Fast Recovery rectifier diodes

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Eva said:
powerbecker:

The software package that comes with my oscilloscope produces huge pictures, and I can't reduce them because when I do so the lines disappear and everything gets distorted. Try to configure your video adapter to 1024x768 temporally so that you can see them more comfortably.


My TEK TDS produces a big BMP file -- I pull it into Photoshop and "save-to-web" as a GIF -- there are other programs which do this as well.
 
Wonderful thread...

Eva, nice work...very nice.

1. Standard diodes are minority carrier devices. Schottky's are majority carrier devices.
2. Standard diodes need all the minority carriers removed from the depletion zone, this is done by external lead current, or by recombination.
3. Recombination is affected by temperature, doping density, doping profile (linear, exponential, etc..), and impurities (gold or platinum diffused in, or ion implantiontion to provide lattice defects. Impurities will exact a heavy reverse leakage penalty I don't recall if gold or platinum is worse...memory's fading.....ion implantation may or may not be permanent.
4. Schottky's require only that the capacitance be discharged by the external connections.

During TRR, currents within the circuit can flow in the opposite direction...a schottky is just a cap...a PN will stop current flow gently, abruptly, quickly, or slowly...a two by two matrix of possibilities.

Old school TRR methods used 1/2 amp forward, 1 amp reverse, time to 1/4 amp reverse was the measure.

Newer tests use recovered charge. Different, but the same.

TRR does not change over 5 orders of magnitude current scaling..in other words, if you use 1/5 milliamp, 1 milliamp, 1/4 milliamp, you'll get the same numbers...as long as the die is the same temp...the 1/2 amp can heat the junction.

5. Forward bias junctions have very large capacitance. Reverse capacitance is heavily dependent on reverse bias...as EVA stated.
6. Don't worry about TFR...it's kinda fast.

7. Measurement of TRR at the subnano level is easy to do once the circuit issues are understood...but for line supplies, is unwarranted. Standard fare diodes are 2 microsecond..



Cheers, John
 
Accidentally, measured initial primary current surge of Talema 15VA toroid 2 x 18V, with PSU and nominal load connected.
 

Attachments

  • odber_naraz_s.gif
    odber_naraz_s.gif
    43.8 KB · Views: 1,581
Would some fast MUR3060 (30A, 600V) Diodes be suitable for a planned HV CFL lamp project?

I made a 12V DC SMPS board that outputs approx 170V AC squarewave at 50khz.

I want to run a Compact Fluorescent ballast & bulb normally meant to run the 120VAC, 60hz from the wall socket. Inside the ballast, it normally has a voltage doubler to double the wall voltage, and I measure 320V DC before it powers the ballast & bulb.

I want to use the MUR diodes to double the 170V at 50khz from my SMPS to 340V DC then directly power the ballast with that.

Will those 600V diodes be safe at such a high voltage and high frequency? Could I use 680uf caps from computer SMPS and the diodes still be OK?
 
RE - fast Schottky diodes

same experience with Parmeko tube amp - replaced 10 amp 1000 volters-piv- with ordinary silicon power diodes and got harsh shocking sound -- and again for GZ34 with 1a 1000 piv then with 10 amp 1000 piv- in Quicksilver monoblocks with magnificent sound. But way too much HT. THESE DIODES are perfect for valve amps in bridge arrangement and very cheap Schottkys too- meant for solar cell panels. RSVP.
 
As an ounce of practice - to all my detractors I suggest just try these cheap diodes - axial 10 amp 1000 PIV - meant for solar cell panels - and lo - hear the tremendous difference for yourself in sound quality ~ and be amazed, BEFORE theorising why - I challenge other opinions with this simple practical test. QED.
 
No need for a fancy fast recovery diode at 50/60Hz.
Agree for tube amp B+ supplies.

For ss amps, the load current can be significantly higher, and some people use over-sized power transformers (low winding resistance, and substantial leakage inductance) along with huge amounts of first filter capacitance, so diodes get turned off with significantly higher dI/dt as dV/dt at turn-off can be exacerbated by the leakage inductance energy.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.