Buck regulator

I am having an issue uploading the noise image. I upload it but it doesn't seem to be coming up as an option

I have an issue with noise using a TPS5420 buck regulator. The attached image shows the noise zoomed in. That high peak has a frequency of about 50Mhz . How would I be able to reduce it down as much as possible. My schematic is also attached. The design is taking from the data sheet and the online calculator Texas Instruments do. I have tried putting in extra capacitors but it doesn't seem to do much good.

http://www.ti.com/product/tps5420

Thanks for the good documentation. First you have to make sure the noise is really there. If you are using a scope probe the ground lead is likely picking up radiated noise - try a direct coaxial (think RF) connection to the output. a BNC conector to some 50 ohm coax the other end soldered to the output with very short leads. then see if the noise is still there. Now - if you still have noise there (and how much are you reading?) then add a pi filter on the output with an inductor (sized for the DC current you expect) and good low ESR caps. You may have to adjust the compensation components to take into account the addition of the filter. What is the voltage, current required and what level of output noise you are looking to achieve?

Thanks and let me know how I can help

Also, along with the ESR, pay attention to the resonate frequency of the capacitors. If there is a 50MHz noise component, then your capacitors need to have a resonate frequency (or _self-resonant frequency_) at least twice as high.

If you go to our Power Supply Design Center, there is an article about good probing to eliminate the noise that is not "real".

www.ridleyengineering.com/design-center

Article 12 is about power supply noise and the issues Mr. Parmenter is referring to above.