scope of project : take atx power box out of old pc
project outcomes : bench supply for 3.3v,5v,12v, adjustable voltage under 12v,
: PD charging USBS
: supply power for PINECIL solder iron
: fuses to all outgoing power supply’s
where im at : i have the power supply and have checked that its working bu jumping the standby to ground.
: will be using buck converter to supply adjustable voltage under 12v
: have made housing out of Perspex
THE QUESTION: is there a PD module that i can use for the USBs???
This charger can be driven from the 12V supply of the ATX PS.
Note that just taking the power control lead to ground is not sufficient to convert the ATX PS to a bench power supply.
Firstly, the 3.3V monitor lead must be tied to 3.3V. Without doing this the PS might be continually attempting to bring the 3.3V supply up to the required voltage, when in fact it’s already there.
Secondly, one or more of the power outputs should have a load resistor to put a permanent load on the unit. 200mA or thereabouts is suitable, and 12V is probably the best one to use, although experimenting may be required. Without this load regulation may be poor or the the unit may run much hotter than normal.
Hay Jeff, thank you for contributing. You say some testing may be needed, can you link something to explain what this would entail, I’ve read that not all Atx power supply’s need the resistance, when I take the standby to ground (green lead) all the relevant wires power up and the fan comes on, is there a test to determine if it needs the resistance.
Not questioning your knowledge just wanting to understand
You can’t really test the 3.3V monitor input, because if the PS hasn’t already blown up then there is some inbuilt protection, and what that protection might do is unknown. For instance, it might limit the current on the 3.3V output, or it might just disable the control mechanism and leave the output to find its own level - the only way to find out would be to examine the circuit. That behaviour is not part of the ATX spec (it just says that the facility must be there) and each unit could be different. Connecting it to 3.3V is very simple insurance.
The best test for the load resistance requirement is that the PS shuts down without it! This is the suggested action in the ATX spec for ‘no-load operation’. However not all supplies will shut down immediately, and it can be difficult to determine if it was supposed to shut down or it is designed to continue operating under ‘no-load’. But even if it shuts down properly under ‘no-load’ this can cause problems with a bench supply, when you don’t expect power to be coming and going. You could test for a minimum load requirement by measuring the ripple and temperatures with and without the load installed, and choosing the arrangement that gave the best results. This can also be regarded as insurance, although it is a little more difficult to arrange.
Of course, the best option would be to determine the required installation details from the manufacturer’s specification, but not many ATX power supplies provide that level of detail in their specs.
I was just rereading your comments. The first Atx I had was giving out 4.4v on the 3.3v lines. Most likely needed the load on the monitor line to calm it down.ill get the resister and check it out
That’s not the load resistor - it’s the 3.3V monitor. That lead needs to go to any 3.3V output, preferably right at the terminal. It’s the way that the PS checks that it really is outputting 3.3V. It’s the brown wire on pin 13 and is usually of noticeably thinner section.
