ADL8100: About S-Parameter comparison result

The pink and blue traces are ADI supplied data. The pink trace (external bias tee) has lower insertion loss as expected. The red trace is customer generated deembedding data. You need to use Thru line for deembedding and thru lines are different for external bias-tee and onboard bias-tee boards. Let's double check you're using correct thru line and measurement data combination. 
Since the board with external bias tee has no extra bias tee components on the signal path, it's natively simpler and this is an advantage for better deembedding and showing inherentperformance of the chip. I suggest using the external bias tee board and thru for that purpose.

Hi Yus-san,

Thank you for your reply.

In fact, Red data was provided from Eamon-san, please see the below thread.

ADL8100: about how to calculate the S-Parameter of IC - Q&A - RF and Microwave - EngineerZone

If correct you said, should he compare the red data with the blue data because, the red data made from the board without bias-T?
However, I am thinking, in fact, the data of only IC must be better performance compare with the IC + board data in gain specificatoin.

Best Regards,

Tomohiro

Hi Tomohiro-San,

If correct you said, should he compare the red data with the blue data because, the red data made from the board without bias-T?
-Correct

the data of only IC must be better performance compare with the IC + board data in gain specificatoin.
-Correct


I repeated the deembedding again and see the plots below. Gain performance of ADL8100 IC shown with red trace is better than the board measurement including trace loss etc.

Hi Yus-san,

Thank you very much for your support.
I confirmed your data and I think it is correct data.
Could you provide me with S-Parameter data (S11, S12, S21 and S22) as IC only?

Best Regards,

Tomohiro

Adl8100_DeEmbedded_extBiasBoard.zip

I'm glad I can help. 

Best regards,

Hi Yus-san,

Thank you for your support.

Customer sent to me his comments and questions.

Could you give me your advice and answer for him.

Please see the below comparison data as S11, S12, S21 and S22.

Please check this data so it is in blue data in above graph.

His comments and questions are below.

He said he cannot judge data in red is correct or not if data in pink is base data.

Question:

1. Where the RL of the S55 is improved, the gain change of the S65 is slight, why whereas the two de-embedded data show that the change in RL is sensitively connected to the gain?

2. Why the changing value is so large when compare with S22, S44 and S66?

3. Is it correct there is little changing value at the each isolation (S12, S34 and S56) ?

Best Regards,

Tomohiro

I think the questions refer both old and new deembedded data, I couldn't be sure. Therefore, I don't think I understand the questions. Let's just only use external bias tee measurement and recent deembedding data I shared previously.

ADL8100ACPZN_Non-AFR_SParameter_25C.zip
8117.Adl8100_DeEmbedded_extBiasBoard.zip