Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: Prev 1 [2] 3 4 Next   Go Down
  Print  
Author Topic: Chinese tube manufacturing analysis  (Read 22447 times)
W8JX
Member

Posts: 13268




Ignore
« Reply #15 on: September 18, 2011, 09:17:44 PM »

As far a China goes, is anyone surprised?
All they do is try to copy and then change the process to cost as little as possible and make junk to export.

Labor is cheap there so no need to cheapen process. They just have QA issues. I have some 16 year old Chinese tubes in my linear and they still work fine. USA priced itself out of tube business on a lot of tube types. Were it not for China there would be a lot less tube amps in use today and many in a early grave.
Logged

--------------------------------------
Ham since 1969....  Old School 20wpm REAL Extra Class..
KM3F
Member

Posts: 908




Ignore
« Reply #16 on: September 18, 2011, 09:46:30 PM »

Yes your right on that Tom.
.
A recent experience in the automotive field is their PRC coils made for use on Fords coil over plug motors.
Purchased 10 units because the price was low and lost every one of them within less than one year.
They cost me a lot of money by damageing the solid state switched in the PCM so it was a hard lesson to learn having to have a dealer replace the PCM, programming and loss of a cat assembly for one bank due to melting from raw gas being burnt in them caused by dead cylinders..
The coils suffered from shorted turns and open windings.
The FORD oem coils are the standard and even they will fail after some amount of time years and mileage.
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #17 on: September 19, 2011, 06:27:44 AM »

As far a China goes, is anyone surprised?
All they do is try to copy and then change the process to cost as little as possible and make junk to export.

Labor is cheap there so no need to cheapen process. They just have QA issues. I have some 16 year old Chinese tubes in my linear and they still work fine. USA priced itself out of tube business on a lot of tube types. Were it not for China there would be a lot less tube amps in use today and many in a early grave.

Wrong on both counts.

The more human element is put into a process, the less reliable it is. Especially when it is a parallel build and multiple people build identical components. When good manufacturers build tubes everything is fixtured and one person does all of one process in a continuous build. This way they are all the same, and they either pass or fail as a group.

The worse way to build is multiple people doing the same job, not using fixtures, and then switching jobs. 

To understand reliability we have to look at dozens or hundreds of tubes from multiple batches. You have no idea what quality or reliability is really like based on a sample of two tubes from one batch.
Logged
G3RZP
Member

Posts: 1224




Ignore
« Reply #18 on: September 19, 2011, 06:36:58 AM »

>You have no idea what quality or reliability is really like based on a sample of two tubes from one batch.<

Depends on batch size. For space grade ICs, we were allowed no failures any batch of quantity under 10,000. Since we never did that many S grade devices in a year, we had to have no failures or the batch was downgraded to mil grade or to civilian grade depending.
That included at wafer level, too. Although at that time, we were still using 4 inch wafers for a lot of product.
Logged
W8JX
Member

Posts: 13268




Ignore
« Reply #19 on: September 19, 2011, 07:11:22 AM »

As far a China goes, is anyone surprised?
All they do is try to copy and then change the process to cost as little as possible and make junk to export.

Labor is cheap there so no need to cheapen process. They just have QA issues. I have some 16 year old Chinese tubes in my linear and they still work fine. USA priced itself out of tube business on a lot of tube types. Were it not for China there would be a lot less tube amps in use today and many in a early grave.

Wrong on both counts.

The more human element is put into a process, the less reliable it is. Especially when it is a parallel build and multiple people build identical components. When good manufacturers build tubes everything is fixtured and one person does all of one process in a continuous build. This way they are all the same, and they either pass or fail as a group.

The worse way to build is multiple people doing the same job, not using fixtures, and then switching jobs. 

To understand reliability we have to look at dozens or hundreds of tubes from multiple batches. You have no idea what quality or reliability is really like based on a sample of two tubes from one batch.


Tom you are wrong here. Labor cost is a big factor and while humans add a variable element the less they cost the more time the have to spend on it. Without doubt there is room for improvement and build process. Tubes like 811's 572's, and such are very low tech and do not require automation to build. Also if labor was not cheap many manufactures would not have moved suppliers to china. China can make some good stuff. To look at the quality of 811 tubes in not a measure of what China can build well.
Logged

--------------------------------------
Ham since 1969....  Old School 20wpm REAL Extra Class..
K8AXW
Member

Posts: 7036




Ignore
« Reply #20 on: September 19, 2011, 08:09:55 AM »


[/quote]

Tom you are wrong here. Labor cost is a big factor and while humans add a variable element the less they cost the more time the have to spend on it. Without doubt there is room for improvement and build process. Tubes like 811's 572's, and such are very low tech and do not require automation to build. Also if labor was not cheap many manufactures would not have moved suppliers to china. China can make some good stuff. To look at the quality of 811 tubes in not a measure of what China can build well.
[/quote]

I concure.  The Chinese are at the same place now where the Japanese were when they first started to export to the US.  I still remember the term, "Japanese junk!"  Now that phrase isn't heard anymore because after about 10 years of getting feedback on their "junk" the Japanese government jerked their industry up by their hair and told them that all exports will meet acceptable quality standards or it wouldn't ship.


The Chinese haven't reached this point yet.  No doubt the multiple tube assemblers is a factor but the manufacturers are presently responsible for what goes out and we will continue to import "Chinese Junk" until they adopt the same philosophy as the Japanese.

I'm sure the Chinese aeronautical and space programs don't use the same crap they export to us!  Which means they CAN make good stuff!
Logged

A Pessimist is Never Disappointed!
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #21 on: September 19, 2011, 10:42:39 AM »

Sorry, but I bet neither of you has seen what it takes to build tubes.
Not using fixtures and building in parallel is not a good method at all for reliability.

A sample of two tubes means nothing when tubes are built randomly (not using proper fixturing and have a variety of people build a single component).

You get two out of a batch of 100, with each grid (or other element) built by a different person a different way. With 10 or more things to vary at every stage, it is impossible to predict reliability overall in a batch, let alone over many batches.

When Eimac had problems, and they did on occasion, one or two tubes bad would give a good idea of the entire run. That's because every component inside the tube in that batch was likely to be identical. For example we received a batch of 3CX1200A7's that had very low emission life. Every tube in that batch had the same problem.

With the Chinese method, it is impossible to get an idea how other tubes from the same batch will work.

G3RZP pointed out wafers and testing IC's. Every IC from a wafer is processed the same way. If a wafer was cut up in pieces, and 10 different parallel processes were used all without fixturing and rigid controls, it wouldn't mean anything if two worked correct except two worked correctly.

73 Tom
« Last Edit: September 19, 2011, 11:26:46 AM by W8JI » Logged
KX5JT
Member

Posts: 217




Ignore
« Reply #22 on: September 19, 2011, 12:10:05 PM »

As a lover of vacuum tubes, this is all very disconcerting.  It seems we have to all live with this unless some company steps up to the plate and does things more old school... then we wouldn't want to pay for the end result which would likely be way more than hams would care to spend.  What's the solution?  Mexican tubes?
Logged
N2EY
Member

Posts: 5063




Ignore
« Reply #23 on: September 19, 2011, 02:20:55 PM »

The more human element is put into a process, the less reliable it is. Especially when it is a parallel build and multiple people build identical components. When good manufacturers build tubes everything is fixtured and one person does all of one process in a continuous build. This way they are all the same, and they either pass or fail as a group.

The worse way to build is multiple people doing the same job, not using fixtures, and then switching jobs. 

To understand reliability we have to look at dozens or hundreds of tubes from multiple batches. You have no idea what quality or reliability is really like based on a sample of two tubes from one batch.

I agree with what W8JI is saying. I would say it somewhat differently, though.

Here's my version:

The way to build good tubes is to design the manufacturing process for maximum uniformity of parts and product. This means eliminating every possible source of non-uniformity.

In practice that means precision jigs and fixtures wherever possible. It means automation whenever possible. It means having the same people, or preferably the same person, making sure it is done the right way every time.

Just because 3-500Zs have been made for almost 50 years doesn't mean the process is simple or easy to do right. Old technology isn't always simple. Ignoring what the Ancient Ones knew and did often results in a rediscovery of why they did things the way they did.

---

The Big Problem with transmitting tube manufacture in the USA is that the market is too small.

Suppose that tomorrow I won the lottery and decided to set up a 3-500Z factory. And suppose Eimac decided to just give me all the engineering data on how to build 3-500Zs, free.

And suppose I was able to set up a plant, with all the necessary tooling and equipment, to make top-quality 3-500Zs.

How many could I hope to sell in a year? How many could I hope to sell 5, 10, 20 years from now?

73 de Jim, N2EY

Logged
W8FU
Member

Posts: 6




Ignore
« Reply #24 on: September 19, 2011, 02:43:58 PM »

Perhaps the USA market is small, but worldwide the market is large enough to interest the Chinese.  I suspect the problem is not the size of the market but the fact that it costs more to build these things in the USA. If tubes can be built here of higher-quality than in China, one would think that should be enough to justify a higher price. Unfortunately, too many customers buy on price alone with quality taking a backseat. Eventually it becomes rather academic because if there is only one source for product you are forced to take what you can get.
Logged
N2EY
Member

Posts: 5063




Ignore
« Reply #25 on: September 19, 2011, 03:02:54 PM »

Perhaps the USA market is small, but worldwide the market is large enough to interest the Chinese.

Maybe. How big is the world market for 3-500Zs? Does anybody use them besides hams?

I suspect the problem is not the size of the market but the fact that it costs more to build these things in the USA.

How much more?

For years we've been told that it costs too much to make things in the US, that American labor costs too much, etc., etc.,etc. But nobody says exactly how much more it would cost.

And we wind up paying anyway, not just in lost jobs and revenue, but in replacement cost.

If tubes can be built here of higher-quality than in China, one would think that should be enough to justify a higher price. Unfortunately, too many customers buy on price alone with quality taking a backseat. Eventually it becomes rather academic because if there is only one source for product you are forced to take what you can get.

All true, but there's a lot more. For example, many industries get their economies from scale, which forces the small folks out of the game.

There's also price expectation. I'm going to look up what a 3-500Z cost back-when and see what that equates to in 2011 dollars. I suspect it's a pretty surprising number.

73 de Jim, N2EY
Logged
KX5JT
Member

Posts: 217




Ignore
« Reply #26 on: September 19, 2011, 03:06:22 PM »



Maybe. How big is the world market for 3-500Zs? Does anybody use them besides hams?

I recall reading some 1 KW AM BC Transmitter used them in pulse modulation mode but I would doubt very many of those are still in service.  I would believe the US Market IS very significant to the Chinese tube manufacturers.  Besides Japan, we still have a very large ham population relative to the rest of the world.
Logged
G3RZP
Member

Posts: 1224




Ignore
« Reply #27 on: September 19, 2011, 03:19:30 PM »

Tom,

we went the other way round. If 1 in a hundred failed, the whole lot was failed. It could be that 99 out of a hundred passed, but the requirements didn't allow that. 1 failure was a lot failure.....usually, a wafer failure.

If 2 different wafers in a batch of 40 wafers had a failure, it meant the whole batch of 40 was condemned for space grade use.

Which is why 50 pieces of whatever it was, if S grade, would cost $100,000 in the mid 1980's.
Logged
N2EY
Member

Posts: 5063




Ignore
« Reply #28 on: September 19, 2011, 03:35:42 PM »

I would believe the US Market IS very significant to the Chinese tube manufacturers.  Besides Japan, we still have a very large ham population relative to the rest of the world.

Let's do the math!

There are about 700,000 US hams, of whom about half (350,000) have General, Advanced or Extra licenses which permit high power on HF. (3-500Z isn't a good choice for VHF/UHF).

Let's assume that, of the 350,000 US hams who could run high power on HF, 10% actually do. That's 35,000.

Of those, assume that 20% use 3-500Z tubes in their amplifiers. That's 7,000.

And assume they each buy one new 3-500Z every 5 years. That's 1,400 a year.

Not much of a market. And as the quality 3-500Z are sold, the demand drops!

btw, I looked up the prices back-when. According to my 1964 Newark catalog, a 3-400Z cost $34 ithat year, which equates to about $240 today. A 3-1000Z was $119, which works out to $840 today.

73 de Jim, N2EY


Logged
KX5JT
Member

Posts: 217




Ignore
« Reply #29 on: September 19, 2011, 03:42:46 PM »

I would believe the US Market IS very significant to the Chinese tube manufacturers.  Besides Japan, we still have a very large ham population relative to the rest of the world.

Let's do the math!

There are about 700,000 US hams, of whom about half (350,000) have General, Advanced or Extra licenses which permit high power on HF. (3-500Z isn't a good choice for VHF/UHF).

Let's assume that, of the 350,000 US hams who could run high power on HF, 10% actually do. That's 35,000.

Of those, assume that 20% use 3-500Z tubes in their amplifiers. That's 7,000.

And assume they each buy one new 3-500Z every 5 years. That's 1,400 a year.

Not much of a market. And as the quality 3-500Z are sold, the demand drops!

btw, I looked up the prices back-when. According to my 1964 Newark catalog, a 3-400Z cost $34 ithat year, which equates to about $240 today. A 3-1000Z was $119, which works out to $840 today.

73 de Jim, N2EY




Amazing.  It's a wonder that any are still being made anywhere then.
Logged
Pages: Prev 1 [2] 3 4 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!