thanks for your all answers and replies. ı gained a lot of experience. ı am gratefull to all of you! ı have fixed.

 

What was the fix? Just tying the grounds properly? Or something else? A response could help others that come upon the thread even years down the road and is one of the reasons the forum is here.

 

What was the fix? Just tying the grounds properly? Or something else? A response could help others that come upon the thread even years down the road and is one of the reasons the forum is here.

Yes. Reason is just tying whole gnds together. Connecting arduino's gnd to battery's neg terminal

 

Why not forget the LM317 and instead copy the classic NPN transistor+zener diode regulator? But use the D/A output as the reference? Of course you will still need to have the D/A common tied to the external supply common.
Basicly you are trying to accomplish what is not possible, and so an explanation of the reason will probably result in a suggestion that can be achieved.

 

Basicly you are trying to accomplish what is not possible, and so an explanation of the reason will probably result in a suggestion that can be achieved.

Bill,
The TS has posted that it's now working OK, by taking the advice of others.

So that: 'what is not possible', is possible when the correct advice has been posted by other members who understand the problem.

E

 

Why not forget the LM317

Two reason not to: the current and over temperature protection makes it nearly bulletproof.

 

OK, But without making that connection that the TS finally made, it was not possible. I stand corrected.
It did take the explanations of others to convince the TS to finally make that connection.
Without that is was never going to work.

 

It was also not possible if the DAC could not sink current, as the TS said. Fortunately, the datasheet disagreed with him and it can sink up to 25mA. 12.5 being needed in this case to make the drop across the 100R be 1.25V.