MC33078: Request for schematic design review

Hi Jeffrrey, 

Jeffrey Chung1 said:

(The output frequency of MA300D1 is 300kHz.)

MC33078 has issued PCN recently and I was wondering if the issues are related to post-PCN op amps. Could you send an IC's image to confirm? The date code on the op amp will likely be manufactured after April, 2024. 

https://mm.digikey.com/Volume0/opasdata/d220001/medias/docus/6154/PCN20240411004.1.pdf

In addition, where MA300D1 are you referring to? Please remove C5 of 5.6pF that is paralleled on top top R5 or 91kohm. After it is removed, please let me know if the oscillation is disappeared. 

If you have other questions, please let me know. 

Best,

Raymond

Hi Raymond,

Let me check a top marking image with customer and send you it.

MA300D1 means RS1 in the schematic and it is proximity sensors. Let me check your recommendation and I will let you know if this issue is cleared.

Regards,

Hi Raymond,

Please check a top marking image(28K) below. I think that the issued devices are manufactured before April, 2024.

And the customer already replaced to new IC, so unfortunately can not be tested for removing C5.

Regards,

Hi Jeffrey, 

Jeffrey Chung1 said:

I think that the issued devices are manufactured before April, 2024.

Yes. 28K refers to manufacture date in Aug., 2022, so this is the pre-PCN op amp. 

Jeffrey Chung1 said:

There is the oscillation issue on the output of MC33078. The failure rate is about 7%.

My guess is that  pre-PCN MC33078 does not have oscillation issue, but post-PCN MC33078 does. Please confirm which FAB has oscillation issue. 

Jeffrey Chung1 said:

And the customer already replaced to new IC, so unfortunately can not be tested for removing C5.

Can you clarify the above statement? The output frequency of MA300D1 is 300kHz - what is actual input signal's frequency? 300kHz may refer to the op amp's cutoff frequency. Please let me know. 

Best,

Raymond

Hi Raymond,

As I told you above, there is the oscillation issue even it is pre-PCN MC33078.

The actual  input signal's frequency is 300kHz,

The customer requested to submit this issue on CPR.

Regards,

Hi Jeffrey, 

The op amp is unable to support 300kHz actual input frequency based on GBWP. You circuit gain is 1 + 91/1.2 = 76.83V/V and GBW = 16MHz, the cutoff frequency for the configuration is approx. 16MHz/76.83 = 208kHz, which is unable to support 300kHz input signal - That is the reason that I am asking. In other words, the output signal may be gained up, but also attenuated significantly at 300kHz. 

For actual 300kHz input signal, the op amp's cutoff frequency should be allocated near 3MHz, a decade higher. So the circuit is not going to work as designed, if you input signal frequency is 300kHz. What is the circuit application?

Jeffrey Chung1 said:

there is the oscillation issue even it is pre-PCN MC33078.

Could you tell me which output node is oscillating?

Best,

Raymond

Hi Raymond,

Thank you for your kind explanation. Let me discuss with customer about your comment. Could these circuit configurations cause oscillation?

And their application is the currency counter.

Regards,

Hi Jeffrey, 

Jeffrey Chung1 said:

 Could these circuit configurations cause oscillation?

Could you send me a screen capture of the oscillation? Your schematic marked two 4 spots, It is not clear which stage has the oscillation?

Best,

Raymond

Hi Raymond,

The black marked spot(Pin 1 of U2A) is the oscillation one. They said that the oscillation waveform is square waveform(no frequency information) swing 0V to 5V, but they have not a screen capture of the oscillation. And they said that the real cutoff frequency is more than 300kHz because of using Vcc=5V. Also, they use this OPAMP on all of models, and there is this issue on only 7 devices of 100 devices.

Regards,

Hi Raymond,

Let me check it with customer and update it.

Regards,

Hi Jeffrey, 

Ok. Please let me know the test conditions about the oscillation. 

Best,

Raymond

Hi Raymond,

Raymond Zhang1 said:

So the question is what is resistive value between  V0R1 and VR02 nodes under the observed oscillation.

They said that V0R1 and VR02 terminal connections are not open even though they couldn't check the exact resistive value between both because there is no board occurred the oscillation. As a result, they decided to change OPAMP from MC33078 to LMH6643.

Thank you for your aggressive support until now.

Regards,