I've retried the circuit on the breadboard, all new parts; 28V supply, L7812CV regulator, heat sink and 2 capacitors. The GPS device is still restarting within a few seconds and I am getting 11.75V across the device now.

I can't seem to wrap my head around why this occurs, might just have to stick with the double regulator design seeing as that works. Maybe if I get another batch in I'll try a different set of 12V regulators from another supplier.

Thank you all for your inputs. This has been most insightful. If I ever have more questions I would be more than happy to see replies from any of you again. Wishing you all the merriest Christmas and a happy new year.

 

Have you tried an LM317? There might be nothing in it, but I always thought it was better than the 7805 for difficult loads.

 

There might be nothing in it

Nah. That would be typical of a fake IC from a dubious source.

 

Have you tried an LM317? There might be nothing in it, but I always thought it was better than the 7805 for difficult loads.

I considered adjustable regulators but with the resistors and capacitors, the price per GPS device would have been higher than my initial design. Seeing as I'm now using two voltage regulators, it may work out cheaper to use an LM317. Thanks for the suggestion!

 

I've retried the circuit on the breadboard, all new parts; 28V supply, L7812CV regulator, heat sink and 2 capacitors. The GPS device is still restarting within a few seconds and I am getting 11.75V across the device now.

If I saw this I would question the grounding and whether or not a huge capacitor (470 uf or larer) on the input and output of the L7812 would help things. I have not theory about this but that is what my immediate action would be. Take note that a diode across the regulator to dump the output capacitor and protect the internal pass transistor once the external supply is switched off is a good idea.

 

I considered adjustable regulators but with the resistors and capacitors, the price per GPS device would have been higher than my initial design. Seeing as I'm now using two voltage regulators, it may work out cheaper to use an LM317. Thanks for the suggestion!

Did you try this new build with the resistor load before you put the GPS module on it ?
Was the output voltage stable for all input voltages ?

Once you have done that,
put a larger load on , to take 100 mA,
is the output of the LDO stable for all input voltages ?

Once you re confident in the LDO,
then add the more expensive GPS module in place of the load,

Don't risk the GPS for the sake of a few dollars of LDO