Hi
Have these 3 version of regulators.
Wich one will be best or suggestion to other.

I have 12-18 dc input and will need outputs at 8v 200mA and 5v 20mA.

 

use the circuit in pic3, with all the caps connected to input and outputs , assuming the input voltage is 18v, then you will be dropping 10v @ 200mA thats 2W so a descent heatsink is required.

 

None of the above.

Reason being there is no outputcap on the 5V output. These are very old designs and they tend to oscillate, especially with no or light loads.

The large input cap may or may not be needed depending on the input source.

The large output cap again may or may not be needed depending on the nature of the loads.

Since it is far easier to remove parts than to add them later plan forthem all, and if you wish later remove them one by one and see what happens.

As someone once said, "you can't put too much cap on a power regulator."

 

What do you think the 1N4148 on the input is there for?

If it is there for reverse polarity input protection, it needs to be rated for much higher current. The goal is blow an upstream fuse before the diode vaporizes. A 1N4148 will vaporize long before any fuse...

 

Input voltage is around 12-14v dc from maybe a 230v ac adapter
i will need 8v dc 200 mA and 5v 20mA for driving PIC 16F1509

Any scematics to a decent and simpel power regulator ?

 

None of the above.

Reason being there is no outputcap on the 5V output. These are very old designs and they tend to oscillate, especially with no or light loads.

The large input cap may or may not be needed depending on the input source.

The large output cap again may or may not be needed depending on the nature of the loads.

Since it is far easier to remove parts than to add them later plan forthem all, and if you wish later remove them one by one and see what happens.

As someone once said, "you can't put too much cap on a power regulator."

You can put too much cap on the output!

If it discharges backwards through a 78xx, the regulator is toast.

A lot of people piggy-back a 1N4001 pointing backwards so it doesn't conduct in normal operation, if the input cap drops faster than the output at power off, the diode bypasses the discharge current.