88Mhz ripple on the Vds causes radiated emissions
Hı I have designed a 72 W power supply but there is radiated emission issue.. my power supply exceeds emission levels between 70-300Mhz..
In order to understand the reason, I checked the turn off characteristic o f my switch and I see the attached image...
During turn of instant there is a 50 Mhz high frequency ripple on DS voltage, (It is different from ossilation waveform due to resonance between leakage and capacitance of switch)
What can be the reason for this?
Comments
Dear Worf,
I did not add R between Diode and R snubber, but I will try,
But I could not understand how this will effect 88Mhz ripple,
Best regards,
sorry I added the wrong file, snubber schematic is now attached,
best regards,
Dear Worf,
I connect serial 5.6R between RC and Diode on the snubber but it does not make any change,
I also add serial Ferrite bead between Drain and Primary of transformer,
For the setup, still I see the same ripple on the DS,
Best regards,
Your DRAIN Voltage waveform does not look healthy. Something is fishi. Are you following the recomendatios for the PCB latout included in the data sheet of the controller?
What is your Transformer leakage inductance and your primary inductance values? can you attach a waveform of the DRAIn current?
Dear Worf,
Yes drain current looks like unhealty, ı see large negative currents on the mosfet, can this be normal due to anti parallel diode?
My Lp is 341 uH and Leakage is max 10uh( taken from the manufacturer)
I attached the waveforms, one point is when I disconnect voltage probe , signals seems a bit healty,
Talking about your last pictures. The Blue trace is not the DRAIN current. You are measuring something else but it is not the current going inside the DRAIN pin of the controller. . You need to cut the DRAIN trace connecting the DRAIN pin of the controller and put a jumper wire. One side is the wire is connected to the DRAIN pin of the controller and the other side of the wire is the connected to the rest of the DRAIN circuit.. Then you hook your current probe to the wire
Talking about your first picture (top left side) I see the voltage waveform is in 50V/div scale. Is this scale right? What is your input voltage in this conditions?
Did you follow the PCB design recommendations posted in the datasheet of the TOPswitch? It is very very very important to follow the recommendations. You need to have very short traces connecting the noisy nodes like the DRAIN circuit.
yes it does not like drain current but I cut the drain input of the swithc and connect a cable directly betweeen drain and trace so may be there must be a something wrong, may be current probe does not work properly.. I will check tomorrow..
yes you are right scale is not 50 V it must be 100 V, this is because my voltage probe has x200 scale but osiloskop does not support this so I use x100 for scope so the values are half of the real ones but current must be true... Real Input is 85 V so graph is true..
I follow the PCB design recommendation... I also make two other designs with top swtitch for 15W and 36 W, there is not any problem on these designs.. so this is why strange for me... I also changed the layout many times, I changec the x1Y1 caps value and location, and also the input filter, but these does not help me.. so there must be some problem with schematic, snubber, diodes etc...
Thanks for your help, I will try to change snubber values or the output shottky diode with UF diode.. May be something will change..
Thie ringing is due to the parasitic componets of your circuit. may be possible to reduce them. Can you please post your schematic? I would like to see your DRAIN snubber circuit. The very firt thing I ussually do is to add a resistor in series with the blocking diode connection the DRAIN to the clap circuit.