EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Saitama on May 25, 2020, 09:51:11 am
-
Hello! I posted this on other topic, but i think this belongs here.
I have taken interest at designing bench power supplies, but they are not very reliable, so now I am looking for information on designing them. Do you have any suggestions on where should I start reading? Any form of info is fine.
Thanks for reading my post :)
-
Almost all store bought linear PSUs use designs based on the HP/Harrison topology or sometime called "floating"
In this thread, the opening poster has reverse engineered such a design.
https://www.eevblog.com/forum/projects/help-with-how-this-psu-works/msg2897950/#msg2897950 (https://www.eevblog.com/forum/projects/help-with-how-this-psu-works/msg2897950/#msg2897950)
And where it all likely began, https://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1962-07.pdf (https://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1962-07.pdf)
As you likely have already discovered, there are many designs to be found on the internet that don't work properly.
As one member has put it, "they have warts"
The design I used in my last bench supply project actually evolved from one of these dysfunctional internet findings.
https://www.eevblog.com/forum/projects/linear-lab-power-supply/375/ (https://www.eevblog.com/forum/projects/linear-lab-power-supply/375/)
-
A very useful resource for designing the transformer, rectifier, and filter circuits for a linear power supply is the free software "PSUD2" available from Duncan.
http://www.duncanamps.com/psud2/ (http://www.duncanamps.com/psud2/)
The user interface is a bit idiosyncratic, but it gives useful results on the power-up evolution, surge currents, capacitor ripple currents, ripple voltage, and output voltage under load. Note that the transformer is modeled from its no-load voltage and series resistance that limits the voltage under load, while manufacturers normally specify the voltage under load: you need to extrapolate the no-load voltage from that and whatever "regulation" spec the manufacturer has. The ripple current through the filter capacitors may be especially important when choosing components for a high-power system. When comparing specs on suitable electrolytic capacitors, naturally the maximum ripple current varies with physical size, and the choice may affect reliability.
-
A good reference for linear power supplies is the "Kepco Power Supply Handbook", which is not available online but can be purchased. For reliability on a supply that must operate over a wide range of output voltage from a fixed rectifier output, down to low output voltage at high current (maximum stress on the pass device), they invented a circuit that uses two transistors and a power resistor, reducing the dissipation on the transistors automatically (and getting the resistor real hot) in the worst case. I used one for an electromagnet, where the resistor was a nichrome coil wrapped around two ceramic pillars directly in the blast of a fan. Kepco linear supplies are damned-near bulletproof.