So Why are NOS Tubes Superior and Why.......

[Brewmaster]

So why are NOS tubes superior and why can't they be accurately duplicated?

There are several new production tubes I am happy with but they don't compare to old manufactured tubes. Is it because there was superior tooling, vacuum, overall better manufacturing techniques, or maybe all of the above. Many of the reissues tubes fall short of those tubes they are trying to duplicate. Some are good tubes in their own right but not true duplicates much in the same way many reissue amps are not true reissues.

So why are they superior?

[tarzanalog]

So why are they superior?

Here's you're answer in a nutshell:

http://www.epa.gov/history/timeline/index.htm

FWIW, I consider myself an environmentalist... but vacuum tube manufacturing is the one area that I wish was exempt from regulation.

[Brewmaster]

That is a major part of the reason why manufacturing in this country is on a decline as a whole and much manufacturing has moved overseas.

Why can't the overseas plants tubes match the quality of what came before?
Saratov had plants making tubes well before New Sensor took them over. One of the things that struck me was the NOS Saratov EL-84's I got recently. These are way better than any current tubes made by Sovtek in the same place. Also, isn't the Tesla plant the same one that makes "JJ's now?

Just asking.

[tubetek]

Why do you think they (AssPitt/GrooveTubes/foreign investors) paid so much to buy the old GE tube manufacturing stuff from here in KY. several years ago? (And promptly shipped it overseas) They got a supply of the actual plate material (now used up) that GE used prior to shutting down in the 70s!!! As I understand it, there was an explosion involved in the process, not to mention the salts, metals etc. that the EPA would have a fit over today. The cost of manufacture for such a niche market explains it all. AND, I don't believe some of the materials can be made now at any reasonable cost. Don' I wish it was different!
tubetek

[blaren]

No....TESLA roznovs were made in Czek and JJs are made in Slovakia.
There are a TON of reasons...some urban legend, some true. Some people will tell you that tubes that were manufactured in any former iron curtain country were inferior. They had poor QC and a high failure rate. Whether that's true or not, I guess it's a "relative" statement....maybe inferior to western tubes available at the time but, evidently, not inferior to NP tubes.
Not only are there the environmental problems of aquiring materials and disposing of waste safely and economically....and the added expense that would be incurred if the vacuums were what they used to be....but....there's also this...
Don't forget that back in the day this market was HUGE. There was peobably more competition. Don't forget, tubes were IT. Imagine a SMALL airborne radar system in a little Canadian AirForce plane. It would have over 5000 tubes running the radar system. They didnt want them to fail. In cold war days, tube technology was a highly guarded military secret. I've heard it said that even if you bought all the tooling from the Blackburn plant (which someone did...I cant remember who), you wouldnt get the "recipes", which, were destroyed. Kinda like the Mayan language. There are still some old timers who worked there and know some stuff but the recipes/books are GONE!

I thought there WERE a couple of American outfits still making tubes.? I'd imagine that a PROPERLY made current production tube would probably retail for more than a nice NOS one. Heck.....I was paying $45ea before SOVTEK and PM came-out in the '80s.

[briane]

So why are NOS tubes superior and why can't they be accurately duplicated?

yep, mainly its barium...though there's other metals used in the originals that also cause horrible diseases in the workers and those living nearby plants...

I have found lots of good new stock tubes...not quite the same majic or mojo, but still solid and very good sounding. And also I generally consider, for a well made tube, the newer tubes to be more dependedable, just cause I have seen some strange burnouts in NOS tubes occasionally.

[blaren]

wow...I would never say that any np tube is more dependable than a NOS one. I mean....a tube can fail for a multitude of reasons but I would trust ANY old tube (except maybe an Ei) over ANY new production tube in any of my amps any day of the week. It's like a "hands-down" thing. I've seen PLENTY of NP preamp tubes pack it in for no(?) reason. Well...for no reason other than maybe they were tired of sounding like CRAP and felt everyone would be better off if they died.
Not very often do I see an old amp (with 40 or 50yr old preamp tubes) come in thad had a bad old pre-tube. Amps with a "blown" new preamp tube are MUCH more common.

All that said...I've never tried the new TungSols though (12AX7) so I cant speak about them.

[pylet2000]

There are new production tubes that work quite nicely. I like the Tung Sol (I was way skeptical at first too, it proved to be a winner) and the JJ ECC803S (long plate version). They sound nearly as good as any NOS tubes, and I'm starting to notice a lot more burnout/microphonics/noise on NOS tubes anymore. Let's face it, they are getting older and glass is mildly gas permeable over time... A tube that 40 years old has probably been handled and rattled a lot more too. Reliability is starting to be shaky in both new and old tubes.

Outside of environmental issues, which certainly are a factor, I think the economics of the market are the main drivers for mediocre modern tube performance. You ever priced a new Western Electric 300B? Phenomonal sounding tube with incredible QC, but the prices reflect it to the point that the only market capable of supporting it is the HiFi realm (unless you want to design very expensive single ended triode guitar amps, might sound pretty sweet...) Other environmentally harmful industries have been able to come up with better and safer ways of doing things, but they usually had enough market demand to justify the R&D. The vacuum tube doesn't have much practical use in the transistor world, and as such doesn't get much research effort.

[jckid66]

imho,
back then people actually took pride in their work...

BIll

[drgonzo2]

There are many reasons for the decline in quality of current valve production. The most important (imo) are:-

1) Reduced marketplace. Pre the mid 60's, almost everything powered electrically had a valve in it somewhere. TV & radio are the best known uses, but there were many other items which used valves in daily use. One example is the telephone - admittedly, there were no valves in the handset, but at the exchange? Valve city. The market was HUGE. IIRC, there were 15 valve manufacturing plants in the UK alone. Worldwide, there were some 300+ valve manufacturing plants. Today, there are (to the best of my knowledge) 6 plants worldwide making regular production valves. (This figure does not include plants producing 'specialist' valves, such as Eimac & Western Electric.) Plant's like Mullard Blackburn produced several million valves a year - today the worldwide demand for valves is roughly 2-2.5million valves a year...

This reduction in the marketplace has meant that companies producing valves have had to compete for sales in a way that, prior to transistorisation, they had never had to do. The easiest way to do this, is to drop the price. This leads on to the next major issue...

2) Materials Quality. Valves in the 50's & 60's used new, 'virgin' metals. Companies like Mullard, RCA, Sylvania, et al, used tonnes of metals each year. The metals they used were smelted from ore to their own specifications. These 'virgin' metals were of a much higher quality than the metals used today, which are made using partially recycled materials. Why is this? Cost. Is is a lot cheaper to use partially recycled metal than it is to use new metal. It's massively cheaper to use a standard (partially recycled) alloy than it is to use a proprietary alloy of virgin metals...

All done to keep the cost down.

Which leads to the next problem:-

3) Production Quality. Valves used to be made to much higher production tolerances, for the simple reason that if they weren't, the user would go elsewhere. When there were multiple plants producing valves, each plant had to keep it's quality high, or soon word would spread that there were problems with that plant's production, and sales would dry up. Valves had to meet spec, or they wouldn't sell - hence quality was kept high. Unfortunately, keeping quality up costs money. QC inspectors cost money, skilled workers cost money, and precision tooling costs lots of money. While the competition was there, the money had to be spent, to stay competitive with your rivals. With the vastly reduced market nowadays, the competition really isn't there anymore. Quality control has been allowed to slide by the wayside, largely in the interests of reducing costs.

One of the biggies:-

4) The Military. The military forces were always one of the largest users of valves. They used millions of valves a year, and bought millions more 'just in case'. They had the buying power to force manufacturers to build to 'military spec', which cascaded down to the civilian production. Then the technology changed. This has lead to them having large stocks of valves which were surplus to requirements, so they sold them off. Cheap. So, not only were new valve manufacturers competing with each other on price, they were also competing with a massive supply of cheap old valves. So they dropped the prices as much as they could, just to get sales.

Which leads on to the final, and probably most significant problem:-

5) The Price of valves. I have a boxed NOS Mullard ECC83, dated 1971. Price £1.79. In 1971, my father bought his first house, a pretty standard 2-bed. Price £1800. In other words, a preamp valve cost 1/1000'th the price of an average house. Today, an average UK house is about £135,000($260,000). Apply the same price ratio, and a valve should cost about £135($260). And yet they don't. In fact, they cost less than 10% of that. Why? Because the consumer isn't prepared to pay that sort of money for a frivolity (and believe me, valve amplification in this day & age is considered a 'frivolous' pursuit) when IC's can do a similar job for so much less.

I'm sure if someone went to a current valve plant and asked them to build a run of valves to the original mullard specification, and told them that they were prepared to pay $100 PER VALVE that met the stated spec, you'd get a top quality product which would meet or exeed the toughest expectations. The problem comes when you ask them to make a valve for $2, so that you can sell it at $20. At such a price, the manufacturers can't afford the machinery, materials or training to make a top quality valve. So they make the best they can, using machinery that, in many cases, is up to 40years old. They keep production up by skimping on time consuming processes (like pumping down the vacuum), and send out batches of valves completely untested. Hence the rise of companies like GT, who do the job that would originally have been done by the factory QC dept.

All because the consumer want's to pay the least amount of money...

It's all rather depressing really, to realise that the decline of valve quality has largely been due to our unwillingness to pay for a quality product...

... G

[kleuck]

There are new production tubes that work quite nicely. I like the Tung Sol (I was way skeptical at first too, it proved to be a winner) and the JJ ECC803S (long plate version). They sound nearly as good as any NOS tubes, and I'm starting to notice a lot more burnout/microphonics/noise on NOS tubes anymore. (

I agree, ECC803S are awesome new tubes.
And my JJ EL84 sounds justs as good (and powerfull) as my NOS RFT.
Time will tell if the reliabilty is good.

[Keith2112]

5) The Price of valves. I have a boxed NOS Mullard ECC83, dated 1971. Price £1.79. In 1971, my father bought his first house, a pretty standard 2-bed. Price £1800. In other words, a preamp valve cost 1/1000'th the price of an average house. Today, an average UK house is about £135,000($260,000). Apply the same price ratio, and a valve should cost about £1350($2600).

Great point, but it appears as though you made an error in your arithmetic. If the tube would cost 1/1000 as much as the house, your price should be $260, not $2600. Still an awful lot to pay for a valve, but a little less painful.

I always think that using comparisons such as these is the only real way to look at how things have chnaged over time. People often bemoan how costly things are today when the reality is that things have really never been cheaper. Even gasoline is only slightly more expensive today in relative terms to what it was at its high point in the early 80s, at least here in the US. And things such as food and clothing have plummeted in real terms to what they were only a few decades ago.

Compare the price of a vintage Fender amp new in the 1950s to the median annual income with the price of a high end boutique amp today to the same median income and the difference is not as glaring as one might initially suspect.

[PeterS]

Imagine a SMALL airborne radar system in a little Canadian AirForce plane. It would have over 5000 tubes running the radar system.

I have heard many times the statement that old old vacuum tube computers had terrible problems due to tube reliability--- some calculation takes 12 hours to complete and one of the thousands of tubes always fails during that time. But I wonder if that is really true? I heard the number that a (original) B52 has over 2000 tubes, because all the controls and so on use servo motors powered by tube drivers. You'd think if tubes were so unreliable that would just not be practical.
I know they have ways of designing things to be robust and redundant so that if a tube fails it doesn't e.g. disable the landing gear, but still, you'd think if you were going to have any significant number of tube failures per year they would simply not have built the plane that way.

[WhiteSummer]

I read the same thing somewhere, but it later said that the tubes werent any good back when the compy was just starting out. They replaced the tubes and stuff with the newer high quality tubes and didnt have nearly as many problems. I could be wrong though. The better grade tubes are problably todays NOS tubes. Computers have been around since the 40s I believe, and it seems like a lot of the great NOS tubes come the 50s-70s. But I have no idea what Im talking about so dont trust me.

[drgonzo2]

Kieth2112, my bad. I've corrected that...

PeterS, as I understand the situation, the early valve computers tended to blow tubes whenever they were switched on. Once they were up & running, they were pretty reliable. As understand it, bugs in the program were a more common problem. By that I mean bugs (insects) physically getting into the switching relays & jamming them... The B52, otoh, is a bit of a special case. As a strategic bomber, it was designed to carry & deliver nuclear devices. When such devices go off, they release a shedload of electromagnetic radiation, which will destroy any unshielded transistorised electronics. This is considered 'not good' if those electronics are the flight control systems of the delivery aircraft. Valve based electronics can survive an EM pulse, which gives the flightcrew a chance to get home. Thus, the B52 used valve based avionics systems. The vast majority of early warning missile radar sites were (still are?) valve based for the same reason - it's a bit pointless having such systems if they go blind after the first bang...

Cheers... G

[PeterS]

When such devices go off, they release a shedload of electromagnetic radiation

I'd forgotten about that aspect. But it wasn't so much that I was thinking 'why didn't they use transistors' as that if the tubes weren't reliable they would have used something mechanical or hydraulic for flight control.

I think the 5881 tube was developed as a highly robust version of 6L6 for such applications.

Now back in the 80s I heard (may be total nonsense) that after the US stopped doing tube R&D, that the Russians continued, and developed tube 'integrated circuits' where you have a solid ceramic block with numerous small tubes within, for military applications.

[kleuck]

Tube technology is the only technology that can survive a nuclear explosion.
Understand : a tube-based radio would be the one and only way to communicate with the rest of the world after the bomb.
It is THE only reason, with the cold war, why the manufacturer tubes have not disappeared during the 70-80's.
The only countries that were still manufacturing tubes were : China USSR and USA....figure why ?

[Merlinb]

2) Materials Quality. Valves in the 50's & 60's used new, 'virgin' metals. Companies like Mullard, RCA, Sylvania, et al, used tonnes of metals each year. The metals they used were smelted from ore to their own specifications.

+1. Mullard smelted its own ore in house!! I didn't believe it until I saw this educational film made by them: http://www.youtube.com/watch?v=VV7cimRWah0
(At work we have to use very pure aluminium smelted specially for us in Italy, otherwise it won't anodise consistantly. We're always getting reject front panels due to "stringers" (stripy grain defects), so I now appreciate that all metal isn't made equal, even with todays production methods!)
That film was made quite a long time ago, before they introduced automatic assembly lines for some valves, which includes the ECC82/3. Nowadays I suspect they're all made by hand again, with the usual intollerances that involves.

5) The Price of valves. I have a boxed NOS Mullard ECC83, dated 1971. Price £1.79. In 1971, my father bought his first house, a pretty standard 2-bed. Price £1800. In other words, a preamp valve cost 1/1000'th the price of an average house. Today, an average UK house is about £135,000($260,000). Apply the same price ratio, and a valve should cost about £135($260). And yet they don't. ... G

Now let's not get carried away, house prices have increased disproportionately compared to other goods. A man could build his own house for less than 6 months wages in 1971!
Better to use something like the retail price index. £1.79 in 1971 comes out at £10.67 to 18.73, depending on who you ask http://www.bankofengland.co.uk/educatio ... index1.htm.

Which is about twice what you'd expect to pay for a new one.

[dotfret]

Just once in a while I can feel useful around here ...

Let us be clear about explosions and barium. This is associated with the getter. The valves, when assembled, have most of the air pumped out, but to function properly they need a harder vacuum than can be provided by pumping. The getter solves this problem. What most people call "the getter" is actually a support for a special tablet, which contains barium and nickel - the tablet is the real "getter". When most of the air has been pumped out, the valve is sealed - the pip on the top is the remains of that seal. The valve is placed in a strong magnetic field, which causes the getter tablet to explode, leaving a shiny deposit on the top of the envelope, and the getter support behind on the valve assembly. The explosion consumes the residual oxygen and nitrogen inside the valve. The strange metallic shiny stuff which results mops up and consumes any traces of air that may enter the valve via micro leaks.

The modern makers still use getters. You can see that shiny metallic stuff on the inside of the glass. That is not why old valves sound better than new ones.

The reason that old valves sound better is because old valves contain a radioactive metal called thorium in the mystic mud that the old makers used as a cathode coating. This is not a horribly dangerous radioactive material, so the tiny amount in your NOS valves is unlikely to harm you. But on an industrial scale, it requires much tighter control in this day and age, because the hazards are better recognised than in the past. As a consequence, new valves are made avoiding the use of thoriated cathode materials.

This is like a lot of industrial problems - someone will find a way around it eventually. Just like 20 years ago it was almost impossible to make a colored LCD screen, or an LCD display over 9" square. But for now, old valves still sound better than new ones ,,,

[dotfret]

To address some other comments above ...

One great reason for making large quantities of valves in the Soviet Bloc was, and still is, telephone exchanges. Aside from the military applications, valves were used in massive quantities for telephones. The most sought-after 12AX7 for hifi is the frame-grid Tesla E83CC, and the absolute best examples were made for the Czech telephone service - you can spot these because the gridposts are gold-plated.

Now, talking about making valves in the old-fashioned way, Russian exchanges were focused on using the 6n23p, which is very similar to the E88CC/ECC88/6DJ8?6922. I believe that the Voskhod factory is still making some of these - I have seen pictures of a batch with 2003 batch codes, and I own some which have 2004 batch codes. These valves appear to be just as good as, if not better than, any NOS valves from the same source.

Voskhod is a town where a lot of Russiam electronic components originate, and it is still a "closed community" - no foreigners are allowed in, and the inhabitants need a permit to leave town. They may be just knocking them out in the old-fashioned way - but IF they have a SAFE process allowing the use of thorium, they have a technical advantage over the rest of the world, and they are not exploiting that YET ...