Has anybody ever tried to build some form of PSU without transformers, using just a ragulator, say a LM338? I'm thinking something like just the rectifier, a capactitive voltage divider, perhaps, and then the LM338. How about it?...
I just spend some bucks with a pair of large TO-3 LM338, and now I'm realizing I could have spared the twice-as-expansive pair of transformers I bought!
I just spend some bucks with a pair of large TO-3 LM338, and now I'm realizing I could have spared the twice-as-expansive pair of transformers I bought!
Why not?! There are caps and diodes robust enough!
Right now I'm using a rectifier bridge that can handle 1000V "repetitive reverse voltage", and I know there are some capacitors near mF range that can handle 3-figure voltages... If we trust in the capacitive voltage divider, we might even just pile up four 50v 4700uF caps, and we are fine!...
I started to think about this because the LM338 datasheet says that it can handle very high voltages, you just have to keep a 40V difference to the output. So if we want a 40V source, we only need to get the mains from 180V to 80V...
Test it up for me and tell me if it works!
Or do I smell fried chicken?? 
The only really dangerous aspect I see right now is losing the current limiting of the transformer...
Right now I'm using a rectifier bridge that can handle 1000V "repetitive reverse voltage", and I know there are some capacitors near mF range that can handle 3-figure voltages... If we trust in the capacitive voltage divider, we might even just pile up four 50v 4700uF caps, and we are fine!...
I started to think about this because the LM338 datasheet says that it can handle very high voltages, you just have to keep a 40V difference to the output. So if we want a 40V source, we only need to get the mains from 180V to 80V...
Test it up for me and tell me if it works!
Or do I smell fried chicken?? The only really dangerous aspect I see right now is losing the current limiting of the transformer...
For some low-power applications that don't require line isolation, like powering a microcontroller in some appliance, it's ok to use a ballast capacitor followed by rectifiers, a limiting zener and a linear regulator if it's required... See the attached schematic, diodes should be 1N4007...
This approach is not recommended for audio circuits for quite obvious reasons, though.
This approach is not recommended for audio circuits for quite obvious reasons, though.
Current capability depends on ballast capacitor size. The best that can be obtained with a reasonably sized capacitor (2.2uF 400V) at low line conditions (160V on 230V mains) is 50mA or so. Double the capacitance figure for 120V mains (80V low line).
The circuit shown was sized to provide 1/10th of that.
The circuit shown was sized to provide 1/10th of that.
- Status
- Not open for further replies.