0
\$\begingroup\$

I am building and testing the transmitter side of a wireless charger.

The image below is a rough representation of what I am doing, U1 is powered by an IR2110 driver at 5kHz off a 555. V1 is 30VDC and the flyback diode is a 10 amp ultra fast recovery diode( cant remember exact model)

When measuring across the load on a scope I get a just the positive cycle of a sine wave ( so what a sine wave looks like after a rectifier), its voltage however, comes solely from the mosfet and if I am to increase and reduce the supply it does not effect the wave. I can't for the life of me figure out why?

Could this be to do with the internal diode inside the MOSFET? Would an H-bridge configuration work better in this situation?

Why does this not work? Because in my head the voltage across L1 should be drain signal plus V1 shouldn't it?

enter image description here

\$\endgroup\$
0
\$\begingroup\$

The diode (D1) is the problem - think about what is happening - you draw some current through the inductor then use the mosfet to "release the spring" - what should then happen is that the LC tuned circuit resonates and this would naturally produce a "recoil" voltage that rises up to twice the supply rail and down to nearly ground. The action of the capacitor restricts the peak voltage to 60V and pushes ALL that energy magnetically towards the load.

Having the diode there just returns most of that energy to heat in the diode. Try a simulation without the diode.

Think about it another way - the long term average voltage across an inductor must be zero (any scenario period) - this MUST mean the waveform rises above and below the Vcc rail. The IRF630 can handle 200V so this won't be a problem unless you mis-drive the MOSFET and take so much energy into the coil that the capacitor cannot adequately take away all that energy before hitting 200V. If you think this might happen use a 30V zener in series with the diode you currently have installed.

\$\endgroup\$
  • \$\begingroup\$ Thanks for the reply, your point about D1 makes sense as before I added it in the circuit did work, however with the pulsing MOSFET, when the iductor current collapsed it would flow back through the MOSFET (when it was open) and rapidly heat up and eventually blow the driver so I had to add it in as protection. Would there be another way of protecting the driver without D1? \$\endgroup\$ – Tom Hobart May 9 '16 at 17:56
  • \$\begingroup\$ The inductor current will flow through the 10uF capacitor unless you have it wired incorrectly. If you are trying to pulse too much current through the inductor the dv/dt change on the drain may force some current through the internal DG capacitance of the MOSFET - maybe remove D2 and, above all check the layout of components. \$\endgroup\$ – Andy aka May 9 '16 at 18:05
  • \$\begingroup\$ C1 is a PET film cap and have double checked all the wiring. I am using a 24VAC adjustable supply which goes through a rectifier and voltage reg so I am able to increment Vcc from 0-30V slowly and before I added D1 it would work up until around 40% power at which point it would collapse (while also heating up the driver) I will try removing D2 tomorrow to see if it helps at all. \$\endgroup\$ – Tom Hobart May 9 '16 at 18:27
  • \$\begingroup\$ What pulse width are you hitting the mosfet with and what amplitude. Also, what is the coil made from? \$\endgroup\$ – Andy aka May 9 '16 at 18:55
  • \$\begingroup\$ The driver is outputting 15V at 5kHz so 0.1ms pulse. Coil is enameled wire, 1.6mm or 14AWG, wound with a diameter of around 7cm \$\endgroup\$ – Tom Hobart May 9 '16 at 19:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.