I'm trying to charge a capacitor from 20 uF a 500Vpp circa , in less than 1,5 seconds starting from a voltage of 12-15 Vdc . Once uploaded, I need to download it via short .
I managed to load it using the diagram I show you below .
The problem that arises is at the moment of the discharge of the capacitor C4 . The mos-fet Q1 remains in an anomalous condition that I show you via an uploaded image .
It does not get out of this condition until I turn off the power and then give it back again .
Maintaining the capacity of C3 at 22nF this anomalous condition does not present itself but , increasing this capacitance consequently reduces the capacitance reactance and therefore I can charge the capacitor C4 in the time required by me ; doing so though , the problem of the abnormal mos-fet condition comes up.
How could I solve this problem ?
Delete the 7808, it internally has a short-circuit protection that limits the current delivered to max 1.5A and you need at least six peak.
The limiting resistance must be much lower, I propose two resistances from 0,1 in parallel.
The 22nF capacitance used as the electrolytic charge current limiter seems excessive at the switching frequency. You need a reactance of about 10k to have a complete recharge in about a second without excessive current peaks.
One thing you haven't specified is whether the coil core can withstand current spikes of at least 6A without saturating.
For proof I deleted the 7808 and short-circuited the current limiting resistor that is R1.
With 22nF at a frequency of about 45Khz I have a reactance of 160 ohm if I'm not mistaken .
The above problem occurs if I increase the capacity of the 22 nF .
I needed to increase the capacitance of C3 to charge the capacitor C4 in 1.5 S circa
Only IDC = data is shown on the coil date 2.2 (max) A