TOP250 Opto Feedback
We are using the TOP250 in a powersupply. We have a feedback circuit with a opto and a TL431. Is it possible to power the opto with an external powersupply? Or should the opto (with gain resistor) always be connected to the power output, for the regulation loop?
I've done some tests with this and i notice that the output voltage, when i power the optocoupler from an external supply rises up to much higher levels (19V) than it should be (normal about 7V when powered as prescribed). I just can't find the reason why there could be any difference in the source of the voltage for the opto, the opto has to turn on when the voltage supplied to the ref pin of the TL431 reaches 2,5V, right?, which then interupts bias voltage and lowers the voltage.
In the case of a to high output voltage the primary transober broke down, indicating there must be a way to high primary voltage (or spike), this looks like that when using an external ps for the opto, no regulation is present.
Comments
Is this because the opto is missing the spikes / ripple in the output when powering externaly, thus giving bad regulation?
I forgot to mention that i do connect the ref pin of the TL431 to the output using a resistor divider, this do makes it (part) of the loop. But Still the opto has to be connected to the power output?
Brell -
The TL431 feedback topology has multiple different feedback paths and transfer functions. On page 7 of AN-32, they show the TL431/Opto feedback setup. The resister divider formed by R and the 10k resister make up the slower outer loop of the TL431. The feedback path created by the 470 ohm resister (connected before the post filter) forms the faster loop of the TL431. Both of these feedback paths need to be connected for the TL431 feedback circuit to function correctly.
The primary side of the opto is supplied power from the bias winding (the second output shown in figure 6 of AN-32). This is the side of the opto that you can power from an external DC supply. This can be useful for debugging purposes, especially for early prototypes where you might have some leakage inductance issues or coupling problems with the bias winding.
If you're having some problems with the initial design and want a simpler feedback topology, you can pull the TL431 out of your PCB and implement the feedback circuit shown in figure 5 of AN-32. This is another useful tool that can help simplify debugging of your power supply. It can be helpful to use a zener/opto feedback circuit initially to get your power supply up and running. Then, once other issues have been resolved, you can optimize the performance of the power supply by switching to the TL431 feedback circuit.
Here is the link for AN-32:
http://www.powerint.com/sites/default/files/product-docs/an32.pdf
A couple of other good references are from OnSemi and Switching Power Magazine:
http://www.onsemi.com/pub_link/Collateral/TND381-D.PDF
http://switchingpowermagazine.com/downloads/15%20Designing%20with%20the%20TL431.pdf
-The Traveler
If you're having problems with the bias supply in your design, you can power the primary side of the opto with an external power supply for debugging purposes. If you do this, make sure you're using an isolated AC source and a floating DC supply to avoid any ground loops.
On the secondary side, you need to power the opto from the output of the supply. If you power the secondary side of the opto from a DC supply, you've broken your control loop and your power supply will not regulate correctly.
-The Traveler