My brain got stuck at Power Supply question

[anotherjim]

Must not forget physical size when choosing parts. >1A diodes and bridge rectifiers tend to have thicker leadout wires. These can be too much for most of the perfboard/stripboard pcb out there, although you can drill the holes out a little to make them fit, it isn't ideal.
There used to be types of boards with wider spaced, bigger holes than the common 0.1" matrix that was ideal for PSU/power projects, but I haven't seen any of that for many years.

[KarenColumbo]

Must not forget physical size when choosing parts. >1A diodes and bridge rectifiers tend to have thicker leadout wires. These can be too much for most of the perfboard/stripboard pcb out there, although you can drill the holes out a little to make them fit, it isn't ideal.
There used to be types of boards with wider spaced, bigger holes than the common 0.1" matrix that was ideal for PSU/power projects, but I haven't seen any of that for many years.

Ah, well, I never touch perfboard of vero or stuff like that. I etch 'em myself - and AllPCB do my double-sided things.

[KarenColumbo]

Well say hello to my little friend!

Handmade in Vienna, Austria. 6 x 12V, 1 x 15V and 1 x 18V secondaries for 9V, 12V and 15V DC. Cost 140 Euros (tax included) - but I think they would do this for less if only they got to know me better

[KarenColumbo]

> feel free to go wild.

+1

I would have told you: don't fool with 50V diodes even if they are ample. The cost of 400V diodes is so close that they are often same-price at DIY retail. The extra voltage rating does no good or harm UNTIL you need it. R.G. figures with everything nominal you may be 1V shy on your 18VAC winding(*). I'd expect 20% un-sag at no-load. "240V" is very nominal and while my 125V is steady, other folks report wild variations (to over 130V!). And then there is elevator-motor spikes and lightning near-misses. 2X over-rate is really not enough, 4X is better, and frequently 10X is same price.

Also cheaper by the dozen and usable for other things.

Here's your choices for the part you bought:
VS-2KBP005  50 Volt 2.0 Amp $2.26
VS-2KBP04  400 Volt 2.0 Amp $2.26
VS-2KBP10 1000 Volt 2.0 Amp $2.43
You see that 400V is same-price, so why buy less? 1,000V is only pennies more. If you are opposed to high voltage, the 400V part will cover anything you do with ample margin for spikes and mistakes. If you might work near hundreds of Volts, the 1KV part may be worth the 17 cents.

(*)Oh, and don't worry about 51V on your "50V" parts. There was a time when "50V" meant you were sure to never see a 100V breakdown (they'd sell it as 100V and higher price.) But the days of sort-out are declining. The price-list above suggests they aim for 1000V and get >400V almost every time. They list a 50V part for buyers who take a BoM too literally, or in case they have a bad day in the kitchen, but I bet you are really getting hundreds-Volt parts at the 50V price.

Thanks a lot for this hint - I will do an order round this weekend and THINK before I click this time

[KarenColumbo]

Couldn't find too much about that topic @ Google - maybe I asked the wrong questions. But this should suffice, I gather. [...]
Am I on the right track with those? (I'm still reading in on the topic power supplies, but a quick answer would help me edge closer to the gist of things)

I suggest you go read http://www.geofex.com/Article_Folders/Power-supplies/powersup.htm at geofex for some detailed info on power supplies.

It is usual in the power supply industry to start with the desired DC output voltage and current and then work backwards through the regulators, filters, rectifiers and transformer, specifying each part earlier in the chain of power from the AC wall socket. But if you are doing what DIY people mostly do, and buying a transformer first, you can calculate what output current and voltages are possible from that.

You probably already know your desired output voltage - usually 9Vdc for pedals.
You will want to regulate it. The two common solutions are the 7809 regulator or the LM317 adjustable regulator. Each of these regulators (and three-terminal regulators in general) need about 2V "headroom" above their output voltage to stay in regulation. So you need to supply them a minimum of 11Vdc (for a 9V output) to avoid having them saturate and letting ripple and noise flow through.
You can reduce ripple to almost any desired level before the regulators by specifying ever-bigger capacitors. As illustrated in "Power Supply Basics", the filter caps get a pulse of DC at the peak of the AC wave feeding them through the rectifiers, and then have to carry the load all by themselves until the next AC wave peak. So the voltage on the caps peaks at each AC peak, then runs down almost linearly At a rate of dv/dt = I/Cfilter. So you can make ripple as low as you want by making Cfilter bigger.

But you can never make the voltage on the filter caps bigger than the peak of the AC power wave minus the rectifier forward drop. So your transformer has to make enough peak voltage so that your regulator never gets below 2V + 2*diode voltage above the desired output voltage. Note that this includes low AC line voltage, so you have to design for more than this.

Your transformer that you ordered already has a fixed amount of output voltage for any given AC power line voltage, so you don't get the luxury of specifying it. You get whatever you get, and have to hope it's OK. In general it will be, but when you finish your testing, you'll know.

As for your question about rectifiers, always ensure your rectifiers can support at least twice the peak of the AC voltage your transformer puts out. Voltage rating is cheap in rectifiers, so feel free to go wild. I think you had one winding at 18Vac if I remember correctly. That's 18Vac*1.414 = 25.45Vpeak. So 50V rated rectifiers would be just barely enough on this one winding. But 100V, 200V, 400V rated rectifiers will not cost more than a few pennies more. Get the higher voltage ones; use 100V or 200V. Current is another issue. There is a bit of calculus to get there, but a full wave rectifier "sees" an RMS current that is about 1.6 to 1.8 times the DC current that leaves the filter capacitor that it feeds. This is worst for very large capacitors and low ripple current. Fortunately again, current rating is cheap in low-power diodes. You can get 1A rectifiers for pennies each. That will support half an amp of DC current out.

Read "Power Supplies Basics", and ask any questions that come up.



To get 11Vdc minimum on your power supply filter caps, you need the peak of the AC voltage feeding your rectifiers to be

Thanks again, R.G. - this will take up my weekend. Looking forward to learn a few more things

[KarenColumbo]

I still can't get my mind off those, though: https://www.xppower.com/Product/ICZ09-Series