advertisement


Valve lifespan question...

JTC

JTC

PFM Villager...
Hey folks, simple question here: if I have a valve amp with power tubes (e.g. EL34, KT88, whatever) is the lifespan of those valves linked to the volume I listen at or simply the amount of time the valves have been powered up.

This is just curiosity on my part, but I wonder if a (say) 50w valve amp playing at super low volumes for 95% of its life would get significantly more playing time out of its valves than one playing at higher levels? Or is it relatively unrelated to actual volume/load? Also, is the power consumption noticeably higher when the amp is playing louder (again, just curious)?

thanks.
 
I would say no, but what could prolong the life of the power tubes is related to the type of power supply, slow start up time but there are others who know more than me.
But the valves are running constantly so changing the volume would make no difference.
 
Afaik power consumption will be pretty much constant, and won’t vary with volume (or, if the amp has the ability to switch between them, triode vs ultralinear modes either, despite the difference in output wattage).

Bit miffed about that with my amp, which I thought used about 200W less than it actually does (wrong info on the website). Still, at least I’m with Bulb so it’s all renewable!
 
Depends on the type of amplifier but in general valve amplifiers will run at a fairly constant current unless pushed very hard. As such it will make little difference if it is in use or merely powered up, of course if you leave it powered up but not in use for long period this is a pointless waste of power and valve hours.

It also won`t make much difference what mode they are used in, what will make a difference is how close to the maximum dissipation they are running at. Amplifiers which are conservatively run will have much longer lasting output valves.

Signal valves will have a much longer life, (tens of thousands of hours) and will not concern themselves with the existence or non existence of actual signals.
 
Volume level will have no effect on lifespan . The bias point of the output valves will determine how long your valves will last .
 
I don't think it makes any significant difference. It's a bit like car fuel for me...get something decent and it'll help the car in every way...same with valves. Buy a good make and they'll sound better and last longer IME.Unless your amp runs 2 or 3 very exotic valves it's not a big expense. I used to reckon 1000-2000 hours from a set, and as Barry says ^, almost a lifetime from the signal valves.
 
early said:
Volume level will have no effect on lifespan . The bias point of the output valves will determine how long your valves will last .
Click to expand...

I’ve always been curious about this with regards to hi-fi vs. guitar amps, e.g. I’d have expected to get a longer lifespan out of the valves in my Leak Stereo 20 than say Brian May does from pretty much the same valve set in a flat-out and highly distorting Vox AC30. I guess my question is does distortion/over-driving cause valve damage?
 
The valve doesn`t know about distortion, it only knows about dissipation, if the Guitar amp achieves its effect with a flaky power supply and an undersize output transformer it doesn`t follow that the valve is being overdriven.

It probably is though, but then I suspect that normal usage hours of a guitar amp is a lot lower than a hi-fi system used for several hours a day (say).
 
Most push-pull valve amplifiers are running in a rich class AB, so thrashing them does increase power dissipation and shorten life.
 
davidsrsb said:
Most push-pull valve amplifiers are running in a rich class AB, so thrashing them does increase power dissipation and shorten life.
Click to expand...

Actually, if you are running in the class A region, which will be the vast majority of the time, disspation in the valves will go down.
The dissipation is "shared" more to the speakers.

Naff all chance of this making any difference with Tannoys and with music that resembles music though.
 
Barrymagrec said:
I think the ratio of peak to average in normal music will mean it makes very little difference.
Click to expand...

Agreed. Plus the amount of class A is pretty high in most valve amps and in fact up to 30W ish (generalising) the majority are class A. With higher power amps then theoretically they should wear faster at high volumes but we are talking stupidly high complete headbanger type volumes here, way above what any sensible person would ever play at, and it would have to be used like that for a good percentage of the time to make any difference.
 
With modern higher power valves you've done well if you get enough life out of them to notice wear... this especially applies to higher power amps and using eg KT88's and similar. They will often just "blow", due to dodgy manufacturing and parts, well before they wear out... The higher the power relative to the number of output valves used in parallel per channel the worse it gets. eg 4 KT88's per channel giving 100W should be far less risky than one pair giving 85W as much higher voltage is needed in the latter case, with the proviso of course that there are more to go wrong in the first case. It's rather a lottery!
 
A related question, while this thread is still active - do start ups and shut downs affect valves?

I.e. putting aside considerations such as sounding better when warmed up, would a valve be happiest and longest lived (in terms of total hours lifespan) if it were left on the whole time, or doesn't it make any difference if it's frequently switched on an off?
 
Dave***t said:
A related question, while this thread is still active - do start ups and shut downs affect valves?
Click to expand...

If the system is in constant use, e.g. early valve computers, radio station etc, then yes, there is a benefit to not turning off for the odd hour or so it may be unused. In every other scenario powering off is hugely advantageous. Leaving a hi-fi amp powered is just insanity as they tend only to be used a couple of hours a day.

A gentle power-up is beneficial though, a lot of guitar amps have a ‘stand by’ switch, some rectifiers such as the GZ34 ‘soft start’ etc. At the museum I volunteered at pre-covid we use a huge variac to bring the large valve computer up gently. It starts with the heaters on at half voltage and the different supplies are brought in gently at a predetermined time to preserve valve life (it has over 500 valves).
 
Valves are in many ways extremely similar to lamps. How many times have you switched on, or off, a filament lamps and heard it go "ping"? The same could easily happen in a valve, but that is just one of very many failure mechanisms, even for a simple filament lamp. For either a lamp or a valve, going from room temperture to very hot in next to no time, produces huge stresses.
 
One of the reasons for valve computers having slow switch on for the heaters was that the heater transformers were huge high current components with a very low output impedance and so would provide a massive overcurrent surge to cold heaters. Domestic amplifiers have small transformers and tend to be partly self limiting as far as switch on surge is concerned.

If you look at the regulation figures for a range of transformers you will see how the regulation between no load and full load reduces as the VA rating of the transformer increases.

Slightly off topic, I once visited a very large valve analogue computer (to look at a tape drive), the maintenance engineer very proudly told me that they had recently saved 3 Kw of heat by changing from metal rectifiers to silicon diodes in the DC heating circuit for the valves.
 
A constant current heater supply can be beneficial to heater life... in theory anyway! The lack of one is why many Mullard valves glow like a 20W bulb for a second when power is first applied.. it amazes me that they seem to survive this no problem!

Either applying HT without heaters on or vice versa can be very bad for valves. Special versions of valves were developed for computers to get around this.
 
Vinny said:
Valves are in many ways extremely similar to lamps. How many times have you switched on, or off, a filament lamps and heard it go "ping"? The same could easily happen in a valve, but that is just one of very many failure mechanisms, even for a simple filament lamp. For either a lamp or a valve, going from room temperture to very hot in next to no time, produces huge stresses.
Click to expand...

I’d still argue valves were a very robust device assuming the circuit was good. There are many Quad IIs around still running their original GE KT66s and I have a lot of used small signal valves (ECCxx types) that measure and sound fine. I certainly have no issue using very rare and collectable/valuable valves of this type as I suspect they’ll see me out, plus I’ve got some spares. I’m currently listening to vinyl in my main system and the Verdier preamp is stuffed with 6 very nice Mullard valves. I’m even using good NOS stuff in my guitar amp, the position 1 there is a early ‘60s Mullard box-plate that is about £200 worth if you believe eBay (I paid nothing even approaching that). I’ll likely get 20+ years out of it, in fact it will almost certainly see me out.


The computer (SSEM replica) at the museum was assembled 22 years ago now, mostly from period-correct NOS components, but there are a lot of used parts too. You would be surprised how infrequently we have to change valves given it has 500+ of the things. It is a fairly rare occurrence, and mostly because a whole area of circuitry is ‘off’, which is often traced to an out of spec ancient carbon resistor, or even a circuit error where something was borderline to start with. It ran up to four days a week for about four hours a day, so that is quite a lot of use over 22 years. Way, way more use than anyone is typically going to give a domestic hi-fi amp. The only caveat is as stated above we bring it up gently and the valves used are proper high-quality just post-war NOS stuff, Mullard, GE, RCA etc as most of the valves are no longer made at all (it mainly uses EF50s & EF55s). No naff Chinese valves there at all.

I would definitely recommend buying good valves. They really are not created equally either sonically or from a reliability perspective.
 
Barrymagrec said:
One of the reasons for valve computers having slow switch on for the heaters was that the heater transformers were huge high current components with a very low output impedance and so would provide a massive overcurrent surge to cold heaters
Click to expand...

The same is true of (very) high power filament lamps, now almost? totally? a thing of the past. I am unsure where running gear started in terms of power but certainly once you got to kWatts of filament lamp, for theatre and outdoor flood lighting, the filament gets/got to be very thick wire and at switch on it was basically a dead short. The automatic ramp-up running gear for a few kW of theatre lighting is/was a huge bit of kit and hugely heavy - WAY larger than any running gear needed for any discharge lamp that I have ever seen, and I have seen them up to something like 20kW.
 
You must log in or register to reply here.


advertisement


Back
Top