Building a symmetrical psu B1 buffer

[music2diy4]

Salas,

Thanks for looking at it.

I didn't think that the pin outs would be that big a deal with the 2SK170 use in the earlier Shunt PS drawings showing the Gate and Source pins shorted ( J210 could still be inserted the same manner, 2N5245 would have to "face" the opposite direction of the board layout).

But with the other adjustments you touched upon, it may not be worth the tweaking at this point.

David

[Ian444]

I did a bit more work on the compiled "build guide" the other day, a few additions and a fair bit of trimming. Any suggestions for improvements/alterations are welcome. Maybe it could be combined with the wonderful BOM, maybe not... Hope it helps...

[jean-paul]

And here the PDF with some typos changed.

[Salas]

Originally Posted by Ian444

I did a bit more work on the compiled "build guide" the other day, a few additions and a fair bit of trimming. Any suggestions for improvements/alterations are welcome. Maybe it could be combined with the wonderful BOM, maybe not... Hope it helps...

Originally Posted by jean-paul

And here the PDF with some typos changed.

[Salas]

Originally Posted by music2diy4

Salas,

Thanks for looking at it.

I didn't think that the pin outs would be that big a deal with the 2SK170 use in the earlier Shunt PS drawings showing the Gate and Source pins shorted ( J210 could still be inserted the same manner, 2N5245 would have to "face" the opposite direction of the board layout).

But with the other adjustments you touched upon, it may not be worth the tweaking at this point.

David

You can use alternatives in the led ccs and bjt collector ccs positions if you don't mind the pins hassle and figuring it out personally. Also you can see what j201 can do with what resistor as figured in my hypnotize as a psu only guide and get your desired voltages. But this is beyond the GB boards thing, its no practical alternative for general guiding and will create immense confusion in the threads. I.e. You are on your own if you fancy to modify. Those jfets are noisier than 2SKs non the less.

[grufti]

The one thing I've seen that ends up causing lots of grief for no good reason is missing jumpers. My build guide would start like this:

'Make sure you install all jumpers first.

In any build using 12V relays you need to install three jumpers. Install one jumper at "AC IN". Then find the white line right next to the two 220R resistors on the left output and insert a jumper across the two solder pads. Finally insert a jumper across the resistor pad marked "Note" in the 12 volt power supply for the relays.

In any build using 5V relays you need to install two jumpers. Install one jumper at "AC IN". Then find the white line right next to the two 220R resistors on the left output and insert a jumper across the two solder pads.'

[Ian444]

Hi Grufti,

Good point...it is difficult to know how much detail to put in, or what the experience level of all the builders will be. Usually I build with an overlay and a schematic, and enough info to know any special requirements, e.g. in this case there are component matching requirements. That's what I was aiming to fulfill with the brief build guide, just enough info and links to get the job done.

If you want to go into more detail, for sure, go ahead, and from the 2 paragraphs you wrote above it looks like you have a real talent for this! Feel free to write a more detailed guide if you wish, I'm sure it will be appreciated.

[wintermute]

Hi Salas,

Sorry if this has been asked before, but I'm wondering why the swap of the 1M and 220K compared to the original B1. ie the original has a 1M shunt resistor on input and 221K on output, but the schematic at the begining of this thread (not the hand drawn one) has 220K on input and 1M on output.

I was also curious about the power supply caps. Is the shunt reg so low in ripple you can get away with only 100uf in the supply and nothing locally on the B1? It's a dramatic difference to Nelson's brute force approach with 2 X 15,000 uF

Tony.

[Salas]

Has to do with my suggested log pot value and the way I want to subtly help its balance between channels. 1:10 //.

Yes its very low in ripple. Bigger caps across its Vout just ''slow'' it down. Of course there is enough rectification filtering at 4700uF.

Its interesting when you listen side by side to two so basically same designs but differently executed as a whole. There are so many PCB's given on the GB, that we will possibly get feedback by people having original B1s also. Will be educating on how people perceive or not little stuff in the grand scheme of things.

[AndrewT]

The Pass PSU is simply an RC filter.
The Cap is sized appropriately to the value of the R used. It is not brute force.
The regulated supply starts from a completely different specification, but ultimately tries to achieve as little noise as possible, much the same as the RC filter tried to do.