sounds incredible compared with a solid-state counterpart.

Chris
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Chris,

You seem to be suggesting that transistor amplifiers (solid state) are digital. That is not the case.

Power amplifiers using thermionic valves or transistors were still analog devices. Digital encoding and processing techniques didn't happen till long after the initial introduction of transistorized amplification.

The output stages of amplifiers have to be analog because loudspeakers (including the ones in headphones) are analog devices.

BTW - I tend to suspect the argument regarding clipping is a load of ball baloney that was concocted by a guy in marketing. There is no particular reason why exactly the same effect can't be reproduced by other means, although that does not mean that anyone has actually done it of course. Your valve amp may really sound better than anything else available.
« Last Edit: 14/07/2010 18:32:52 by Geezer »
 

Offline Bored chemist

  • Neilep Level Member
  • ******
  • Posts: 8665
  • Thanked: 42 times
    • View Profile
Valves are big. They can absorb a fair amount of energy without heating up much.
Transistors are small. They can not.
Also, a valve can survive being baked to a few hundred degrees; it will work fine once it cools down again. Transistors fail, irrevocably,  below 200C.
If there is some glitch that means that the power that is meant to be fed to the speaker gets dissipated in the valve or the transistor, the big heavy slow thing is more likely to survive because iyt won't get as hot, and if it did, it wouldn't matter so much.

You really don't need anti-static packaging for valves.
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Valves are big. They can absorb a fair amount of energy without heating up much.
Transistors are small. They can not.
Also, a valve can survive being baked to a few hundred degrees; it will work fine once it cools down again. Transistors fail, irrevocably,  below 200C.
If there is some glitch that means that the power that is meant to be fed to the speaker gets dissipated in the valve or the transistor, the big heavy slow thing is more likely to survive because iyt won't get as hot, and if it did, it wouldn't matter so much.

You really don't need anti-static packaging for valves.

That's part of the reason.

Bipolar transistors (the type used in early transistor amps) as somebody pointed out much earlier in the thread, are current amplifiers. In crude terms, you shove a current in one pin, and it tries to pass a current between the other two pins that is N times greater, where N is the gain of the particular transistor (N is anything but constant, even for a single transistor).

Valves on the other hand are a lot more like variable resistors, which is a bit ironic, because that's where the name "transistor" came from, despite the fact that valves are a lot more "transistory" than the first transistors.

The clue is, that despite their name, bipolar transistors don't really have any resistance! So, if you put them in a situation where they can pass a lot of current, they will cheerfully try to do that until they overheat and blow up internally.

On the other hand, valves are intrinsically current limiting because the maximum current flow is a function of the geometry and the supply voltages.

BTW, bipolar power transistors are virtually static immune, and you don't need antistatic packaging for them either.

 
 

Offline daveman

  • Neilep Level Member
  • ******
  • Posts: 1002
    • View Profile
Open circuit the power has to go somewhere, if there is no load  would it also heat up and allow fry these stages. Not to mention sometimes feedback.
Am I close or an I a bit disillusioned in this area? From time to time this happens?


No load, no current, no power, just voltage rail to rail.

 

Offline tommya300

  • Hero Member
  • *****
  • Posts: 655
    • View Profile
Open circuit the power has to go somewhere, if there is no load  would it also heat up and allow fry these stages. Not to mention sometimes feedback.
Am I close or an I a bit disillusioned in this area? From time to time this happens?


No load, no current, no power, just voltage rail to rail.


Wait a second let me clarify my post, the way I read it I can see it is not clear, it sounded like I was looking at the circuit of the amp in a Voltage bias state only with no audio input? Sorry Dave thank you for pointing that out to me. I read my post and the first thing came to mine WHAT am I talking about. I better make sure I get better sleep, saw more logs, before I post stuff like that.
Let me rephrase...
 
If the Driver section is pumping an audio signal into the base terminal of the final transistors while there is no load or improper loading on the collector relative to ground of the finals, there is a Power problem!

This depends on an instant in time duration and amplitude driving the finals the signal will kill the final push and pull circuitry.
The voltage drop will occur at the substrate level of the final transistor, distruction of the transistor will be faster than a blink.
I think it is almost similar to overdriving the amp.
.   
« Last Edit: 30/07/2010 04:51:32 by tommya300 »
 

Offline SeanB

  • Neilep Level Member
  • ******
  • Posts: 1118
  • Thanked: 3 times
    • View Profile
Geezer, you have never used bipolar RF power transistors, they are very static sensitive, not to mention load sensitive, heat sensitive and also have that lovely pink ceramic casing that carries the tag Beryllium Oxide - Caution Toxic. Only good thing was that when installed they really work well.

Valves are so hard to damage because they have a lot of surface on the anodes, and they will happily run till bright cherry red on overload, whilst the bipolar dissipates all it's heat in a tiny bit of silicon around 5 by 5mm, that has about the same heat density as the surface of the sun in the worst case.
 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Geezer, you have never used bipolar RF power transistors, they are very static sensitive, not to mention load sensitive, heat sensitive and also have that lovely pink ceramic casing that carries the tag Beryllium Oxide - Caution Toxic. Only good thing was that when installed they really work well.


No. I was thinking of something more like a 2N3055 which, I suspect, has about the same RF response as a slug :D

Yes, power density has a lot to do with it, but don't forget that "tubes" are intrinsically current limiting and they are dealing with relatively small currents whereas power transistors are dealing with relatively high currents and the only thing (within the transistor itself) that limits the current is the current gain.
« Last Edit: 31/07/2010 07:59:53 by Geezer »
 

Offline SeanB

  • Neilep Level Member
  • ******
  • Posts: 1118
  • Thanked: 3 times
    • View Profile
Geezer, I worked on equipment that was poorly designed, and had a switching transistor that was poorly heatsinked, and badly driven. Peak current was 40A and average was 10A, well beyond your common 2N3055. I used to select the transistors based on BE saturation voltage, as a few mV here made a difference ( along with selecting a 2N2219A driver that had a high gain and would saturate with the limited drive) between surviving a while or going short circuit in worst case or merely unsoldering itself at best. Eventually I got a good batch of BUX40 that would survive, the batch was at the top end of spec for Hfe and most did survive.

A 2N3055, depending on who made it, often has a gain roll off at under 10kHz, not what you would like in an amplifier. I did use some as active loads, by soldering leads onto then and dropping them into a ceramic cup ( old tea cup) filled with cold water. Driven with a 555 timer and used to test power supplies for load stability, they generally worked well until all the water had boiled off, then glowed white hot and burnt out.

 

Offline Geezer

  • Neilep Level Member
  • ******
  • Posts: 8328
  • "Vive la résistance!"
    • View Profile
Ah yes! The 3055 would not be the ideal choice for an output stage.
 

Offline tommya300

  • Hero Member
  • *****
  • Posts: 655
    • View Profile
Ah yes! The 3055 would not be the ideal choice for an output stage.

Can economics dictate and dominate the use of the 2n3055, when the application is just for a PA system? The vocal speaking bandwidth needed is just to accomadate 1KHz +/- 10KHz. Or for real cheap sounding Ghetto Blasters.
 

The Naked Scientists Forum


 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums