My [in development] homemade DC PSU of 30 V / 3 A
Hi. To power future electronic projects of mine (have a few in mind), I've decided to make a DC power supply. Not a simple one, like 5 V / 500 mA or something like that - a power PSU (if you can call it that).
(There are more on the Pictures folder)
- Should be able to generate from 0 V to at least about 30 V between any pins (30 V because it's around the necessary to control the CRT of an oscilloscope I bought - Tektronix 2201)
- Should be able to drive minimally high currents (else it won't be able to do much)
- Must be as efficient as possible, but at least 60% (the more, the better though - with any ideas that come to mind)
- Must have short-circuit and current limiting protections
- Low to very low output ripple (mega maximum of 5%, ideally 1-2%, or even better 1% or less - I read that "For a 5 vdc supply 50mv ripple is an acceptable figure." (1% then), but also that that CPUs need 1% at mega maximum of ripple, with less being the ideal)
- Able to cool down itself (fans), preferably dynamically (less power dissipation)
- It's being designed for max 30 V and 4 A (the transformer I bought can provide 34 V /4 A max on the output)
- Max current on all components of 6 A, including the fuse (1.5 x 4 = 6. Not 2x - more expensive... 1.5 should be enough I guess)
- NTC thermistor on the input to limit the inrush current
- 40 mF in input capacitors to have the ripple stay low even on low loads (as of 2022-01-04, that results in 2% output ripple at 28 V / 4 A with a 7 Ohm load --> means that at lower voltages much lower ripple will happen, and at 9 V / 4 A the ripple is of 9 µV (0.0001%), which seems more than perfect)
- Unregulated voltage outputs for generic supply (like the fan, which needs 12 V)
- Automatic switch between the transformer output to save power with a relay and a comparator (if less than, like, 16 V requested on the PSU output, use the 12 VAC winding. More than 18 V requested, use the 24 VAC winding. Else, [haven't taken care of that, but I think it should keep the winding in use]
- Schmitt trigger is in the works too, to fix a problem of instability (at least when I imagine the circuit working, but I think it's right) when the voltage gets too near the op-amp comparison voltage
- Precision regulator to use for the op-amp to compare with the output, while also being a very low regulator voltage (the greater the comparator voltage, the greater the minimum - and I'd still like a very low output voltage to test on very low voltage devices, if I ever want to, like a clock)
To see the calculations for everything, check the Calculations.docx file. I'm writing everything there.
It's being really cool to make this PSU. It's also taking me much time, because I'm not at home to solder anything on it, so I can only develop on LTspice and think if that matches what I think that would happen in practice (intuition and the little experience I still have).
Also, if any component files are missing, tell me and I'll provide them (I found a lot of devices on a ZIP and downloaded them all and I use any device that is comparable to the ones I have available).
Any new ideas and/or improvements are welcomed! (Just giving the idea and/or making a pull request)