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

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Ameritron cost reduction/quality improvement  (Read 8428 times)
WX7G
Member

Posts: 6329




Ignore
« on: October 08, 2012, 09:45:27 AM »

With the high failure rate of the Chinese tubes used in Ameritron amps, and one of the consequences being increased warranty costs for Ameritron, I think tube sorting should be looked into. While this looks like an added manufacturing cost the savings in warranty costs could offset it. The usual MFJ/Ameritron catchy trademarked term could be used to advertise the new process.

A sorting fixture and process flow something like this might do the job:

1) Hipot tube
2) Operate tube with DC on the plate that is ramped up to >2X the plate voltage used in the Ameritron amp the tube is destined for
3) Limit the peak arc current so the tube is not damaged and is actually conditioned
4) Monitor arcing
5) Ramp up plate dissipation to full rated
6) Monitor cathode current to detect a failing cathode

As with all industrial test fixtures it would be designed to survive fail when a tube failed. Automated control and data collection would cut labor test costs.
Logged
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #1 on: October 08, 2012, 10:21:27 AM »

I know the Internet opinion is no one does anything, but this has already been looked at extensively.  :-)

A sorting fixture and process flow something like this might do the job:

1) Hipot tube

already being done

Quote
2) Operate tube with DC on the plate that is ramped up to >2X the plate voltage used in the Ameritron amp the tube is destined for

Not done, and an unworkable test.

If a tube was operated at 2x dc voltage, peak voltage would be around 4x dc. This means an 811 tube would have around 6kV peak anode voltage under high current operating conditions. That isn't reasonable. Almost all perfectly good tubes would fail.

Plus excessive HV can strip metal oxide cathodes and might instantly ruin an otherwise good tube.

All tubes are operated at maximum expected voltage at RF, which results in peak voltages around 2x dc.


Quote
3) Limit the peak arc current so the tube is not damaged and is actually conditioned

Already done

Quote
4) Monitor arcing

Already done

Quote
5) Ramp up plate dissipation to full rated


Done in testing.

Quote
6) Monitor cathode current to detect a failing cathode

Since the emission loss takes dozens of hours, this just isn't practical. No one is going to cook a thousand or more tubes a month for a ten or twenty hours each. That would be 10,000 tube-hours a month minimum.

That would be 231 tubes a week average, more if batches are bad. With a 5 day week, more than 2300 tube-hours is 460 tubes at a time in parallel for 24 hours at 5 days if ten hours run time would ever show anything, which I doubt.

Emission loss generally seems to be very abrupt, not gradual. I base this on the fact there are almost no reports of gradual emission loss, it is almost always abruptly failed emission.

Quote
As with all industrial test fixtures it would be designed to survive fail when a tube failed. Automated control and data collection would cut labor test costs.


There isn't any way to test for most infant failures other than running the tube for a few dozen hours, and even doing that  many would not be caught.

The only practical solution is to make sure materials used in the tube are proper. That would be up to the manufacturer.

73 Tom
Logged
KD0REQ
Member

Posts: 1050




Ignore
« Reply #2 on: October 08, 2012, 10:56:30 AM »

tubes today are a crapshoot.  the only way to guarantee a win is not to play, go for solid state.

but then you have the usual creaky antenna connection opens/shorts issues that blow one or both banks of the output transistors, too.

probably just as many infant failures one way as there are the other.  a tube amp will usually take some short-term abuse, unless it's the tube that's the abuse.  SS amps, nope.

wonder if the power tube problems are worse than, or better than, the "good old days" of the late 20s, early 30s, when anything at any time was subject to a flash hard failure.
Logged
G3RZP
Member

Posts: 4965




Ignore
« Reply #3 on: October 08, 2012, 11:07:05 AM »

When English Electric started making 813s just after WW2, they had the sudden death emission problem. It took them several months to sort out and it turned out to be lower than required amounts of thorium in the filament. (In those days, measuring it was pretty difficult).

It maybe that the Chinese tube makers are looking into the failures, but it could also be that they have gone for low thorium to cut costs.
Logged
W6UV
Member

Posts: 540




Ignore
« Reply #4 on: October 08, 2012, 11:14:53 AM »

tubes today are a crapshoot.  the only way to guarantee a win is not to play, go for solid state.

but then you have the usual creaky antenna connection opens/shorts issues that blow one or both banks of the output transistors, too.

Is there not a way to design a SS amp to detect such a fault condition and act to protect the transistors? Perhaps that would be too cost prohibitive in a low-end amp?
Logged
WX7G
Member

Posts: 6329




Ignore
« Reply #5 on: October 08, 2012, 11:27:18 AM »

W6UV low end (Ameriton SS amps) are protected against load faults.

W8JI, when I said DC voltage at 2X the amp I meant DC with NO RF.

Well then a good effort is being made to weed out bad tubes yet a good number still make it customer's homes where final burn in occurs.
« Last Edit: October 08, 2012, 11:36:56 AM by WX7G » Logged
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #6 on: October 08, 2012, 12:43:48 PM »

W6UV low end (Ameriton SS amps) are protected against load faults.

W8JI, when I said DC voltage at 2X the amp I meant DC with NO RF.

Well then a good effort is being made to weed out bad tubes yet a good number still make it customer's homes where final burn in occurs.


A DC test at 2X voltage is almost exactly the same as an RF test at 1X voltage. Peak dc voltage in normal operation results in a peak anode voltage of about 2x dc. 

:-)

One problem I caught years ago was the Chinese were testing 572B's in an amplifier running 1500 volts dc supply. That only applies 3000 volts. At or around that time the Russian tubes were either real old stock, or were being made in China and labeled as made in USSR.

In the 1980's a company in California was buying old scrap Eimac tubes, cleaning them, replating them, and selling them as new under the Penta Labs name.

http://www.zoominfo.com/#!search/profile/person?personId=122908991&targetid=profile

http://www.justice.gov/usao/cac/Pressroom/pr2010/138.html

It's a strange world.


Logged
WB2WIK
Member

Posts: 20666




Ignore
« Reply #7 on: October 08, 2012, 02:01:36 PM »


In the 1980's a company in California was buying old scrap Eimac tubes, cleaning them, replating them, and selling them as new under the Penta Labs name.



http://www.justice.gov/usao/cac/Pressroom/pr2010/138.html

It's a strange world.




That wasn't the 1980s, it was September 2010.

Funny about Penta...they were advertising "matched pairs" of 3-500Zs a few years ago and since they're about two miles from my office, I called them and asked what they did to match them.  I was assured they're all "dynamically matched."  So I said, "That is great!  I'm right around the corner and if I can watch you match some, I'll buy a dozen."

Click.  They obviously didn't want me dropping by.  I wonder if they're still there?  I could drive by and have a look from the outside, anyway.
Logged
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #8 on: October 08, 2012, 05:21:45 PM »

That wasn't the 1980s, it was September 2010.

Ameritron bought a bunch of tubes from Penta Labs. They were re-branded Eimacs from the late 1980's or 1990's, when Eimac was having problems with batches of 8877's. This prompted Eimac to start stamping the Eimac logo into external anode tubes so the name could not be washed off and the tube relabeled.

I'll try to get the exact date, but I'm sure this happened. I'm just not sure of the date, but I'm positive it occured BEFORE Eimac started stamping the "tube" logo in the anode connector of the 8877 tube, because that is what prompted the change.

I was a little shocked to see the 2010 date, and don't know if the wheels turn that slow or that stems from new activity. Anyway, looks like he plead nolo contendere to the 2010 charges.
 
Quote
Funny about Penta...they were advertising "matched pairs" of 3-500Zs a few years ago and since they're about two miles from my office, I called them and asked what they did to match them.  I was assured they're all "dynamically matched."  So I said, "That is great!  I'm right around the corner and if I can watch you match some, I'll buy a dozen."

Click.  They obviously didn't want me dropping by.  I wonder if they're still there?  I could drive by and have a look from the outside, anyway.

That wasn't the only "click". There apparently was the sound of a lock clicking:

http://www.corporationwiki.com/California/Chatsworth/steven-jay-sanett/42364294.aspx


There being a finding/verdict of GUILTY, defendant has been convicted as charged of the offense(s) of: Mail Fraud in violation of 18 U.S.C. § 1341 as charged in Count 1 and Fraud Involving Aircraft Parts in Interstate or Foreign Commerce in violation of 18 U.S.C. § 38(a)(1)(C) as charged in Count 2 of the Information.

The Court asked whether there was any reason why judgment should not be pronounced. Because no sufficient cause to the contrary was shown, or appeared to the Court, the Court adjudged the defendant guilty as charged and convicted and ordered that: Pursuant to the Sentencing Reform Act of 1984, it is the judgment of the Court that the defendant, Steven Jay Sanett, is hereby committed on Counts 1 and 2 of the Information to the custody of the Bureau of Prisons for a term of 12 months and one day. This term consists of 12 months and one day on each of Counts 1 and 2 of the Information, to be served concurrently.

It is ordered that the defendant shall pay to the United States a total fine of $70,000. The total fine shall bear interest as provided by law.

Upon release from imprisonment, the defendant shall be placed on supervised release for a term of 3 years. This term consists of 3 years on each of Counts 1 and 2, all such terms to run concurrently under the following terms and conditions:


« Last Edit: October 08, 2012, 06:02:53 PM by W8JI » Logged
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #9 on: October 08, 2012, 06:05:11 PM »

I confirmed the rebranding with 8877's to Ameritron was in the 1980's.

Here is another link for the same person from the 1980's:

http://articles.latimes.com/1986-09-28/local/me-9765_1_iranian-government/2

Quite a long history.

We wonder why the world laughs at us.
« Last Edit: October 08, 2012, 06:08:33 PM by W8JI » Logged
KM3F
Member

Posts: 526




Ignore
« Reply #10 on: October 08, 2012, 07:56:03 PM »

My thoughts are that MFJ might be large enough to consider starting a tube division at some point if it gets bad enough.
If Corning Glass would still make the stems and envelopes or some other supplyer could, in house assembly with grids and heaters in a clean room and an index machine to exhaust them, it might be the answer to quality issues.
There has to be some index machines around yet somewhere.
Even in house close QC still has problems at times.
Last 3-500 were made right in my back yard.
I worked at the plant years ago making microwave devices.
 
Just wild thoughts.
Logged
N4CR
Member

Posts: 1702




Ignore
« Reply #11 on: October 08, 2012, 08:24:12 PM »

My thoughts are that MFJ might be large enough to consider starting a tube division at some point if it gets bad enough.
If Corning Glass would still make the stems and envelopes or some other supplyer could, in house assembly with grids and heaters in a clean room and an index machine to exhaust them, it might be the answer to quality issues.
There has to be some index machines around yet somewhere.
Even in house close QC still has problems at times.
Last 3-500 were made right in my back yard.
I worked at the plant years ago making microwave devices.
 
Just wild thoughts.

I can only imagine that in this day and age of technology that vacuum tube manufacturing could be retooled from the ground up. The problem, of course, would be recouping your expenses.
Logged

73 de N4CR, Phil

Never believe an atom. They make up everything.
G3RZP
Member

Posts: 4965




Ignore
« Reply #12 on: October 09, 2012, 03:30:07 AM »

If you made GOOD 6L6s, KT66s, KT88s, 211s, 2A3s and the like, the audio fraternity would pay handsomely! That could fund the good 572s, 811s and 3-500Zs.
Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #13 on: October 09, 2012, 06:18:09 AM »

If you made GOOD 6L6s, KT66s, KT88s, 211s, 2A3s and the like, the audio fraternity would pay handsomely! That could fund the good 572s, 811s and 3-500Zs.

Even with all the audiophiles buying tubes, it is still a fringe proposition.  If you made quality tubes that had a long service life, you'd eventually put yourself out of business because of the low overall demand.  The days of hollow state are long gone.

Gene
Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #14 on: October 09, 2012, 06:22:04 AM »

My thoughts are that MFJ might be large enough to consider starting a tube division at some point if it gets bad enough.
If Corning Glass would still make the stems and envelopes or some other supplyer could, in house assembly with grids and heaters in a clean room and an index machine to exhaust them, it might be the answer to quality issues.
There has to be some index machines around yet somewhere.
Even in house close QC still has problems at times.
Last 3-500 were made right in my back yard.
I worked at the plant years ago making microwave devices.
 
Just wild thoughts.

LOL!  With the (non-existent) quality control of MFJ overall, can you imagine what kind of tubes they would turn out?  At least as bad as the Chinese!  

They'd also have to hire (and pay handsomely) some tube guys with the expertise and after watching the MFJ factory tour, I doubt that would happen.

Gene
Logged
Pages: [1] 2 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!