Walt Jung shunt Design

The current source dissipates (10 - 5) * 0.13 = 0.65 watts. Its case rises to 58C which is 33C above ambient air. That means its thermal resistance from case to ambient air (theta_CA) equals (33 / 0.65) = 51C per watt. That's a pretty impressive transistor, since most TO-126 medium power transistors are rated 100C per watt.

{And if there's no load on the regulator output, the shunt device dissipates the same: (5 - 0) * 0.13 = 0.65 watts}
 
Last edited:
@Vunce
I finished the design, I preferred to put two pads in the case I would like to implement the cascode for the dn2540, so as to be able to experiment with new solutions with a small daughter pcb. I tested the operation for an output voltage of 5v and a current of about 80mA, everything works correctly, as soon as I complete the assembly of the pcb I will start preparing everything to be able to power the dac and see if everything is fine. I am attaching the eagle files if someone might need them.
Ad maiora.
 

Attachments

Here is the new UnivReg_122714 for 3.3V output voltage, still in testing and so far I am very satisfied.
 

Attachments

  • 3V3 wj shunt.jpg
    3V3 wj shunt.jpg
    184.5 KB · Views: 349
  • Like
Reactions: YashN
@Vunce
I finished the design, I preferred to put two pads in the case I would like to implement the cascode for the dn2540, so as to be able to experiment with new solutions with a small daughter pcb. I tested the operation for an output voltage of 5v and a current of about 80mA, everything works correctly, as soon as I complete the assembly of the pcb I will start preparing everything to be able to power the dac and see if everything is fine. I am attaching the eagle files if someone might need them.
Ad maiora.
Could you post your sch for + and - ckts?
 
@Tix88
Thanks for the additional clarity Tix88.
Given the various changes since 12/27/14 on this topic, I am reposting the design document of 12/27/14, as it was updated in 2018.

The PDF can be found at: https://refsnregs.waltjung.org/UnivReg_122714_121718c.pdf

Note that the RED level shift LED should run at high current as set by the 220ohm R3 (not 1k!).