The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> noise requirement for VDD of a RF power amplifier
https://designers-guide.org/forum/YaBB.pl?num=1157350535

Message started by richard88 on Sep 3^rd, 2006, 11:15pm

Title: noise requirement for VDD of a RF power amplifier
Post by richard88 on Sep 3^rd, 2006, 11:15pm

Hi,
When designing a LDO to serve as power supply for a RF PA, I came across different datasheets : some require integrated noise e.g. 30uVrms from 10Hz to 100kHz. I'm wondering why this bandwidth 10Hz to 100kHz is of interest for PA ? I thought RF PA would be operating at much higher frequency ?

Thanks,
Richard

Title: Re: noise requirement for VDD of a RF power amplif
Post by Ken Kundert on Sep 3^rd, 2006, 11:45pm

At the transmit frequencies, noise from the supplies would be blocked by passive bypass and decoupling networks. However, noise at low frequencies could get by the passive noise reduction networks and mix with the carrier in the transmitter and be translated up to the transmit band.

-Ken

Title: Re: noise requirement for VDD of a RF power amplif
Post by richard88 on Sep 4^th, 2006, 10:47am

Ken,
Thank you very much for the explanation.
I wonder if the passive decoupling network comes from the RF PA or merely due to the output cap of the LDO ?
The closed loop transfer function of the LDO with the output cap shaped like a low pass filter and also shaped the output equivalent noise unless something goes wrong like low phase margin. But I'm not sure if this low pass function is enough.
Also, I have seen some DCDC switching regulator powering RF PA. Using pwm (pulse-width modulation) switching, there are harmonics tones at mulitple of switching frequencies. I suspect there is an "internal noise passive decoupling network" in RF PA to filter off this harmonics, not sure if the speculation is correct.

Thanks,
Richard

Title: Re: noise requirement for VDD of a RF power amplif
Post by loose-electron on Sep 4^th, 2006, 1:33pm

Agree with Ken on the up conversion of noise -

Another additional item - the LDO noise content will come in two different forms, depending on the LDO - A bipolar LDO will have much less LF noise (flicker content is much smaller) and the CMOS LDO will have a good amount of flicker noise under 5KHz that needs to be dealt with using local decoupling.

When you refere to a "RF power Amp" - if this is the output PA of a cell phone, these are typically connected straight across the battery, and not underneath an LDO (thinking CDMA and GSM RF PA's here) - So in those cases, it is a different setup.

Jerry