Capacitor locating question

[Dxvideo]

Hi all,

A short question;
Is it important the position of the rezervoir cap on board?
Means;

In this PCB layout, do you think the places for CSN and CSN1 are suitable? Why I ask that? Because if you look closer, CSN is more closer to left side transistors while CSN1 is on right side.
What can you say about that?
PS: I have a 22000uF capacitor bank per rail before this board..

[jackinnj]

When using the LM4702/LMExxxx try to attain as much symmetry as possible -- the supply lines should be decoupled as close to the power pins of the driver chip as possible, and don't bother with a ground plane.

You can mount the decoupling capacitors on the bottom of the PCB -- this will give you a bit more free real estate on the top of the board. It's only important to have easy access to the potentiometers for bias adjustment.

[Dxvideo]

Hmmm...
I can cancel the ground plane, thats not a problem.
However in this positioning, do you think the power lines are not enoughly close to the LM4702s power pins?
In fact I was worrying about the output transistors, that theyre not enoughly close to power input points..
Thats strange, because I did not declare the circuits driver chip! And if you could have a correct guess about the driver IC then you will also know the other IC. Whats it doing in there?
Just for fun...

[peranders]

If you have a ground it is still important how the currents flow. It's especially important that the currents from the big smoothing caps don't interfere with the signal ground currents.

[pacificblue]

Not so difficult to know the chip. It's written on the PCB on the left side.

[Dxvideo]

Originally posted by peranders
If you have a ground it is still important how the currents flow. It's especially important that the currents from the big smoothing caps don't interfere with the signal ground currents.

Ok I will cancel the ground plane and make a star ground point there. With this way, I will also split the power and signal gnds.
And as you can see in the layout, my smooting caps GND line goes to GND point fully dedicated (via wires on bottom side). So I think they will not be a problem.
Do you agree?

Originally posted by pacificblue
Not so difficult to know the chip. It's written on the PCB on the left side.

So, make a guess, whats the IC (Dil8) bottom of LM4702? And what for? A tip: its not a preamp or buffer!

[pacificblue]

You see, what I mean? Much more difficult to know, if it's not written there. And I'm not such an expert for those things anyhow.
Looks like an op amp you use as DC servo. Which op amp? Have to guess. OPA2347?

[Dxvideo]

Right,
Its DC servo, but not an OPA... However youre lucky because it was also written on the left side!
Anyway,
I am working on a LM4702+SAP15+DC Servo project. I am planning to use LF412 (as National guys used for BPA200). I will inform you about the results..

[jackinnj]

Those SAP15, 16 transistors from Sanken have been replaced by the STD03N and STD03P -- the major difference being the elimination of the emitter degeneration resistor -- this resistor couldn't take the heat (so to speak). You can pull the emitter's output from the "Sense" pin and use an outboard degeneration resistor. There are still SAP transistors around, no reason not to use them as long as you are advised of this one device shortcoming.

[Dxvideo]

But in the SAP's datasheet, if you take consider the destruction point of the internal degeneration resistor then it seems enoughly strong..

But I can put an extra pinout for an external resistor. Youre right.