Since I got my Amateur Radio Operator license a few months ago I've been in the process of putting together my station. One of the aspects of this project is to build a reliable power source to run my various radios and amplifiers. I've successfully converted a couple of old(ish) ATX PSUs to output 13.8v to 14v, with some working better than others. Each are rated at ~20 amps on the 12v rail.
I plan on documenting the various components with pictures in this thread so we can see what components should be upgraded in order to supply more current/power. I have a fried PSU that can source 60a on the 12v rail so I may pull components from it to increase the power capacity of this supply.
I converted these by removing the 5v regulation and using a voltage divider on the 12v rail to regulate the PSU (the PWM controller sees 5v through the voltage divider). If I increase the voltage above 14v the PSU won't power on. Voltage is what I consider fairly stable -- 14v at idle and the lowest I've seen is 13.8v under a roughly 200w peak load (~15a).
I also added a gang of supercapacitors on the output. Total capacitance from this should be roughly 1F. Unfortunately I'm afraid the total ESR of the 7 capacitors in series is too high to do much good (each capacitor is rated at 80 mohms esr).
The problem I've come across is the transient nature of the power consumption of the SSB transmission mode is randomly powering down my PSU.
I don't believe the PSU is going into overpower or overcurrent protection; it will shut down at random whether I'm using the amplifier (15a max) or running barefoot (less than 1/4 the power of the amplifier). I can sometimes talk for upwards of 20 minutes without a shutdown and other times I can't finish a sentence. Once the PSU starts to power down it will do so regardless of the power settings.
I don't believe temperature is a factor; the room temp has been anywhere from 40°F to 85°F and if anything the PSU will shut down more often when it's cold but I have no way to quantify that. I have never felt any heat coming out of the back of the supply (high speed fan runs directly off the 14v rail).
It seems to be able to run more stably in AM mode which has a more constant power draw with more average power used but less peak power. However, when talking on AM in medium or high power mode the PSU makes a slight squeal corresponding with my voice. This noise doesn't occur in SSB mode.
I have replaced a few bulged output capacitors with higher value components - both in voltage and in capacitance - though they came out of another PSU so admittedly I don't know their ESR (just that they aren't bulged and my Fluke meter measured the correct capacitance).
Strangely even with a modest ~250 watt load I’ve noticed that the lights slightly dim in my shack when I’m talking in SSB mode (following my voice). Might larger input capacitors alleviate this problem? It should be noted that my 2kw electric heater and ¼HP grinder don’t cause the lights to dim appreciably. Could this possibly be what is causing the PSU to shut down?
I plan on documenting the various components with pictures in this thread so we can see what components should be upgraded in order to supply more current/power. I have a fried PSU that can source 60a on the 12v rail so I may pull components from it to increase the power capacity of this supply.
I converted these by removing the 5v regulation and using a voltage divider on the 12v rail to regulate the PSU (the PWM controller sees 5v through the voltage divider). If I increase the voltage above 14v the PSU won't power on. Voltage is what I consider fairly stable -- 14v at idle and the lowest I've seen is 13.8v under a roughly 200w peak load (~15a).
I also added a gang of supercapacitors on the output. Total capacitance from this should be roughly 1F. Unfortunately I'm afraid the total ESR of the 7 capacitors in series is too high to do much good (each capacitor is rated at 80 mohms esr).
The problem I've come across is the transient nature of the power consumption of the SSB transmission mode is randomly powering down my PSU.
I don't believe the PSU is going into overpower or overcurrent protection; it will shut down at random whether I'm using the amplifier (15a max) or running barefoot (less than 1/4 the power of the amplifier). I can sometimes talk for upwards of 20 minutes without a shutdown and other times I can't finish a sentence. Once the PSU starts to power down it will do so regardless of the power settings.
I don't believe temperature is a factor; the room temp has been anywhere from 40°F to 85°F and if anything the PSU will shut down more often when it's cold but I have no way to quantify that. I have never felt any heat coming out of the back of the supply (high speed fan runs directly off the 14v rail).
It seems to be able to run more stably in AM mode which has a more constant power draw with more average power used but less peak power. However, when talking on AM in medium or high power mode the PSU makes a slight squeal corresponding with my voice. This noise doesn't occur in SSB mode.
I have replaced a few bulged output capacitors with higher value components - both in voltage and in capacitance - though they came out of another PSU so admittedly I don't know their ESR (just that they aren't bulged and my Fluke meter measured the correct capacitance).
Strangely even with a modest ~250 watt load I’ve noticed that the lights slightly dim in my shack when I’m talking in SSB mode (following my voice). Might larger input capacitors alleviate this problem? It should be noted that my 2kw electric heater and ¼HP grinder don’t cause the lights to dim appreciably. Could this possibly be what is causing the PSU to shut down?
Depending on what the PSU uses, it may be a bad idea to swap the 5V and 12V rectifiers. Most likely the 5V uses a TO-247 30A schottky and the 12V uses a 12 or 16A ultra fast recovery diode. If so, then swapping them would probably throw all the voltages out of whack. If there's a schottky on the 3.3V, swap that one with the 12V. It probably has 13007 switchers, which are good for usually 250W but depending on the size of the main transformer it may not even be worth it to replace the switchers.
Half of the computer problems is caused by bad contacts 
Comment