SSLV1.1 builds & fairy tales

[schubert]

Here's some pics of my B1 before and after installing the SSLV, and a closeup of one of the Jupiter output caps.

 

[Salas]

Very classic Scandinavian like box. I like.

 

[Salas]

which Is that jupiter

which mkp do you recommend to use with BIB

That Jupiter is their current brown skinned higher temp line. Use something like Multicap in the Zobel.

 

[jimchan]

Dual Polarity Salas

I have been trying to persuade the Group Buyers to adopt a single PCB for both the -ve and +ve regulated supplies.
With care, any of the existing PCBs can be converted to the other polarity.

I would rather buy all the PCBs identically and use the build manual to guide me to achieving which polarity I require.

But even that complication is not always required.

Simply build all -ve regulators and use a dual secondary to power the regs. This gives an equal or better performance Dual polarity regulated supply than having to buy specialised +ve & -ve PCBs and having the problem of "left overs".

Hi Andrew / others,

Newbie here : I am building two Salas boards to create -12v 0v +12v output to power an Opamp.

Can I 'just' connect the 0v from the two boards to create a reference 0v ? or will I get current drawn from one to the other (ie: short circuit situation).

I have only single transformer 15 output to feed both boards.

Thx
Jim.

 

[Marra]

Its all in here

https://docs.google.com/document/d/...U6YYamCqlmk/edit?hl=en&authkey=CM7d-IkL&pli=1

 

[janos_904]

Hi, I kept building 3 boards in a row . My boards power TP's Opus DAC and the transceiver board. As a Vref cap, Oscon was replaced soon. Weren't a good choice there. SSLV1.1 works well, I like it very much! C105, C103 caps are Nichicon Muse KZ's: 1000 uF/25 V and 47 uF/25 V.

The schematic is OK. Shows an optional Zener also that may had tricked you. Objectively yes, OSCON is a good filter. Subjectively not my preference there.

 

[Salas]

Final success PSU+client application system picture we like to see as always.

 

[jimchan]

Thanks Marra,

My transformer is dual secondary, feeding 15v to each Salas board. I get nice 13v output from each board.

To make a dual polarity PS, I assume the 0v from each board should be made common (?). Should the common 0v also be grounded ?

 

[Salas]

If your client circuit board has a common point, there they go the F0 S0 wires from both boards and they make a symmetric system. Grounding to chassis or not its a matter of how your design is sorted out and what is recommended, i.e floating or not there.

 

[jimchan]

Thanks Salas, have got it all up and going now, powering a OPA827.

Real audible improvement over the 7812/7912.

 

[janos_904]

I promise you to take some pictures soon. So far I have too many LDO based (useless ) power boards close to the DAC. Also, the PLL (3,3 V) of Wolfson WM8804 , AVDD (+5 V), and AVDD(L+R) (+5 V) of WM8741 chip are powered by NewClassD regs. My BIB boards work as pre-regs (providing 1x 5,5 V, and 2x 7.0 V). Directly connected Salas BIBS are my goal , but there are 100 nF C0G/NP0 Ceramic caps at the legs of power input of the chips. So, any advice are welcome to plan the correct output of BIB!

Final success PSU+client application system picture we like to see as always.

 

[jimchan]

R Type or Torroidal

Any suggestions about the 'best' transformer to power the Salas when used for an Op amp or for DAC chip?

Currently using a single R Type for both, but will probably add a second transformer and more Salas.

 

[jimchan]

I promise you to take some pictures soon. So far I have too many LDO based (useless ) power boards close to the DAC. Also, the PLL (3,3 V) of Wolfson WM8804 , AVDD (+5 V), and AVDD(L+R) (+5 V) of WM8741 chip are powered by NewClassD regs. My BIB boards work as pre-regs (providing 1x 5,5 V, and 2x 7.0 V). Directly connected Salas BIBS are my goal , but there are 100 nF C0G/NP0 Ceramic caps at the legs of power input of the chips. So, any advice are welcome to plan the correct output of BIB!

I have similar considerations - I have 100uF 35v Cerafine at the legs of the power input of the OPA827. I guess these are not necessary now the shunt has replaced the regulator (?) I also assume the 'sense/feed' line should be connected as close as possible to the chip power inputs (?)

I will remove these caps tomorrow and inform any difference. Oh, and maybe send a pikkie

 

[Salas]

Those have to do with local needs of the chips in the context of layout, original supply characteristics and general noise coupling if measured and analyzed best during system design. It takes objective and subjective tests to see for potential changes & differences at best one's sense and measurement gear / ability. Closest sensing helps the supply read the dynamic consumption better.

 

[Salas]

Could be difference in consumption per polarity, some difference in CCS setting and/or actual value tolerance in the reservoir caps.

"Go Advanced" button under message window and scroll down to attachments manager menu to upload your pics.

 

[jimchan]

Shunts mounted on the heatsink of a donor amp. Powering a OPA827 of a DAC. DAC not final mounted yet.

Salas, can you explain to me why the LEDs of the +ve board turn off quicker than the -ve on power off ?

 

[jimchan]

The +ve LEDs also extinguish much quicker (ie: immediately) when no load attached. -ve rail takes 3-4 seconds.

 

[Salas]

What is your CCS current on +Ve and -Ve will go a long way explaining why -Ve has more energy reserve when powering it down.

P.S. Always use some kind of load, its safer.

 

[jimchan]

I have 1v across 5ohm r101 and 201 : 200ma

Noted re: load.

 

[Salas]

Maybe something with the symmetry of the transformer or the reservoirs at power off tilting the burden to one force side. But it does not seem in error when working. When in no load are the force and sense properly tied to each respective one?