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 [5] 6 7 8 9 10 ... 13 Next   Go Down
  Print  
Author Topic: Goodbye tubes.  (Read 7248 times)
KM1H
Member

Posts: 5566




Ignore
« Reply #60 on: August 11, 2019, 05:19:01 PM »

Quote
NASA demanded that test data on space qualified parts be kept for 25 years after delivery. That data was on 15 inch discs....25 years later, I believe that there was one specialist firm in the US who could read such discs and charged accordingly! But data book from those days are still available....

I spent many hours working on IBM and Memeorx drives with those disks all the way back to 10 mb single platter ones.
Logged
W9IQ
Member

Posts: 3563




Ignore
« Reply #61 on: August 11, 2019, 06:26:09 PM »

These are not your grandpa's solid state devices.

A modern LDMOS device, such as the popular BLF188XR can do 1400 watts at 5 watts of pulse drive. The estimated average (MTTF) lifetime with a 50 volt bus is north of 100,000 years when run with a 90 °C case temperature. That is a 100% duty cycle rating, not some goofy tube ICAS rating.

There is no commercial tube that comes within three orders of magnitude of that lifespan. If you are a solid state skeptic, divide that life by 1000 and you still beat out any tube by a couple of generations of ownership.

- Glenn W9IQ
« Last Edit: August 11, 2019, 06:31:46 PM by W9IQ » Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
AC2RY
Member

Posts: 761




Ignore
« Reply #62 on: August 11, 2019, 09:44:04 PM »

The estimated average (MTTF) lifetime with a 50 volt bus is north of 100,000 years when run with a 90 °C case temperature.

- Glenn W9IQ

I think you make a mistake here - stated life span of semiconductors is usually 100,000 HOURS, not YEARS. Still means 11 years of non-stop operation. Other parts like capacitors will start failing sooner than that.
Logged
K6BRN
Member

Posts: 1355




Ignore
« Reply #63 on: August 11, 2019, 09:45:11 PM »

Some comments...

Quote
Solid state amplifiers are NOT really "instant on". It takes 10-20 seconds to start up control processor and do initial self-check before amplifier is ready to transmit.

Well, my KPA-500 is up any ready ro go in less than two seconds (just tried it) and my Yaesu Quadra is up and running in about 5.  Close enough to "instant on" for me.  And their idle power is almost nothing compared to my tube amps, plus moving them around (minor shock and vibration) does not seem to bother them one bit, unlike tube amps.  BTW, my SB-200 takes about 3 seconds and a buddies Henry 2K takes about five.

Glen (W9IQ): 

Quote
The estimated average (MTTF) lifetime (...referring to an LDMOS device...)with a 50 volt bus is north of 100,000 years when run with a 90 °C case temperature.

Might not want to jump on that bandwagon just yet.

Trust me (or don't).  The practical life of LDMOS devices is not 100,000 years, nor is its true MTTF.  The service life (or MTTF) of a rock isn't 100,000 years.  Just look at the pryamids. Smiley  It's very interesting marketing based on less than brutally honest and realistic conditions generally built on simplified models of the gate oxide. There are MANY failure mechanisms still unaccounted for and there is still WAY too little data on this type of part, plus  major geometry details, metalization, doping profiles, embedded protection structures and many more issues are still evolving.  Which is one reason why they are still not used in Hi-Rel applications.

They have a LOT of promise.  But like so many (relatively) new technologies, unrealistic claims set by companies promoting them can actually poison the market if they are taken as the literal truth.  And frankly, my industry, which is built on Hi-Rel, doesn't trust anybodies claims.  We check.  For a reason.  I recall the long-life claims made when GaAs LEDs first came into wide availability with quoted service lifetimes >100 years back in the '70's.  Did NOT happen - those early devices dimmed pretty fast.  And that was just 100 years.  Remember the long-life promise of Lithium-Ion batteries?  The 787 and quite a few consumer products had some major "oooops!" (and Ka-BOOM!) moments before they matured (so much for long life!).  And they are still maturing.

Give LDMOS time.  It'll mature, too.  In the meantime the current devices seem to be working pretty well in amateur amps.

Brian - K6BRN



Logged
G3RZP
Member

Posts: 1326




Ignore
« Reply #64 on: August 12, 2019, 02:16:47 AM »

Both Voyager spacecraft have been operating for over 350,000 hours. The Swindon factory that made some of the integrated circuits supplied for them has been a hole in the ground for about 26,000 hours....having only lasted about 510,000 hours....

There was a UHF TV tx in Germany which, quite a few years back, was mentioned in the electronics press as having taken a klystrode out of service at 110,000 hours and sent it to a sister station because it still met full specification......but that is unusual, to say the least.
Logged
W9IQ
Member

Posts: 3563




Ignore
« Reply #65 on: August 12, 2019, 07:41:08 AM »

I correctly stated years - but I had one too many zeros. It is 10,000 years. It is not hours.

Even if you discount MTTF by a factor of 100, you still have more than 875,000 hours of continuous operation. There is not any tube that is close to this type of MTTF specification.

Regarding other components that make up the amplifier, of course they will have a shorter lifetime in most cases. This is common to both amplifier topologies.

- Glenn W9IQ
« Last Edit: August 12, 2019, 07:45:25 AM by W9IQ » Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
KM1H
Member

Posts: 5566




Ignore
« Reply #66 on: August 12, 2019, 07:41:40 AM »

The original Eimac 3-500Z's in my one owner 1986 Amp Supply LK-500ZC are still producing the same 1200W at the same 75-80W drive level.  It is rather well known that it has been run hard and put away wet after serious 48 hour contests and lots of DX pileups and more recently as an AM linear at the same 1200W PEP. Those tubes certainly stay gettered Grin

That is certainly not a longevity record of hours producing RF but does show how well the "silly ICAS" spec can work with high quality products.  Unlike the seriously overated Russian and Chinese copy 811's being run as 811A's.

Carl
Logged
KM1H
Member

Posts: 5566




Ignore
« Reply #67 on: August 12, 2019, 08:07:33 AM »

Quote
I correctly stated 100,000 years. It is not hours.

Citation please

Quote
Regarding other components that make up the amplifier, of course they will have a shorter lifetime in most cases. This is common to both amplifier topologies.

In MOST cases??  While electrolytic caps have a finite life the more recent ones are already passing 30 years and a decade longer in some cases. External PS are common for SS amps and are conveniently not mentioned in the complaints columns. Fans are common to all. There are more relays in a SS amp and reported failures are common.

The low power draw on standby for SS is a fallacy as the most popular series of ALS-1300 and 1306 draw about 100W which is often more than indirect heated filament amps of the same or even double the power level.

And the list goes on. I'll stick to tubes thank you as Im more interested in long term reliability than marketing hype.
The several other amps used here all have tube date codes from the 60's to 80's and produce rated power.

I am certainly not against SS where it is needed and worked with it professionally from the 70's at Sr Tech to Sr Engineer positions until retirement.

OTOH the US military is apparently returning to hollow state for some critical applications and the Russians havent stopped AFIK.

Carl
Logged
W9IQ
Member

Posts: 3563




Ignore
« Reply #68 on: August 12, 2019, 08:25:40 AM »

Carl,

Do note that I corrected my post to 10,000 years - and that is with a relatively high case 90 °C temperature. You can go up another order of magnitude with a more aggressive cooling package.

You can use Ampleon's reliability calculator at:

https://www.ampleon.com/products/broadcast/0-500-mhzrf-power-transistors/BLF188XR.html

Regarding my comment that "most" components will have a shorter lifetime, components that might not could include things like baluns, splitters, inductors, PCBs, heatsinks, etc. That is why I said "most" rather than "all".

The idling current issue is a question of the amplifier design and it is not a given with all solid state amps. But then a single 3-500Z will draw ~75 watts on the filament alone if it isn't switched off during standby - also a question of the amplifier design.

My broader point is that modern LDMOS solid state finals now far exceed the lifetime of any tube. Lots of other facets can be discussed but the dagger is in the device MTTF question.

- Glenn W9IQ
Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
W0BTU
Member

Posts: 2268


WWW

Ignore
« Reply #69 on: August 12, 2019, 08:33:23 AM »

That's interesting, but how about the IMD specs as compared to good tubes?

These are not your grandpa's solid state devices.

A modern LDMOS device, such as the popular BLF188XR can do 1400 watts at 5 watts of pulse drive. The estimated average (MTTF) lifetime with a 50 volt bus is north of 10,000 years when run with a 90 °C case temperature. That is a 100% duty cycle rating, not some goofy tube ICAS rating.

There is no commercial tube that comes within three orders of magnitude of that lifespan. If you are a solid state skeptic, divide that life by 1000 and you still beat out any tube by a couple of generations of ownership.

- Glenn W9IQ
Logged

AC2RY
Member

Posts: 761




Ignore
« Reply #70 on: August 12, 2019, 11:18:02 AM »

That's interesting, but how about the IMD specs as compared to good tubes?



If used no higher than -3dB below rated power along with moderate negative feedback transistors have IMD comparable with tubes. When you get into clipping, distortions from transistors raise much more quickly than from tubes. But this was well known for over 50 years, that is why guitar amplifiers still mostly use tubes.

But somehow people complain about distortion at clipping level. Transistors at rated power should only be used for CW mode, nothing else. To amplify any wide band signal -3dB is a reasonable design limit.


Logged
KM1H
Member

Posts: 5566




Ignore
« Reply #71 on: August 12, 2019, 11:47:20 AM »

Quote
But somehow people complain about distortion at clipping level. Transistors at rated power should only be used for CW mode, nothing else. To amplify any wide band signal -3dB is a reasonable design limit.
 

Clipping also occurs with CW especially when the final tubes are biased quite deep into Class C. I suspect similar occurs with overdriven SS.

Carl
Logged
G3RZP
Member

Posts: 1326




Ignore
« Reply #72 on: August 12, 2019, 01:24:53 PM »

Deep space probes do have an advantage in that ambient temperatures are low - but all the heat has to be radiated. Still, 40+ years isn't a bad record for pretty well continuous operation, although there is a carbon filament lamp in a fire station in, I believe South Carolina, that hasn't, it is claimed, been off since 1912.
Logged
W0BTU
Member

Posts: 2268


WWW

Ignore
« Reply #73 on: August 12, 2019, 01:38:44 PM »

I suspect you're right. :-)  Otherwise, key click QRM is inevitable.

Clipping also occurs with CW especially when the final tubes are biased quite deep into Class C. I suspect similar occurs with overdriven SS.
Logged

AC2RY
Member

Posts: 761




Ignore
« Reply #74 on: August 12, 2019, 01:45:28 PM »



Clipping also occurs with CW especially when the final tubes are biased quite deep into Class C. I suspect similar occurs with overdriven SS.

Carl

When you only have CW signal - there is (almost) no IMD - only harmonics, which can be easily filtered out.

Tubes do clip, but raise of high order harmonics and IMD is much slower. That is why tubes can be overdriven (HAMs commonly do that) with less impact on signal quality.


 
Logged
Pages: Prev 1 2 3 4 [5] 6 7 8 9 10 ... 13 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!