¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io

Re: Some observations on linear power supplies


wn4isx
 

Note 3 should read
Add a 0.01uF capacitor with a 27~33 ohm resistor in series across each power rectifier to minimize commutation EMI [radio noise].
?
I have a nice PDF on diode commutation noise and somewhere I have a "story" by? production engineer about how he had to add the RC network to a power supply that had passed military EMI tests for years yet suddenly failed because the diode manufacturer had improved the diodes and they turned off faster, which produced serious EMI.
I know for a hard fact 1968 1N4004 are a lot slower then (most) modern 1N4004 because the silicon is purer and the PN junction area is "more perfect."
For most applications this isn't an added issue for radio power supplies even the 1968 1N4004 required the RC network to eliminate commutation EMI.
?
FWIW and full disclosure, while I hold an extra class amateur and commerical radio engineer license [Formally known as the "First Class ticket", thank you FCC for cheapening my efforts] I am really a shortwave listener.
?
Actually shortwave listener is a bit of a misnomer because I am interested in everything from 1,000Hz [whistlers[ to about 20MHz. I typically go for signals right at my local ambient RF noise floor and any added noise is a problem, so my power supplies must be as quite as I can get them.?
?
At one point I used a R392/URR, portable 24V version of the famed R390A. and used 2 12V deep cycle marine lead acid batteries as my power supply. I'd float charge them when not in use. Eventually it became too much of a hassle dealing with acidic fumes and I was forced to learn how to build a RF quiet 120V to 24V DC power supply. It was an interesting and extremely frustrating process.
?
In 1974 the idea a power diode could produce intense EMI was a confusing concept. There are brute force and elegant solutions to most engineering problems. My solution was brute force, an EE professor took me under his wing and taught me how to finesse my problem, his approach has helped with other EE problems.
?
I am far far from a decent design engineer, I am painfully aware of the areas I'm weak in and don't hesitate to get help from other engineers who in turn come to me for those areas I am an "expert" in,
?
Finesse requires you understand what is going on, clearly the engineers who designed the pedestrian bridge in Florida that collapsed couldn't even manage a brute force approach. While no one reading this is likely to design a bridge, please try to design and build projects that are safe.
?
?
?
?

Join [email protected] to automatically receive all group messages.