Solar capacitor charge converter with TL494?

[Artlav]

If you connect a supercapacitor bank to a solar panel directly, then it would charge, but veeerrryyyy sllooowwlllyy, since the panel only gives that much current.
It's maximum power is only available at it's optimal voltage.
To keep the panel in it's maximum power point, i want to make a step-down converter that would raise voltage as the capacitors charge, without letting the panel's voltage drop too far.

The idea is straightforward - one TL494 error amp is wired to sense the input voltage, the other one is disabled, and the rest is as per datasheet for single-ended operation:

Panel is 17Vmp, 0.6A max current, converter works at 66KHz.

There are two problems, however.
One is that when i connect an empty supercap, it either works quietly or inductor starts whining. There is no noticeable difference in charge rate, but that does not sound good, no pun intended.
Is it because the thing is on a breadboard, or is there a design flaw that makes something unstable?
I'm particularly uncertain about what i should have done with the comparator feedback pin, 3.

Second, as the output voltage approach the input, there will no longer be enough space for the MOSFET gate to turn it on completely, even for the lowest Vgs ones.
I haven't ran into any issues due to that yet, but is it possible that it would enter linear mode, and would just waste power when the load is below the panel's power capacity and no voltage drop is needed?

The rationale for the circuit is a solar phone charger with a supercapacitor buffer.

The converter goes in between the panel and the capacitors.

Without the converter everything works fine until a cloud covers the sun for long enough to discharge the capacitors below 10-12V.
After that, the panel can't recover - forced out of it's max power range, it can't supply enough power to charge both the capacitors and the phone, and the voltage keeps dropping until the charger shuts down.

With the converter i expect the charger to remain functional regardless of the capacitor's charge level, with only the excess power charging them.
Is that a viable solution?

 

[Artlav]

I just realized the existence of P-MOSFETs.

That solves the second problem with minimal changes.

However, with a P-MOSFET the whine became more pronounced, and won't go away at all.
Now it climbs in pitch as the capacitor charges, and everything surges until it reaches 0.8V or so, after what it just whines.

 

[Artlav]

I've put the thing on a proper PCB.
Now the surges are gone, and the whine is more or less a clear tone, slowly changing as the capacitor charge.

Is that normal, or is there a design flaw somewhere?
Apart from the whine it works nicely.

 

[FvM]

C2 adds a pole to the control loop that might be sufficient to create instable behaviour. In addition you should try to adjust the C1 + R5 error amplifier compensation.