The problem about the 2nd harmonics in RFMD RF3267

[tangjinchuan]

Dear all,

I am writing a report to explain every parameters of the specifications of a WCDMA chip RF3267.

But i have a problem about the Harmonic Output (2fo) in its data sheet. The value is -15dBm under Pout +28dBm which converted to dBc is -43dBc. This is the spec:https://www.rfmd.com/CS/Documents/3267DS.pdf

However in wcdma standard spec, 5.11 Spurious Emissions. It has a -30dBm in the 1MHz bandwidth. And it's power is class 3 with a test power value 24dBm(tolerance +1/-3), which means the minimun requirement is -30dBm-24dBm = -54 dBc.

Can i say that this chip does not meet the standard of WCDMA.

Or i may do something wrong in the calculation or consideration.

Please help me!!! Thanks!

 

[BigBoss]

You interpret wrong the given data/spec..
Spurois Emission is different than harmonic level.Harmonic frequency ( 2nd Fo) does not fall down spurious mask of any digital scheme therefore its effect is generally not disturbing effect.
The essential is around main carrier/mask spurious components which may come from "spectrum growth".
Second harmonic is very far from this emission mask/spec.In order to check, you should apply a WCDMA signal at where PA has nominal output power and observe the spectrum in according with WCDMA mask specifications.

 

[tangjinchuan]

Thanks dear BigBoss,

'Spurious Emission is different than harmonic level.Harmonic frequency ( 2nd Fo) does not fall down spurious mask of any digital scheme therefore its effect is generally not disturbing effect.'
1.Can i interpret your words as: if a this scheme works in 2GHz operating frequency then 2fo is 4GHz which is far out side the bandwidth. So there is no need to consider the interference from this. And even the specification in WCDMA standard do not consider this because it's outside. But I'm still wondering for this chip why it still gave this parameter?

'The essential is around main carrier/mask spurious components which may come from "spectrum growth".'
2.Like ACPR right? or More than ACPR?

'Second harmonic is very far from this emission mask/spec.In order to check, you should apply a WCDMA signal at where PA has nominal output power and observe the spectrum in according with WCDMA mask specifications.'
3.Like question 1, so i what's the purpose to check this one under such a high frequency? Check this because it may cause interference for other scheme under 4GHz?

Thank you again!

 

[BigBoss]

Thanks dear BigBoss,

'Spurious Emission is different than harmonic level.Harmonic frequency ( 2nd Fo) does not fall down spurious mask of any digital scheme therefore its effect is generally not disturbing effect.'
1.Can i interpret your words as: if a this scheme works in 2GHz operating frequency then 2fo is 4GHz which is far out side the bandwidth. So there is no need to consider the interference from this. And even the specification in WCDMA standard do not consider this because it's outside. But I'm still wondering for this chip why it still gave this parameter?

'The essential is around main carrier/mask spurious components which may come from "spectrum growth".'
2.Like ACPR right? or More than ACPR?

'Second harmonic is very far from this emission mask/spec.In order to check, you should apply a WCDMA signal at where PA has nominal output power and observe the spectrum in according with WCDMA mask specifications.'
3.Like question 1, so i what's the purpose to check this one under such a high frequency? Check this because it may cause interference for other scheme under 4GHz?

Thank you again!

1-Yes, you don't need to consider this 2nd harmonic because it's far outside.The manufacturer gave this specification because it's a PA and ti may be used for different application out of WCDMA.It's a standard proceure..
2-Yes and in additional to this spurious emissions should have a certain limit. ACPR is a kind of figure of merit of nonlinearity but spurious may have different causes but both should be used to check it out.
3-No, you have to spurious emissions around WCDMA carriers ( suppose 2GHz ) , measuring around 4GHz doesn't make sense..

 

[rf1008]

for a stand-alone PA,the output power of harmonics will not below -30dBm for gereral requirements ,so the additional filter is needed.

 

[tangjinchuan]

Thank you very much!

I have another question about Noise power in the PA:https://www.rfmd.com/CS/Documents/3267DS.pdf

A PA usually comes with a parameter called Noise Figure, but this document has only a term called Noise Power.

The Noise power has a value -95dBm/30KHz for WCDMA in the condition:

-50<POUT<POUT,MAX,
RX=2110MHz to 2170MHz,
TX/RX Offset=130MHz

where the frequency range is not the PA's operating range 1920 -1980MHz.

Why it's over 30KHz?How can i understand this term and such test conditions.

How can i convert this into Noise Figure?

Thanks!

 

[vfone]

Your initial statement was correct. WCDMA ETSI spec (as well GSM and LTE all are the same) have requirements for spurious emissions outside of the band (including harmonics) as in the table below.

These are system requirements for out-of-band spurious emissions, so if a PA fail this spec, doesn't mean cannot be used. WCDMA systems (full-duplex system) always use a duplexer at the output, which will reject a lot of harmonics. You have to do an RF system analysis.

 

[tangjinchuan]

Your initial statement was correct. WCDMA ETSI spec (as well GSM and LTE all are the same) have requirements for spurious emissions outside of the band (including harmonics) as in the table below.

These are system requirements for out-of-band spurious emissions, so if a PA fail this spec, doesn't mean cannot be used. WCDMA systems (full-duplex system) always use a duplexer at the output, which will reject a lot of harmonics. You have to do an RF system analysis.

Thanks, things are much more clear now.