Solution Finder Get Tech Support

dpa switch

Posted by: sheldon on

Hi,

Please find attached the schematic for dpaswitch, 12V/600mA deisgn,

Does this circuit work,

please tell mne modifications in this design

Files

Attachment Size
dpa power supply.jpg 61.46 KB

Comments

Submitted by PI - Traveler on 01/25/2012

Yes you could probably get something like this type of feedback circuit working. Off the top of my head, I can't say if you've chosen the right components and/or values. Modeling the circuit would probably be a good idea. If you build this circuit, I would definitely make sure to evaluate the control loop on the bench (don't just do calculations by hand). You'll want to analyze the control loop at different input and output conditions as well as over temperature.


On another note, you'll want to check and make sure that having a non-isolated design is acceptable in your application. With the way you have your feedback circuit, there is no safety isolation.

-The Traveler

Submitted by sheldon on 01/29/2012

Hi,

Thank You

would you provide me with the control loop calculations tips ,i.e,, formula.
or somethings which are more related to the feedback section,

Thanking You

Submitted by PI - Traveler on 01/31/2012

In reply to by PI_Crusher

I'm pretty bandwidth limited at the moment, so I don't have time to do all the hand calculations for you on this feedback circuit. I can however give you some suggestions.


- With the non-isolated design and lack of primary-side bias winding, you need to connect your output return to the source connection on the IC.
- R7 and C6 appear to be part of a phase-boost network. I would initially remove those and get the power supply up and running without them. You can add them later if you're having problems.
- Your starting values for R6 and R5 are going to need to be larger. Likely in the 500-1k ohm territory. You want to bias the TL431 at approximately 1 mA when the output is at 12V in steady state conditions.
- Since this control loop is much different than what PI Expert can model, you will absolutely have to measure/analyze your power supply control loop on the bench. Expect that there will definitely be some fine-tuning of resister values to get the proper operation you're looking for. Doing this will also let you know if you're going to end up needing the phase-boost network around the PNP transistor.
- If you don't have access to the equipment necessary to analyze the gain/phase of the control loop in your power supply, I would suggest stopping where you are and changing the design to something that PI Expert supports.


I'm sorry I don't have the time to figure out specific component values for your design. If I can help with other aspects, please let me know.


-The Traveler