OPA2182: EMIRR RF signal injection condition

Hello Thomas-san,

Thank you for your reply.
I have following two questions for your answer.

- 1 -
>We standardized our Precision Amplifier Op amp EMIRR testing to use 100 mVpk (-10 dBm) input condition. 

At first, I could not understand why 100mVpk = -10dBm.
If my following idea is correct, I can understand.

---
* dBW = dBm - 30

In above case,

dBW = -10dBm - 30
dBW = -40 

* dBW = 10log(W)

So,

-40 = 10log(W)
W = 1 x 10^-4

According to attached doc, impedance is 50 ohm.
So Voltage can be calculated following formula.

* W = V^2 / R

1 x 10^-4 = V^2 / 50
V = 5*SQRT(2) * 10^-2 (Approx 70.7mV)

Previously, I believe that above is Vpeak, but I noticed this is efficitve value. I need to multiply above value by SQRT(2).

* Vpeak = Vrms * SQRT(2)
Vpeak = 5*SQRT(2) * 10^-2 * SQRT(2) = 0.1V = 100mV
---

Q1 Is my above understanding correct ?

- 2 -
If above my understanding is correct, I would like you to confirm about below.

* My customer observed EMI influence of both OPA2189 and ADA4522-2 (Peak spectrum is arround 50MHz).
According to datasheet, EMIRR spec arround 50MHz is similar, but result of output Vos value is  several times different.(OPA2189 was worse than ADA4522)

At first, as I said above, input condition of OPA2189 is 70.7mV. So, input condition is different from ADI condition.
I assumed this difference can cause EMI influence difference b/w OPA2189 and ADA4522-2.
 
However, I did not know anymore why this difference was observed from your answer.

Q2 Can you explain why this difference was observed ?
(Since customer tried evaluation on their custom board, so Vcc condition, temperature condition is same)

If input conditon of OPA2182 is same as OPA2189, I think same issue will be happened by using OPA2182 as well.

Thanks in advance,

Nachida-san,

Here is the dBm to volts calculation for a -10 dBm power level using an online calculator. The voltage is 100 mVpk, or 70.7 mVRMS for a sine input which is what is used for the EMIRR testing so your results match:

"Worse" is a vague term and doesn't really tell us how much difference was observed between the OPA2189 and the competitor's EMIRR performance.

The OPA2189 EMIRR is presented in the datasheet "Typical Characteristics" curves, therefore there is no assurance that every device will have the same EMIRR as those used for the curve generation. There will be variance between devices. Additionally, we have a standardized fixture and procedure we use to characterize EMIRR as outlined in the information I provided the link to. I do not know what the competition uses for their EMIRR measurements so I cannot speculate for the variance between the OPA2189 and their product. 

It is best to test the OPA2182 in the customer application and see how it performs before making any judgement about its EMIRR performance in that application. Every application circuit is different and their RF environment, PCB layout, supply bypassing, etc. is different than the original EMIRR test platform so different results may be obtained in the application. 

Regards, Thomas

Precision Amplifiers Applications Engineering

Hello Thomas-san,

At least, I understand you can say as shown below.

* EMIRR measurement condition b/w OPA2189 and OPA2182 is same.

So, EMIRR spec b/w OPA2189 and OPA2182 can be simply compared.
Is my understanding correct ?

BR,

Hi Ryuuichi,

me thinks this on-chip EMI filtering should not be overestimated. As the on-chip EMI filtering only works above 10MHz anyway, much more sense makes the good old grounding, shielding and PCB filtering measures in combination with proper supply voltage decoupling and the use of a solid ground plane.

Kai