The AM Forum
March 19, 2024, 10:04:33 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: HV power rectifiers  (Read 4070 times)
0 Members and 1 Guest are viewing this topic.
W4AMV
Member

Offline Offline

Posts: 672


« on: March 22, 2017, 05:11:56 PM »

Good day forum,

I am planning another amplifier after having FUN with a pair of 811A's where the HV supply used a string of 1N5408 diodes. I see an an earlier post that the 10A10 diode is a better alternative and they are reasonably priced from various sources. However, after reading this "old" 73 Magazine article, portion attached, I wonder at what risk are we if we compromise on the quality of the diode manufacturer? Many of the lower cost diodes are manufactured abroad. The 73' article is a good read and available from the 73' archive. Comments and thanks! Alan


* 73_Dec_80.jpg (277.37 KB, 820x876 - viewed 228 times.)
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2644


Making AM GREAT Again!


« Reply #1 on: March 22, 2017, 05:30:05 PM »

Diode technology has come a long way in 40 years.

I use the 10A10 almost exclusively.

--Shane
KD6VXI
Logged
Pete, WA2CWA
Moderator
Member

Offline Offline

Posts: 8154


CQ CQ CONTEST


WWW
« Reply #2 on: March 22, 2017, 05:35:06 PM »

How can you define and/or determine "if we compromise on the quality of the diode manufacturer" I've used diodes in HV circuits for years where I had no clue where they were actually manufactured. If I defined the circuit parameters correctly, they all worked as per their specs. Sort of like asking if diodes from China, are better or worst then diodes from Korea, better or worst then diodes from Japan, better or worst then diodes from U.S., etc.
Logged

Pete, WA2CWA - "A Cluttered Desk is a Sign of Genius"
W4AMV
Member

Offline Offline

Posts: 672


« Reply #3 on: March 22, 2017, 05:46:31 PM »

Fair answers. If today diode PIV is a conservative rating, I agree. I would like to think that if I use a 1kV rated PIV diode on a 2 kV supply that 8 diodes in each leg should provide me with sufficient head room. Furthermore do you dispense with the resistor equalizers as well the shunt capacitors across the diodes?
Logged
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #4 on: March 22, 2017, 07:23:42 PM »

Fair answers. If today diode PIV is a conservative rating, I agree. I would like to think that if I use a 1kV rated PIV diode on a 2 kV supply that 8 diodes in each leg should provide me with sufficient head room. Furthermore do you dispense with the resistor equalizers as well the shunt capacitors across the diodes?

Eight diodes in each leg should be enough,  I would use 7-8 diodes for 2KV rms supply.  Most will say that you don't need the resistors and caps but I always include them.
Logged
N8ETQ
Member

Offline Offline

Posts: 795


Mort


« Reply #5 on: March 22, 2017, 07:40:52 PM »


     

         Yo'


      Nothing wrong with the trusty 5408
'cept I hear they are obsolete. 6A10 or 10A10's are
a good sub. if you can take the larger footprint
and lead size.

                I read somewhere years ago that due
to the consistency from unit to unit credited to
improved manufacturing technique's that the R/C
combo across each diode is unnecessary. I have not
bothered with those in years and have not had any
issues either.


                I have also had good experiences
using "Zero Crossing SSR's" for input power to the
supply, Eliminating the need for a soft start setup
when hitting a high uF Cap bank.

GL es 73

/Dan


Logged
WB2CAU
Member

Offline Offline

Posts: 342


« Reply #6 on: March 23, 2017, 11:01:03 PM »

1N5408 rectifier diodes available here for $0.06 each:

http://bit.ly/2obb4ed

Logged

"Life is tough, but it's tougher if you're stupid." -- John Wayne
w7fox
Member

Offline Offline

Posts: 102


« Reply #7 on: March 24, 2017, 02:20:28 AM »

I have been reading this thread with some alarm.  I have been using 1n4008 type diodes for all my power supply projects, without a single failure for years.  I get most of my stuff from Allelectronics or Mouser.  I check the surge rating, inverse voltage and current, and these old standbys have always worked.  Have I just been lucky?

Best regards,
Fox
Logged
W4AMV
Member

Offline Offline

Posts: 672


« Reply #8 on: March 24, 2017, 02:48:31 AM »

I have been reading this thread with some alarm.  I have been using 1n4008 type diodes for all my power supply projects, without a single failure for years.  I get most of my stuff from Allelectronics or Mouser.  I check the surge rating, inverse voltage and current, and these old standbys have always worked.  Have I just been lucky?

Best regards,
Fox

Hi Fox, I suspect not. Perhaps this old article was off base in relation to today's devices. However, there are "bad" suppliers out there so it peaked my concern. I spoke to folks at Cree. They supply 1700 V PIV SiC diodes. They do suggest R equalizers despite the fact they have conservative PIV specs.   
Logged
N8ETQ
Member

Offline Offline

Posts: 795


Mort


« Reply #9 on: March 24, 2017, 08:29:43 PM »




    yO'


        I guess it all comes down to ROI.
Are you willing to spend a few bux (Maybe $10)
to save a 25 cent diode?

        Plus the "Real Estate" and time. I'm not.
Seems akin to puting in a circuit breaker to save
your fuses to me.


        Let them fry.. They prolly won't, and if
they do, those diodes are the least of your problems.


/Dan

Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8304



WWW
« Reply #10 on: March 24, 2017, 09:24:28 PM »

The cheap-type diodes often sold to hobbyist market in the past have been populated with the parts that were still functional but failed some test or another. This was in a time when semiconductors were not produced in anything like the volume of today, and the processes were nowhere as well-refined and controlled.

Today, well after that article, unless the part is a counterfeit or marked incorrectly, one can count on the device's ratings from a reputable manufacturer.

Within a lot, meaning a stack of wafers, the devices are uniform enough now to not need to be equalized. This is the result of decades of fine tuning the processes for ever-higher yields, translating to ever-higher quality.

Defective individual chips are inked and marked by automated equipment, and thrown into the recycle when the wafer is sawn onto the 20,000-100,000 little rectifier dice that were made on it.

It is not worth it for a single non-spec part to escape a semiconductor factory today. Just one making its way into the hands of a contract manufacturer's high speed assembly line (Apple? GM?) and into an end product which then fails prematurely can cause more trouble than can be easily imagined, including line-down, tens and hundreds of thousands of dollars costs in investigations, after-the-fact QA (decap and x ray and other tests) on many many individual parts, global concalls at all hours of the night, and the recall, scrapping, contracted delivery stoppage costs, and replacement of whole crates of devices just to make sure they got them all. It's very ugly.

I could not dare to speculate as to why the SiC units are recommended to have equalizers. It is always good practice (insurance if you like) with any stack of diodes, or maybe the process as it is today does not permit the same uniformity of reverse specifications at this time, since SiC is 'new' as fabrication technologies go. Just do as the engineer says and all will be well!

All semiconductors should be bought only from the manufacturer's authorized distributors, unless there is some huge compelling reason to buy from a surplus reseller who may not know exactly where the parts have come from unless they have documented lot numbers. You have quality assurance all the way back to the fab when you buy it from an authorized distributor. When you buy it from someone else you have only their good faith or guarantee. Counterfeiting is rampant (more so on costly MOSFETs and the like) and the semicondctor manufacturers don't like that known but the above paragraph is the key to your getting quality products. Just my opinion after 13 years in a semi company and seeing it happen.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
W4AMV
Member

Offline Offline

Posts: 672


« Reply #11 on: March 25, 2017, 02:22:21 AM »

Hi Patrick. Summary well stated. Equalizers were suggested as a precaution, while the PIV rating is conservative, some customers are known to push the envelope.

Alan
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.087 seconds with 19 queries.