LM2051 not working properly. Help!

Verify that HF transformer's windings, (specially the one used for normal operation IC power supplying) are connected in the correct polarities. I'll bet you use at least one winding reversed. And do not mean after rectification, but the winding itself. ALL transformer polarities for this topology are NOT reversible, because of the flux/time figure asymmetry. Notice the point signs of transformer on reference schematics. They're there for a reason. Not that difficult. S.M.

I cannot read component codes and values on circuit, definition not good enough.

1. You seem to be running out of power to the control IC.
2. It is fed by a resistor from the rectified DC and by a rectified supply via a diode from the Transfo.
3. Since the circuit does kick into life, the resistor is doing what it should. But it does not give enough current to meet requirements of continuous operation.
4. So something must be wrong with the diode feed from the transfo?
5. Or your transfo is not right and the overcurrent limit is causing shutdown?
6. But 4. seems most probable. I suggest taking the diode feed off the Vcc pin of the controller and connecting dry cells [ of voltage just below the normal supply level] to the controller common and Vcc.
7. If the kit now works OK then do 8.
8. Load up the diode from the transfo with a capacitor (1 microfarad?) and resistance load equal to that of the controller datasheet maximum. This should show if the diode power path to the controller is good. voltage of the controller] to the Vcc pin and common.

I cannot read the fine details of your schematic. So I do not understand a few things about this circuit topology. On the primary side of your transformer it appears that your drive circuitry is only driving half of your primary winding. It would make better sense to me if you actually have two sets of windings on the left side of the transformer with one winding generating the power for the LM2051. Your circuit output topology is configured to generate two positive output voltages and not the one single 15V output voltage you refer to in your text. Now only one output voltage is regulated via the opto-coupler. As for the output, I usually see some series inductors on the output that smooth the output voltage of switching spikes. (A poor description but I'm not going to attempt fabricating the math from blurred component values.) Now many times an SCR triggered feedback path (your feedback circuitry) uses these spikes to trigger the SCR. Your lack of a Π (pi) L/C filter network on the output may prevent the SCR from being re-triggered on each switch cycle.

Lastly, a common mistake that people make in copying a switching supply circuit is getting the transformer windings and core dimensions wrong. This could also be why you cannot generate sufficient current.

Which application note are you using?

Can you at least show us a blown up copy of the area around the transformer.

Which one of the outputs is the 15V? If you look at this picture from the data sheet you will see that each of the secondaries has its own GND connection:-

You seem to be trying to take two outputs from the same secondary: may be OK, but needs some clarification.

Geez guys..

Thanks for the help.. Found the problem... The flyback transformer used is not really meant for the application.

Thanks!