diyAudio Power Supply Circuit Board v3 illustrated build guide

No, 150w x 2 gives you the minimum power requirement of 300w so at least that. A decent 400va toroidal should do the job. No need to go nuts.
Depending what dc voltage the amp works on you have to choose the correct secondary ac voltage for the transformer.
For my f4 I need dual 24v dc rails. So plus 24v, gnd and minus 24v.
When you rectify an ac signal you end up with approximately (usually slightly less) 1.4 x the output voltage of the tx (abbreviation of transformer).
I used a 400va 240v to 2x 18v.
This gives 18v x 1.4= 25.2v
As there are 2 sets of secondary windings you connect one end of one winding to one end of the other winding and that goes to gnd. The rectifiers take care of the rest and you get a dual rail dc supply of plus and minus 24v.
Therfore you should probably use a minimum cap rating of 35v. Which would be fine in this case as they go from rail to ground (24v) and not rail to rail (48v).
You generally only find 450v caps in tube circuits or maybe (guessing a bit here) switched psu. They also tend have low capacitance, nowhere near enough for this. Your going to need around 40,000uf per rail.
Check the esp website for more details on how linear supplies work.
 
That's better. You saved me $$. Beer on me and this brewery comes from your neck of the woods, sort of.
 

Attachments

  • DSC_0203.JPG
    DSC_0203.JPG
    129.8 KB · Views: 743
Remember to use a ntc thermistor (cl60 or similar check the build guides) on the transformer input or the soft start board. You get a lot of inrush with 300w+ transformers whilst the caps charge!
the transformer demands an enormous current surge at start up even without any capacitor filter on it's output.
A soft start is to get the transformer started without blowing the mains fuse.

If you have concerns for your capacitor bank, then consider a slow charge circuit.
There is a body of opionion that both the soft start and slow charge functions can be obtained with a Power Thermistor based circuit in the Primary circuit.
 
Right, I see. Is that basically until it's developed a field you don't really get any back emf and so appears essentially shorted?
Presumably, if you were to put something on the primary to limit current it would have some limiting effect on the secondary. With a typical set of values, would it make much difference to the rate the caps charge at, given an ntc maybe 10 ohms when cold, that limits the inrush to 24amps, the caps are still going to charge at whatever rate they feel like.
 
I'm scratching my head a bit here. If my cap has an esr of 22mohm, does that really mean that with a 24v rail it will try to draw at 1091A until its charged? Presumably here, the dc resistance of the secondary would actually limit this. I suspect that this is actually quite a complex thing to work out as there's so much changing at power up.
What happens on the secondary as you turn on and get the massive current on the primary before the fields have stabilised?
 
10ohms of soft start has very little effect on a 240Vac mains supply.
The peak starting current could be as high as 240Vac/10ohms = 24Aac or 340Vpk/10ohms = 34Apk
The CL60 works well for 110/120Vac mains supplies, but expect at least double that for the harmonised mains supplies on this side of the Atlantic.
I typically use between 30ohms (1000VA) and 140ohms (160VA) as soft start depending on transformer VA & type.
 
The Power Thermistor manufacturers publish application notes describing the method of determining the Thermistor resistance value and Power rating for combinations of voltage and capacitance. All of these slow charging Power Thermistors are in the capacitor circuit.
I don't believe that placing the Power thermistor on the primary side with the capacitance on the secondary side behaves as required by the manufacturers' method.

I have never seen a app note example for a slow charge circuit separated from it's capacitance by a mains transformer. I too am not man enough to think through the modelling required to predict how this behaves.
 
I have a bunch of these AC interference caps, but their capacitance is a bit higher than the .0033 microfarad units recommended for the general power supply, and they are polypropylene instead of ceramic. Can I use them in the 110 volt power supply for my F5? These X2 units are rated at 275 vac

https://www.conrad.com/medias/global/ce/4000_4999/4500/4500/4502/450239_LB_00_FB.EPS_1000.jpg


Thank you for any guidance you can give.

Larry Wright
"Zapped" (but looking to avoid the situation in the future)

(I tried to directly post the photo here but it did not seem to paste in properly. Maybe too high a resolution of the photo. Dunno)
 
Hi all. I've one of these excellent looking PCB's coming my way with a set of AlephJ class A amplifier PCBs and matched mosfets, jfets.
This will be my first amplifier build. I'm a mech eng with only a bit of electrical/electronics knowledge so need a bit of your practical advice wrt parts selection. I'm trying to learn more so if you can provide a bit more than yes/no answers to my questions below that will be really appreciated.
I'd like to be a bit future proof with my PSU. Build it so I can try different class A amplifier cards down the track (around the same 25W rating).
1. Is there any benefit or need for me to add some of the optional 0.47R PI resistors for the AlephJ?
2. Based on my reading through this thread so far, input and output snubbers appear unnecessary for a class A amp in the first instance. I assume this could be added later if noise becomes apparent?
3. I'd like some capacitor make/model recommendations please... I realise this question is a bit of a can of worms:) I think using 8 x 15kuF caps will be more than I'm likely to need? Will 25V rating cover me or should I go for 50V caps to cover other ~25W Pass amp options? Happy to pay a small premium for longer life. Happy to pay a small premium for "better sounding" audio grade but I'm not sure if I'll be able to hear the differences. ie. Looking for good sound, good life, good value.