[En-Nut-Discussion] Ethernut 5.0 Schematic Preview

[Alain M.]

Harald Kipp escreveu:
> Alain M. wrote:
>> 1) have you considerd powering the ATMEGA with 3.3V
> 
> Then we either need another 3.3V regulator or the ATmega won't be able 
> to switch off the 3.3V supply. See below.

That would be a very smal regulator, probably sot-23, small and 
innexpensive...

>> 4) if you have a RTC chip, you should have a battery too... it could on 
>> the bottom side of the board so that it does not use any pcb area, or at 
>> least justa connector could be nice.
> 
> Mounting a battery on the bottom side would require manual soldering. 
> Beside that, the board may not fit in the same place as used for other 
> Ethernuts. C11 may be increased to 0.33F. As this is a through hole 
> part, you may attach an external battery (plus a diode) the the pads 
> instead.

0.33F are biiig, if you are to leave that possibility, the layout should 
take it into account. As for the battery, the PCB could be ready to 
receive a socket, even if you don't ship with it. As long as it is easy 
to purchase, it would be better than not having any.

>> 5) F1 could be a Polyswitch, more practical, uses much less board space.
> 
> This had been proposed several times for the other Ethernuts as well. 
> IMHO, a TVS in conjunction with a fast acting fuse is the safest way to 
> avoid all kind of problems. If the fuse blows, something went miserably 
> wrong and it's good to force someone to look to it before replacing the 
> fuse.

I am an early adopter of polyswitch and I love it. Small mistakes happen 
a lot in a lab...

>> 6) This is going to be a Develloper's tool, so IMHO no pin should be 
>> unreacheable (TST, SHDN, A18, A19, A20 maybe more...)
> 
> Mh, well, not really. The majority of Ethernuts are used in real 
> equipment.

I believe that it is used a lot as a development tool as well... You 
should not rule it out. But test points are good enough, specially if 
they are in such a position to form a connector...

Alain