Vacuum Tube Life Expectancy

Here's an interesting "vacuum tube life" story (true story). When my dad passed away, one of things I inherited was his late 1960s Heathkit stuff, radios, amps, and other electronic items he built using kits - all of what he built used tubes. I remember going to the store with our television tubes and testing them to see which ones needed replaced, I don't ever recall going to the store for his radio euipment tubes but I'm sure he must have at some point. Now for the interesting part...I fire up the Heathkit 14W amplifier about two ro three times a year, and dad's old crystal receiver, and a small pair of speakers he had on the system, and before applying the tunes I've always let the tubes warm up. They still work!!! It's a four tube amp, and all four tubes are still functioning just fine. The tubes in his old radio still work too. I keep waiting for the day that I have to get on eBay and buy some tubes for replacments, but they won't give up! I think that is really interesting that after 40 some years the vacuum tubes are still working fine. I know very little about tubes, but maybe the life span has to do not only with warming them up but maybe how much they are used?

The amount of power which translates to HV and current is what shortens tube life. Also how much usage 5000 hrs for Russian compared to 20,000 hrs Telefunken. If the output tubes are not biased this is going to shorten life. The 14 W amp is probably cathode biased which means no bias adjustment and it is not pushed hard such as 100 W amp in a hard metal group. Turning on HV before tubes warm up is another example. My reference for this information is the book "Tube Guitar Amplifier Essentials by Gerald Weber" an expert Tech. in Tube amps and Guitars. Some of the damage I have talked about is very gradual some like lack of bias and running hot (too much current) is pretty quick. If you have an ear for music you can tell when you replace old or damaged tubes the difference in sound. You can even tell the difference between manufactures! All tubes are different and the output tubes need to be matched sets and biased correctly. And as someone said tubes on eBay can cost hundreds of dollars for certain types and are not made any more.

Vacuum tubes are a lot like light bulbs. Minimize the thermal shock of on/off, run them at lower power, run them only a few times a year - all of these things help.

I remember reading that in the old vacuum tube computers with their 1000's of tubes, they would start off each day with a run at 120% of plate voltage and 80% of plate voltage - this would weed out all the tubes that were about to fail, so they could get through the rest of the day without a tube failure.

I Love tubes. The sound is sooo rich when compared to solid state devices. I just picked up a whole box of Misc Tubes (when I tested them using my Antique Tube tester, 95% were still in working order).

Also Recently I picked up a Harman Kardon 3 channel PA system, (Model DA-100, Commander series) It has 3 preamp tubes, and 4 6L6 tubes, for the power section. This 100W will blow the pants off a 1000 Watt transistor amp. It's just a different kind of power. The interesting thing about the Harman Kardon piece is I cannot find ANY (short of 1 single site, who claims to have the schematic, although I'm skeptical, even though they only want 10 bucks for it) information anywhere on the internet on this model. I'm beginning to think it was a military device, or something perhaps not sold as a consumer device. With the case removed, it is beautiful, and built like a rock.

In fact, I (If I ever get enough spare time to finish it up) will be posting a multiple part blog series on tubes, testers, and amps soon. Full with step by step pictures on how to perform proper amp bias ( the other common reason for short tube life), Troubleshooting, and building a handmade custom amp for use as either a guitar head or vocal amp (depending on setup).

It's amazing how expensive tube amps are, you can spend $500-$1000 easy on a basic hand built tube amp.

I would love to find out what this Harman Kardon 3 channel 100W quad 6L6 PA head is worth, (especially because I picked it up for free from a craigslist ad, along with ~50 spare tubes, and a 1947 Vacuum tube tester...all for free). I have had to do troubleshooting on the PA head, and found some easy fixes, such as replacing a bad pot, and a couple of Caps... but it sounds AMAZING!

I am building the famous Fender 57 5E3 Tweed amp that you can see on my logo picture. It is about 12-15 W. One of the first amps built. I have heard this amp which can cost at music stores $2,700 on my space and it sounds good. I am probably building it for about $500 Boutique style with many improvements such as Automatic Standby which I am keeping a trade secret in my Electronics company, overdrive, gain, bright, line out, DC 12.6 for all preamp tubes but it is still point to point. No PC boards, metal film, transistor or op amps.

I do not see my post but my college Prof used to say "Transistors die silently but Vacuum tubes die violently". Another way to kill a tube which I did like in 7th grade was to drop my amp and I broke a tube.

Since I retired I've repaired several tube amplifiers. It seems there are some "secrets" of the long tube life:

• operate them inside the factory limits, about at 75-80 %
• Tube amp with tube rectifier - or with HV turn-on delay
• protect them from mechanical shocks

Proper biasing is essential: you have to measure the cathode current despite the grid voltage. For smaller amps auto-biasing is a good solution.

Russian tube are "bullet-proof". I like them because those are less expensive and their expectable lifespan is not shorter (or sometime longer) than the lifespan of the similar other tubes. The reason is very simple: Russians used the same tubes in military applications.

A nice Hammond HR-40...

Yes, overdriving output tubes will definitely shorten tube life for obvious reasons but you are propagating the same old tired myth about cathode stripping. It has absolutely nothing to do with a standby switch, there is no documentation that shows a standby switch lengthens tube life, only anecdotal claims. Cathode stripping, under NORMAL conditions does not happen, period. What is shortening the tube life is excessive screen grid dissipation and excessive tube current caused by overdriving it. It is a normal practice for guitar players to overdrive and that will shorten tube life, guaranteed. The preamp tubes last a long time because they are low current and are protected by much higher component values, overdriving the input merely causes clipping of the signal (another characteristic guitar players like) but the signal levels are still low enough to not cause excessive current.

You are also correct in that a poorly designed amplifier is going to eat tubes. This is only blaming the tubes for either poorly designed or abusive use, not the tube's fault. The real purpose of the standby switch, which guitar amp manufacturers installed, was because performers wanted to go on stage, flip a switch and start playing. This did minimise power dissipation in the output tubes during this period which could run into quite a long time. This did have the side effect of lengthening tube life some but it was not the real purpose of the switch nor does this switch prevent imaginary cathode stripping. You are quite wrong about tubes not liking high voltage being applied while cold, nothing can be farther from the truth, just more smoke and mirrors.

Before anyone starts going to extremes with claims otherwise, I have documentation from the tube manufacturers themselves which proves these claims are quite false. Extensive testing was done on large quantities of tubes for the military by the tube manufacturers to determine life expectancy and it disproves the myths that have been circulating. In short, it is the misapplication and misuse of tubes which causes failure not these silly myths. There is also data in tube data sheets which disproves the myth of applying voltage to a cold tube and causing damage. Just what do you think is happening when you turn on the power switch and the high voltage is applied to that cold tube rectifier? If this myth were true, tube rectifiers would indeed have short life spans.

If you want short tube life, then overdrive them hard, if you want long tube life, operate them within the manufacturer's specs. This is not as easy as you may think it is to do and many unintended design flaws cause tube failure.

By the way, very heavy filtering of the high voltage power supply can cause unintentional tube damage, including cathode stripping, both at turn-on but especially at turn-off.

It is OK to disagree with what I say based on the Book by Gerald Weber "Tube Guitar Amplifier Essentials" an expert on tube amps and guitars but I am pretty much quoting him on tube damage that can occur. Maybe you could read that book and email him about the smoke and mirrors?

Another reader of this thread pointed something out which I will immediately correct. I inadvertently left out a very important sentence which I usually put in these discussions about tube life and cathode stripping. Here it is:

All of my comments and reference documentation refer to the specific condition of NORMAL OPERATION within manufacturer's specifications, it is certainly possible to cause cathode stripping under certain conditions, nearly all of them pertain to a malfunction (of the tube or circuitry) or use above and beyond manufacturer's ratings. I will also add that in a small percentage of tubes, manufacturing defects can cause a malfunction anywhere from immediately to many operating hours later, this is NOT an normal operating condition.

My experience and research through many old, hard to find references have backed up my statements. I do not give any weight to anecdotal material, only solid, repeatable measurements and experiments. Yes, I have personally seen tubes that have suffered cathode damage, from the minor to the major and in virtually every case it was caused by a malfunction or design flaw. It can be very difficult to determine the source of damage in some cases and in many cases, presumption prevailed.

I know that the current round of published 'opinion' claims otherwise but they have not produced any substantial proof of their claims, nearly all of it is anecdotal or guessing. I've also heard the "better safe than sorry" nonsense, sorry but I don't buy it! Good solid, engineering does the job best, not Band-aids and non-sense.

My apologies for leaving out an important piece of information.

Regards.

To Maveric Manic:

Sorry if it sounded like I was jumping on your back, that was not my intent. I realize that most people must rely on what they find out there and unfortunately, a lot of what is out there is unfounded tripe and assumption. That is one of the reasons I sometimes jump in and try to 'correct' things but I'm often met with responses similar to yours which is a 'reasonable' response given what you have read.

I am trying to get a paper published on the subject but I have run into surprising resistance for some reason, perhaps because it goes against the current published information and these publishers are very often the ones who are publishing the misinformation.

I can relate one item to you, at one time I also believed that the maximum voltage that could be applied to a tube was that published under the plate voltage specifications but I also had a feeling that that was not correct. I found out in a publication from Philips in the '60s that my suspicions were correct, there is a specification that specifies the maximum amount of voltage that can be applied to a cold (off) tube and it is considerably higher than the operating plate voltage spec. Unfortunately, this spec has not been widely published, particularly here in the U.S. but, none the less, applies to all receiving tubes. I refer to several examples of this specification in my paper. I also give a conservative estimation figure to use for tubes which do not have this spec published.

Also, do not forget, that like all electronic parts, tubes have what is called a premature failure rate, just like all solid-state devices. A certain percentage will have some kind of random failure during the initial few hundred hours of operation, EVEN if the tube is operating within normal specs and also like solid-state devices, there will be an increasing random failure rate at the end-of-life. This is the familiar bell curve. There is absolutely nothing that one can do to prevent these failures, they are going to happen no matter what you do.

For instance, in the case of a filament failure, let's say you are operating the filaments on DC and ramping the filament voltage up, conventional theory says that will prevent failure, not true. It may prolong the eventual failure by some small degree over just turning it on cold but it is going to fail prematurely. I have had very few tube filaments fail and the leading cause by far, was due to mechanical shock not heating and cooling.

I have heard many times the excuse for these band-aid proposals, "better safe than sorry." This is nonsense, if the tube is going to fail prematurely nothing you do will prevent it, if your circuit design is flawed in one way or another, it will fail. I have not seen any concrete data which shows these band-aids provide any advantage over a good design, while most of them are harmless, they end up costing wasted money. I can tell you that designing a good, stable amplifier is not easy, there are many alligators in the swamp that can bite you without you even knowing that they are there. Screen grids are notorious for real time signal problems, the analysis of the AC signals on that grid can be quite difficult and a miss-step here can cost you a tube. The fact that you may have the correct DC voltage/current on the screen at idle does not mean your tube is going to be running correctly with a signal.

It is unfortunate that the cost of many tubes today have been run up to obscene values but that is the so-called free market, tube dealers are loving it! Not that much of it pertains to availability either, there are some buyers out there who can afford to pay ridiculous prices and they spoil it for the rest of us. I was just told that some particular brands of 6SN7GTs have been selling for $475 a pair, if that isn't insane, I don't know what is!