diyAudio Power Supply Circuit Board v3 illustrated build guide

PSU decision

Hello!

I'm a novice DIYer. I want to build either an M2x or an F4 + suitable front end. But first I want to get the power supply right.

While the build guides use a regulated supply, several conversations in the forum threads revolve around capacitance multipliers, and - please correct me if I'm wrong - some designs use both methods..?

Mr. Pass' commentary on power supplies (link) has the following paragraph in the section titled Regulation:
Another way to regulate is by using constant current sources, which feed the circuit a constant current that does not fluctuate with supply voltage. A good constant current source can improve regulation for low power front end circuits by a factor of 100, and combined with supply voltage regulation, gives really excellent performance at little cost.
Can anyone shed some light on this and/or point me in the direction of more relavent information?

Cheers,

-Confused.
 
My OCD kept me staring at it but I couldn't quite place what was wrong. So glad I caught it before powering anything up...
 

Attachments

  • IMG_3060.jpg
    IMG_3060.jpg
    1,004.2 KB · Views: 465
Forgot to bridge the ground planes?

I did exactly the same thing when building the power supply for my F6 and thankfully caught it too.

I think chanthing got it. I accidentally reversed polarity on one of the caps. Had a nagging feeling for several days something was off every time I was working on the amps then it finally sank it. I'm a bit stunned I made the mistake in the first place but we're all human after all.
 
Good morning all,
I just finished reading this entire thread. It is very enlightening.
I have a question about filter cap voltage rating. I retired this summer and finally have time to finish a HB started many years ago. My transformers are dual 45 volt. The question I have is will 80 volt rated caps be adequate or do I need to go with 100 volt.

Thanks for helping out,
Ralph
 
Hi Ralph, I think the most conservative design calculations would go like this:

  • Assume AC mains voltage can wander plus or minus 10% from typical value
  • Assume transformer secondary voltage is 45VAC at full load, 15% higher at no load
  • Assume spectacularly efficient rectifiers whose voltage drop is only 0.2 volts even at full load

Since the transformer's secondary voltage equals the primary voltage times the turns ratio, for every 1% increase of mains (primary) voltage you'll get exactly 1% increase of secondary voltage.

Thus the secondary voltage at no load can theoretically reach (45 VAC) * (115%) * (110%) = 56.9VAC

56.9VAC rms, converted to peak DC value, is 56.9 * sqrt(2) = 80.5V DC.

But 0.2V of this is dropped across the rectifiers. So the capacitors see 80.3V DC.

Which means that, in the worst possible scenario, caps rated 80V maximum will be in violation of the manufacturer's do-not-exceed spec. You can decide whether this is acceptable, or unacceptable, risk of fire and/or loudspeaker destruction. I myself would choose 100V capacitors and sleep soundly. But I am not you.
 
Thanks for the reply Mark,

I do not seem to find any 10000mf caps rated at 100v that will fit on the diyAudio power supply board.

My choices seem to be:
1) Use 80 v caps and hope the perfect storm never hits
2) Use much lower total capacitance than the 40k / rail I planned on
3) Don't use the boards at all and point to point the filter section

If I use 1.2v for the bridge drop I get just under 80v. I wonder what the ramifications of running the caps this close to the rating are?

Again thanks for the reply and any other thoughts are welcome.

Ralph
 
Thanks for the reply Mark,

I do not seem to find any 10000mf caps rated at 100v that will fit on the diyAudio power supply board.

Perhaps 10,000 microfarad (u) or (μ) is what you seek? Be sure to use the correct value in your search. 10 Farad caps will be rare or non-existent.

I happened to be on DigiKey, and these two are in stock in the USA.

Blocked

Blocked

There are likely more available from other manufacturers with slightly differing specs or through alternate sources. I did very quick search for 10,000 uF, 100V, <=35mm diameter, snap-in.

Hope you find what you seek.
 
Thanks for the reply,
The m instead of µ (forgot to use the option key with m)

And thanks for the link - for some reason these did not come up in my searching -

Has anyone tried to use this 22.5mm lead spacing on the DIY boards? Bend the leads. or solder on wire extensions?
 
Hi Ralph, I think the most conservative design calculations would go like this:

  • Assume AC mains voltage can wander plus or minus 10% from typical value
  • Assume transformer secondary voltage is 45VAC at full load, 15% higher at no load
  • Assume spectacularly efficient rectifiers whose voltage drop is only 0.2 volts even at full load

Since the transformer's secondary voltage equals the primary voltage times the turns ratio, for every 1% increase of mains (primary) voltage you'll get exactly 1% increase of secondary voltage.

Thus the secondary voltage at no load can theoretically reach (45 VAC) * (115%) * (110%) = 56.9VAC

56.9VAC rms, converted to peak DC value, is 56.9 * sqrt(2) = 80.5V DC.

But 0.2V of this is dropped across the rectifiers. So the capacitors see 80.3V DC.

Which means that, in the worst possible scenario, caps rated 80V maximum will be in violation of the manufacturer's do-not-exceed spec. You can decide whether this is acceptable, or unacceptable, risk of fire and/or loudspeaker destruction. I myself would choose 100V capacitors and sleep soundly. But I am not you.

Another thing to consider (as I mentioned in this thread long ago) is that the MTBF of an electrolytic goes up drastically if you derate working voltage, temperature and/or ripple current. Cornell Dubilier has a calculator here: Cornell Dubilier Plug-In Thermal / Life Calculator

Bottom line - it does not pay to run large electrolytics anywhere near their rated specs if you want long-term reliability.
 
Hello all,

I am just about to start my first project, a Pearl 2 using a Universal Power Supply from the DIYAudio shop which I've already purchased. I am a total beginner, therefore, this project will be a significant challenge.

Thanks to all the members here and especially 6L6 for the tremendous build guides and contributions.

In the best traditions of DIY audio, I plan to over-specify the build🙂 I have some questions some of which have been raised before but I am not 100% clear and others are new for me, I would be grateful if the group can help me with my thinking.

1). The infamous snubber caps and resistors CX1, CX2, CS1, CS2, R11 and R12 - from my understanding of earlier posts although better to include but it is also possible to leave these parts out?

2). I plan to use 6,800uf 50v caps, 27,200 uf per rail - would it be okay to use 10,000uf, is there any benefit, the costs are similar.

3). I understand from posts by 6L6 that I should remove all the 0.47R 3 watt resistors and replace with 2, 10 ohm 3 watt resistors, one on each rail? Have I understood this correctly?

4). I have seen Pass labs and other members use filtered IEC inputs on some preamps, I am planning to use a Schaffner FN9290 (Mouser 631-FN9290-4-06) 4 amp dual stage filter or alternatively a Schaffner FN9280 (631-FN9280-4-06) 4 amp single stage both are combined IEC inlet, switch, fuse and filter, is there any benefit of the dual stage over the single stage and should I be ordering the 2amp version instead of the 4 amp version?

5). From Wayne’s schematic, I need to add a noise suppression cap of 0.0033uf before the transformer, since I plan to use a Schaffer filter above is the suppression cap still required? If it is required, will this Kemet work Mouser ref: 80-PME271M433MR30?

6). The 200v bridge rectifier diode block (Vishay GBPC3502-E4/51) is out of stock at both Mouser and Digikey, I plan to use the same Vishay diode block but in a 400v version (GBPC3504-E4/51), any problems here?

7). Primary earth, grounding hardware to the chassis, can I use any nuts, bolts and washers that I can get hold off or is there a set of parts I can buy in a kit for these.

My plan is to get the power supply up and running first and then tackle the Pearl 2 Boards later.

Thanks in advance for your support!
 
Specifications for alternative toroidal tranformers

Hello all,
I'm still trying to understand transformer behavior. In searching for an alternative transformer for the Antek 3218 or 4218 (out of stock for a long time), what are the key specifications for the toroidal transformer? I understand rationale for the 2x18V secondaries. Is there a minimum or maximum current?



The Antek specs are 8.3A and 11.1A, respectively. Most of the other toroidals I have found with 18V secondaries have lower load VA and lower secondary currents (e.g. Lundberg Y23 series at 250VA, 2x18V and 6.3A). Are there particular minimums and maximums for the general Pass Labs DIY 25W amp builds?



Living in the US I understand the need for 115-120V input.



I assume there will also be discussion of frequency, impedance and other fine points, but I'm trying to focus in the immediate issues to finish my kits. Thanks