LTC3308A spikes on output

LTC3308A.asc

Hi,

This simulation does reflect the inductance in the output capacitor which is realistic. The inductance in the PCB and output capacitors will cause HF spikes to be observed on the output. On our eval board X2Y 3-terminal capacitors with very low inductance were added close to the inductor and PGND as close to the input caps as possible to minimize the inductance in the return path. This helps greatly in reducing the HF spikes on the output. It was observed that larger capacitor packages have higher inductance. Using low capacitance caps in parallel with larger caps and a tight layout will help reduce the HF spikes.

Yes, we did get recommendation to use X2Y caps in the past so I will add them.

Thanks

I tried adding X2Y Caps to the simulation 2 each on the input and output but there is still 500mV pk-pk noise and high frequency ringing.

I measured the rise time of the switch node to be 100ps. This seems unrealistic and much faster than it should be.

Attached power point shows measured rise time to be 1.1ns. Also noise is only 4.4mv pk-pk and no ringing.

It seems Ltspice model for LTC3308A is not accurate and does not reflect reality.

5280.LTC3308A.asc

Hi zmicro ,

This is because, LTspice result is based on what the real value must be without all the filtering that is introduced by your measuring equipment which in this case is the oscilloscope.

Note that oscilloscope have limited bandwidth and its probes have filters. Thus, in real world you cannot observe this high frequency because your equipment cannot simply view them for you.

Regards,

PaulDaria