I should know the answer to this (Adj reg question)

[Speakerguy]

OK, I am looking at using an adjustable reg (LT1086 variant) in an application. It says max 30V input for one I am looking at, but for an adjustable regulator, since it is not ground referenced, this is just the input-output differential, right? So if I have it set to output 15V, I can feed it up to 45V at the reg input, correct?

I am just trying to take into account transformer regulation at light loads so I don't do anything that would damage the part. Thanks for any help!

 

[MikeMl]

Yeah, but you have to consider what happens during start-up. If there is a large output capacitor, there could be a transient condition where the voltage across the regulator exceeds the rating....

 

[dknguyen]

The LT1086 datasheet has this footnote for the 30V input voltage absolute rating:

Although the device’s maximum operating voltage is limited, (18V for a
2.85V device, 20V for a 5V device, and 25V for adjustable and12V devices) the
devices are guaranteed to withstand transient input voltages up to 30V. For
input voltages greater than the maximum operating input voltage some
degradation of specifications will occur. For fixed voltage devices operating at
input/output voltage differentials greater than 15V, a minimum external load
of 5mA is required to maintain regulation.

The datasheet actually lists the 30V as a transient absolute max rating and the absolute max imput operating voltage for the adjustable regulators is 25V is listed right after. But this got my curious and I checked the the LM317 datasheet- oddly enough it lists no similar rating.


...no explanations from me since I don't knowo why the LT1086 has the rating but the LM317 does not.

 

[Speakerguy]

Yeah, but you have to consider what happens during start-up. If there is a large output capacitor, there could be a transient condition where the voltage across the regulator exceeds the rating....

I will run some simulations, but the bridge rectifier + power supply filter caps will be charging up during this time as well. The 1086 regulator output should come up once it's input is >16V (due to ~1V dropout and 15V output set point) and charge it's 10uf-22uf tantalum output capacitor faster than the bulk capacitance after the bridge rectifier gets to its max voltage. I'll sim it but I think you are correct that the transient situation on startup is the only concern.

 

[Speakerguy]

The LT1086 datasheet has this footnote for the 30V input voltage absolute rating:



The datasheet actually lists the 30V as a transient absolute max rating and the absolute max imput operating voltage for the adjustable regulators is 25V is listed right after. But this got my curious and I checked the the LM317 datasheet- oddly enough it lists no similar rating.


...no explanations from me since I don't knowo why the LT1086 has the rating but the LM317 does not.

I am using a 1086 variant from a different manufacturer who lists 30V instead of the 25V of the Linear Tech regulator. They must be using a slightly different process....

 

[Speakerguy]

Simulation revealed that everything is fine, but that bypassing the ADJ terminal with a cap to reduce ripple slowed regulator turn on enough for some values of the cap to cause a problem. I'm not bypassing the adj terminal in my design, so no worries. Thanks!