Dave - thank you for taking the time to write this clear and well thought out response, and I most assuredly appreciate the references supplied to answer my why?'s. Please bear with me for one second...
Craig - if you want my threadcrap out of your thread, please let me know and I'll see if I can get a mod to move the off topic posts to a new thread tomorrow. At this point, I don't want to start a new thread without moving posts as I don't think it serves the community well to separate anecdotal comments from conclusions. Your thread, your call - just let me know.
Okay Dave - I'm going to quote you out of order here because your last point is the most important:
As a science, electronic engineering principles are based on strictly defined definitions. Changing an established definition to suit ones needs is fine for the one, but when it is then arbitrarily offered as an established standard, it is as misguided as those who would say their definition of good sound is the standard as well.
I agree. I never stated what I do is an established standard, and in fact pointed out that I'm clearly the odd man out. I apologize if anyone took it to be a standard. Earth shattering changes in established electronics principals are not going to come from a discussion on a tube audio forum, and certainly not from me.
I have no doubt of what you were taught, or what you believe to be true. And designing by the very conservative measures you suggest can hardly be a bad thing. However, the concept you propose in defining what the meaning of a plate dissipation rating is goes against traditional teachings and standards. Consider that:
The RCA RC-30 tube manual (in part) defines plate dissipation as "the difference between the power supplied to the plate of the tube and the power delivered by the tube to the load."
I suppose that in this definition, you could stretch the concept of "power supplied to the plate" to mean what you will. But traditional teaching defines the power supplied to the plate of a tube as that specifically provided from an external source, available for transfer into a load for useful work to be performed, with the plate dissipation being the difference between the two.
While this clearly defines what plate dissipation is then, it still leaves what energies contribute to make up the dissipation rating for a given tube in question. That point can be addressed with the following:
1. Among many of the most notable vacuum tube reference texts, I have not been able to find any references that suggest that a plate dissipation rating considers anything other than power supplied to it from an external power source. If you have any reference texts that are contrary to this, it would be most helpful if you could provide that information for the rest of us. Also --
I do not.
2. In a classic example given in the same tube manual, an exercise is documented of an effort to obtain maximum Class A power output from push-pull triodes (page 31). The first effort produced an estimated plate dissipation of 16.2 watts per tube (using traditional calculations), for tubes with a plate dissipation rating of 15 watts. The text then explains the proper adjustment to be made to bring the dissipation down to the maximum 15 watt level.
Acknowledged and understood.
The point is, that nowhere in this discussion, is the idea of derating the plate dissipation rating ever mentioned because of the 6.25 watt heaters these tubes use.
You asked why heater dissipation and grid current should be exempt from the calculation of plate dissipation. The answer is because these issues (which are real), were almost certainly taken into account when the rating was originally published. Grid current certainly was by specifying a maximum grid #1 DC resistance value. At this point then, it is up to you to show some body of evidence contrary to these traditional understandings.
Okay. I'm going to state some things that may or may not be true, and ask that you help me understand.
If grid current is drawn, those electrons flow to the plate, yes? Wouldn't that mechanism also fall within the traditional definition of 'power supplied to the plate' you stated above?
I undertand the grid has a max dissipation rating and a max DC resistance value, but don't those specs address the energy the element itself can convert, and not the subsequent effect of that work product being delivered to the plate as power?
I also understand we're talking rather small numbers with grid current, so assuming what I'm saying above is directionally correct, it's also conceivable it's considered inconsequential. As a friend of mine once so eloquently put 'ya, but that's 1% of ****all."
As for the heater, I understand that its radiated heat, as you said is a real issue, though does not fall within the traditional definition of power delivered to the plate hence is not considered in the plate dissipation rating. Thanks for providing clear references.
I, as I'd guess anyone, would prefer it to be a bit tidier than that - a reference somewhere that states this was indeed taken in to account. Although your example is very strong and I've no doubt what was taught is represented accurately... you yourself had to say "almost certainly"
So, Craig has tubes that are red-plating 18.5 watts shy of their rated max plate dissipation. Postulating that this is not a normal operating condition:
Manufacturing error?
Incorrect max spec?
What was 'almost certainly' considered back in the day, wasn't?
Something else I'm not considering?
Any one, combination, or all of the above?
Well crud JJ! I forgot to ask you one thing. Do you also factor in the filament heat dissipation in your calculations? Sorry, I have an inquisitive mind and I firmly believe we never stop learning!
I do, BUT please note Dave's references above - it's not standard practice to do so.
