Silicon Chip magazine = Seriously Crazy
Posted: October 16th, 2013, 1:14 pmThe circuit notebook in the 2012/08 issue describes how to get 24V from two PC PSUs, by wiring their +12V outputs in series. But that's never going to work - unless you're in a house with outdated wiring (which may have been the case because the author of that idea is stated to be in Russia, and they did have class 0 (neither earthed nor double insulated - now illegal in most countries) sockets there, and it's apparently common to misuse this to run modern appliances from the old wiring) - because the secondary is connected to mains earth, so all that will happen is that the negative-side supply will be shorted. I can't believe they actually paid the author of that "trick". It also mentions:
But the part numbers quoted are 6A4/P600G, which are 6A 400V standard rectifiers.
Anyway, in the 2012/04 and 2012/05 issues, they described an induction motor speed controller, designed to handle 1.5kW with either a normal PSC (permanent split capacitor) motor designed for single-phase mains or a 3-phase motor wired to operate with 231V between phases (for reference, if you built a three-phase wye system with that value between phases, phase to neutral would give 133V). But it has no PFC, potentially resulting in hell on the grid, and they installed X2 capacitors between each side of the high-voltage DC bus and earth. I'm sure they'll use some excuse involving how important earthing is anyway, but it's inevitable that, sooner or later, someone with a bad earth will plug that thing in. The value of those caps - 47nF - will result in quite a tingle if the user touches the case with failed earthing, but normally not lethal (<5mA). (Class I appliances should still be made as safe as practical in the event of earth failure.)
They also have a habit of running capacitors, electrolytics at least, very close to their rated voltage, such as 16V on a 15V rail, and in two of their amplifier projects (I think they were the Ultra-LD Mk.2 and 3) they used a split supply with a 2 * 40V transformer and 63V filtering caps (three 4700µF parts on each rail). You can figure out the rest yourself, but in short, hope that "80V" transformer secondary doesn't exceed 89V. It's just as bad as using a 350V cap to smooth the rectified mains. (A proper choice of capacitor for that supply would be Panasonic HC in 4700µF 80V 25mm, or alternatively two 6800µF 80V 30mm per rail. But I think they designed around 22mm wire-leaded types.
If that's the case, there are only four Japanese equivalents to them, two in each temperature grade - Chemi-con SMQ and KMQ, and Nichicon VK and VY.) Let's not forget that they don't specify the diameter of the electrolytic capacitors used in their projects, and they never hint at how low ESR they mean when they specify a "low ESR" electrolytic...
And in the circuit notebook in the 2012/05 issue, the one who suggested removing a ballast from a CFL and using it to power a conventional tube won a $150 gift voucher. Not only has it been done before (and that's only one example) but as I've mentioned before, the things can blow up even if you do nothing to destroy them. And there's no guarantee that the integrated ballast will survive tube failure (even the very-cheap-and-nasty ballast I found in that Crompton fitting has a shutdown circuit based on a small SCR).
The publication is either mad or they care more about making their projects easy to obtain parts for, than the actual safety and reliability.
And there's a non-obvious danger with attempting that "24V" idea - if you have a counterfeit mains cord (which can have extremely high resistances), you can get fire if the second PSU is beefy enough, because in that case, it won't shut down.
Schottky diode D1
But the part numbers quoted are 6A4/P600G, which are 6A 400V standard rectifiers.
Anyway, in the 2012/04 and 2012/05 issues, they described an induction motor speed controller, designed to handle 1.5kW with either a normal PSC (permanent split capacitor) motor designed for single-phase mains or a 3-phase motor wired to operate with 231V between phases (for reference, if you built a three-phase wye system with that value between phases, phase to neutral would give 133V). But it has no PFC, potentially resulting in hell on the grid, and they installed X2 capacitors between each side of the high-voltage DC bus and earth. I'm sure they'll use some excuse involving how important earthing is anyway, but it's inevitable that, sooner or later, someone with a bad earth will plug that thing in. The value of those caps - 47nF - will result in quite a tingle if the user touches the case with failed earthing, but normally not lethal (<5mA). (Class I appliances should still be made as safe as practical in the event of earth failure.)
They also have a habit of running capacitors, electrolytics at least, very close to their rated voltage, such as 16V on a 15V rail, and in two of their amplifier projects (I think they were the Ultra-LD Mk.2 and 3) they used a split supply with a 2 * 40V transformer and 63V filtering caps (three 4700µF parts on each rail). You can figure out the rest yourself, but in short, hope that "80V" transformer secondary doesn't exceed 89V. It's just as bad as using a 350V cap to smooth the rectified mains. (A proper choice of capacitor for that supply would be Panasonic HC in 4700µF 80V 25mm, or alternatively two 6800µF 80V 30mm per rail. But I think they designed around 22mm wire-leaded types.
And in the circuit notebook in the 2012/05 issue, the one who suggested removing a ballast from a CFL and using it to power a conventional tube won a $150 gift voucher. Not only has it been done before (and that's only one example) but as I've mentioned before, the things can blow up even if you do nothing to destroy them. And there's no guarantee that the integrated ballast will survive tube failure (even the very-cheap-and-nasty ballast I found in that Crompton fitting has a shutdown circuit based on a small SCR).
The publication is either mad or they care more about making their projects easy to obtain parts for, than the actual safety and reliability.
And there's a non-obvious danger with attempting that "24V" idea - if you have a counterfeit mains cord (which can have extremely high resistances), you can get fire if the second PSU is beefy enough, because in that case, it won't shut down.