Tube life ?

[sdn25]

How long can I expect a set of current production Mullard EL34s (made by new sensor) to last in a cranked (all knobs on 10) 50w jmp 1987? I have it biased at about 50-53%.

 

[StrummerJoe]

Too many variables to even guess.

 

[sdn25]

Too many variables to even guess.

What other information would you need to make a guess ?

 

[StrummerJoe]

Variables would include how often you play, at what volume, length of time, unforseen events like vibrations when transporting, etc....

In most cases they should last years.

I have the same tubes in my Rivera Jake Studio combo. Had them installed back in January. Zero issues. I like them.

I play at shouting volume for 2-3 hours 4-5 X a week, and the very occasional gig at a respectable yet loud volume.

There really is no way to be more specific than that.

 

[sdn25]

Variables would include how often you play, at what volume, length of time, unforseen events like vibrations when transporting, etc....

In most cases they should last years.

I have the same tubes in my Rivera Jake Studio combo. Had them installed back in January. Zero issues. I like them.

I play at shouting volume for 2-3 hours 4-5 X a week, and the very occasional gig at a respectable yet loud volume.

There really is no way to be more specific than that.

For volume, as I mentioned, I have all the knobs on 10 running into a reactive load, so the tubes are under full load, I play for at least 3-4 hours a day, it lives on my table so no movement, since it’s into a reactive load there’s no vibration from speaker cabinets

 

[StrummerJoe]

The hotter you run it the shorter the life of the tube(s). You're really working them hard. Bet it sounds cool.

 

[sdn25]

The hotter you run it the shorter the life of the tube(s). You're really working them hard. Bet it sounds cool.

Yeah it does sound great, so I could expect like at least a year ? Or maybe 2?

 

[StrummerJoe]

Yeah it does sound great, so I could expect like at least a year ? Or maybe 2?

Who knows? There is no way to make that prediction.

I wouldn't worry about it, I would just have extras on hand and the number of a good tech. If there are no techs in your area I would suggest learning how to bias it yourself.

A spare amp is always nice too.

Good luck!

 

[sdn25]

Who knows? There is no way to make that prediction.

I wouldn't worry about it, I would just have extras on hand and the number of a good tech. If there are no techs in your area I would suggest learning how to bias it yourself.

A spare amp is always nice too.

Good luck!

I built the amp so I’m comfortable with biasing it but I’ll consider getting spares, tubes are quite expensive nowadays unfortunately.

 

[paul-e-mann]

How long can I expect a set of current production Mullard EL34s (made by new sensor) to last in a cranked (all knobs on 10) 50w jmp 1987? I have it biased at about 50-53%.

I think thats a fair question to ask the tube manufacturer, they should have that data.

 

[StrummerJoe]

I built the amp so I’m comfortable with biasing it but I’ll consider getting spares, tubes are quite expensive nowadays unfortunately.

Good deals can still be found on Reverb now that the hysteria has died down a little.

 

[StingRay85]

Will all one 10 in a load box, I wouldn't bet on them making it to 2023

 

[Jon Snell]

I have a Selmer T&B 50 in for servicing. Gigged most weekends from new and looks untouched inside.
The Mullard EL34s date from 1970 and still read 96% emmission per valve. Main smoothing are fine, due to regular use as are all of the components as far as I can tell. No need to change those yet then.
It had a noisy ECC83.
All fixed and on its way back to the customer some 250 miles away from my workshop.

 

[sdn25]

I have a Selmer T&B 50 in for servicing. Gigged most weekends from new and looks untouched inside.
The Mullard EL34s date from 1970 and still read 96% emmission per valve. Main smoothing are fine, due to regular use as are all of the components as far as I can tell. No need to change those yet then.
It had a noisy ECC83.
All fixed and on its way back to the customer some 250 miles away from my workshop.

Old mullards are tanks, I have the new ones idk if they are better Than the competition though. I guess time will tell eh?

 

[playloud]

It's a pity there aren't some basic lifetime statistics for these. These would be my assumptions (happy for others with more experience to correct):

- There's an initial "infant mortality" phase, where the tube has an elevated probability of failure (the extent of which varies widely by manufacturer/batch). After that, you're probably largely in the clear for the next few hundred hours of use. A good "burn in period" should hope to catch the unlucky infants.

- A single EL34 probably has a mean lifetime in the low 1000s of hours* (driving a speaker in a guitar amp). Obviously this will again vary widely by manufacturer/batch (generally speaking, I'm guessing old tubes > current production ones here, although I'm sure there's a certain amount of survivorship bias attached to the old tubes still circulating today). I assume this also varies to some extent by circuit/typical amp settings.

- The more tubes you have, the sooner one of them will go wrong. If we had nice empirical probability distributions for the lifetimes, we could estimate this effect outright, but imagine something like the following: if you have 4 tubes with an expected lifetime of 2000 hours individually, the first one will likely go bad at the 1500 hour mark.

Now I'm just curious. Assuming you have a quad of current production (e.g. New Sensors) EL34s, and they've survived the first 10 hours, do we think they'll make it to 1000 hours, cranked? If so, that'd give you almost 18 months at 2 hrs/day, which doesn't seem too bad to me.

* Based purely on hearsay, but using 10000 hours as a hard upper bound because of those "10M" ECC83s that were marketed as exceptionally durable.

 

[sdn25]

It's a pity there aren't some basic lifetime statistics for these. These would be my assumptions (happy for others with more experience to correct):

- There's an initial "infant mortality" phase, where the tube has an elevated probability of failure (the extent of which varies widely by manufacturer/batch). After that, you're probably largely in the clear for the next few hundred hours of use. A good "burn in period" should hope to catch the unlucky infants.

- A single EL34 probably has a mean lifetime in the low 1000s of hours* (driving a speaker in a guitar amp). Obviously this will again vary widely by manufacturer/batch (generally speaking, I'm guessing old tubes > current production ones here, although I'm sure there's a certain amount of survivorship bias attached to the old tubes still circulating today). I assume this also varies to some extent by circuit/typical amp settings.

- The more tubes you have, the sooner one of them will go wrong. If we had nice empirical probability distributions for the lifetimes, we could estimate this effect outright, but imagine something like the following: if you have 4 tubes with an expected lifetime of 2000 hours individually, the first one will likely go bad at the 1500 hour mark.

Now I'm just curious. Assuming you have a quad of current production (e.g. New Sensors) EL34s, and they've survived the first 10 hours, do we think they'll make it to 1000 hours, cranked? If so, that'd give you almost 18 months at 2 hrs/day, which doesn't seem too bad to me.

* Based purely on hearsay, but using 10000 hours as a hard upper bound because of those "10M" ECC83s that were marketed as exceptionally durable.

I have a 50w jmp so I have a duett. Would that mean they would theoretically last longer than your estimate?

 

[playloud]

I have a 50w jmp so I have a duett. Would that mean they would theoretically last longer than your estimate?

Yep (a pair should last longer than a quad, on average), but my estimate is way too imprecise for this to be material here.

 

[Kutt]

I don't think anyone can say how long they will last, but for me personally I'd start getting nervous around the six month mark if you're playing with everything on 10 for upwards of four hours per day.

 

[sdn25]

I don't think anyone can say how long they will last, but for me personally I'd start getting nervous around the six month mark if you're playing with everything on 10 for upwards of four hours per day.

Is anode/cathode current draw the way I can measure when the tubes are at full load?

 

[Pete Farrington]

It’s anode dissipation, rather than current per se, that sets the consumption rate. But with an inductive load, that’s pretty hard to assess accurately, as V and A aren’t locked in phase, W won’t equal VxA.

This thread goes into assessing anode dissipation under load; it gets quite convoluted! https://www.marshallforum.com/threads/hiwatt-dr504-build.121746/

Operational life will depend on what your end point is. GEC note output power dropping below 50% or gm dropping below 80% of initial value.
Using that, KT66 are at least 8000 hours when operated at their abs max of 30W. The abs max of EL34 would probably be somewhat higher than that.
You may notice some degree of degradation to the sound before the above points are reached.

https://tubedata.altanatubes.com.br/sheets/084/k/KT66_GEC.pdf