diyAudio Power Supply Circuit Board v3 illustrated build guide

Rectifier Placement Question

I'm stuffing two PSU v3 boards following 6L6's excellent build guide. I'm now at the point of installing the rectifiers and heat sinks. I'm looking at the datasheet for the Schottky rectifiers (ON MBR20200CT) and see the pinout diagram. I notice there's a marking for polarity on the device (AKA, pin 3). I'm also seeing a small white dot on the leftmost pin on the board which would coincide with pin 1 when the rectifier is mounted to the heat sink and placed in position on the board. Unfortunately I'm not well versed enough in reading the schematic to confirm whether the pinout of the rectifier is correct when aligned with the three holes on the board. It looks like it should be (L to R) - pins 1,2,3. I suppose there's a possibility that pins 1 and 3 could flip, but that would be a bit tricky. I've read enough in these forums to know not to assume anything, so I'm asking for confirmation. Since 6L6 didn't use the bridge(s) in his build, I don't have any visual reference on this.

Also, what is the best method to approach this portion of the build. Mount the rectifier to the heat sink with insulator/grease, bend the pins to align with the holes, then solder, or mount the heat sink first?

Thanks in advance for your reply and many thanks to everyone for sharing their knowledge. You're teaching an old dog new tricks and I'm very grateful (much fun).

-Jeff
 
What is probably an incredibly dumb question from a newbie: what is the purpose of resistors R1-8 and their optional cousins? If for filtering why the decision to use a resistor capacitor filter rather than another type? (choke input etc)

I know these questions are outside the scope of a build guide but I'd appreciate an understanding of this. If too off-topic, perhaps a reply by PM?

Thanks in advance

Steve
 
Thanks Andrew. Would a different filter topology (choke input, Pi or other) offer advantages? (I recognise there are cost and size disadvantages.) I guess I am trying to understand if a "high-end" design would start with this topology or another.
 
A choke input filter has big advantages.
But it has a serious disadvantage.

When current draw goes low, the voltage rises by ~50%.
That rise could damage many components not designed for the 150% voltage.

I'm still struggling with this. I know the defining equation for an inductor is V=L.dI/dt and that inductors smooth voltage ripple by creating an EMF which "opposes" the change in voltage across the inductor. But if the above is true how can a CLC network be effective in providing stable voltage in a situation where load current isn't stable?
 
In the spirit of DIY, I took a chance and pressed on with my PSU build despite uncertainty regarding the rectifier pin orientation. In the hope of helping another beginner like myself I'd like to share what I found.

In the three holes just in front of the heat sinks I put the three leads from the rectifier - Pin 1,2,3 (L to R), and soldered them to the board. I temporarily connected the 2 pairs of transformer secondaries to the board and was pleased to see the bulb tester light, then dim as the capacitors charged. All is good and I have 42.5V DC on the output from 30V secondaries. :D

I realize for most here, the polarity of the rectifiers was obvious, but it wasn't for for me, particularly in regards to the marked polarity on the device (pin 3) and how it relates to the white dot on the board (leftmost hole) since they don't match up. I'd still like to know why that is.

-Jeff
 
diyAudio Editor
Joined 2001
Paid Member
The PSU boards were designed for maximum versatility. Yes they would be simpler if they could only be hooked up one way, but instead they are more of a Swiss Army Knife type thing, which give people building many different kinds and sizes of of amps workable options. Also, like everything in audio, people have different beliefs about how to best implement a PSU and the options they desire. Of course this makes things a bit more complicated for the newbies but you really can't have both.

Since no parts actually ship with the PSU boards, we want to show the various options. One option is to use the monolithic bridges. But people have different ideas of what are the "best" rectifier diodes and the discrete diodes allow more choice of diode type. Most of the monolithic bridges are more generic, offering less choices of diode types. I've used them lots though, they are convenient to install, generally cheap, and the differences in diodes are subtle IMHO.


A bit confused about these monolithic bridges - why show them if they don't ship with the PCB [edit - by which I mean, they are not part of the BOM for the power supply - you provide the bridge rectifier PCBs instead] ? Should I abandon the diyaudio rectifier PCBs and get these instead? They certainly look easier to work with than all the heat sinking, bolting etc (although admittedly still require heat sinking which you are achieving with the chassis in your pic).

As this is the build guide for the power supply that you guys sell, introducing new components that are not part of that kit / BOM is a bit counter-intuitive to me!

A final comment - from memory mine were shipped with the 2 filter PCBs already separated from each other, so the stuff in the guide above about not separating them for a lower noise PCB is not very helpful ;)
 
Last edited:
  • Like
Reactions: 1 user
I noticed that in this build guide no snubbers have been used. I tried going thru the articles on the use of snubbers and it left me confused. I read somewhere that there may be no real need to use snubbers if using Hifred diodes. Is this correct and is this why no snubber caps and resistors have been used in this guide?
Also in NP's article on F5 Turbo Power Supply there is a .003uf cap across the transformer primaries. Does one need to use this even if one is using a softstart circuit?
 
The PSU can ring if the combination of capacitance and inductance and resistance allow a fast transient to give it a kick.

Many PSUs either don't ring, or do so only a little, or infrequently, that extra snubbing seems to be not required.

If there is no fast kick then there is no ringing.
 

6L6

Moderator
Joined 2010
Paid Member
The 0.003uF cap is helpful for AC filtering in all builds, soft-start or not.

Class-A amplifier PSU should have little or no transient loads causing the transformer to ring, so a snubber is not strictly necessary. Andrew is spot on.
 
The 0.003uF cap is helpful for AC filtering in all builds, soft-start or not.

Class-A amplifier PSU should have little or no transient loads causing the transformer to ring, so a snubber is not strictly necessary. Andrew is spot on.

Thanks Andrew and 6L6. I think what you are suggesting is that for our F5V3 application I would be likely better off not using snubber caps and resistors and just use the MUR3020W 's.
Is this 505-MKS2.1/63/5 Panasonic 3300pf 250V X2/Y2 cap acceptable for use across the transformer primaries?
 
Member
Joined 2011
Paid Member
What about [LC resonant ringing stimulated by] the rectifier diodes? (Trying to relate this discussion to the Quasimodo thread. Does the PCB provide for ready implementation of a CRC snubber across the secondaries?)
I could answer your question. But I won't; you will become more self reliant if you learn to answer it yourself. The tools are at your fingertips: red colour
 

Attachments

  • rectifier_induced_transformer_ringing.png
    rectifier_induced_transformer_ringing.png
    109.4 KB · Views: 4,036

6L6

Moderator
Joined 2010
Paid Member
No thermistors between the amp boards and PSU boards. That will hum like a nest of angry wasps.

You can use the thermistor in place of the 10R resistor across your ground-breaker bridge.

You may want to try bringing the input ground to the PSU and/or the speaker ground to the amp boards, one of those combinations will be the quietest.