OPA4354: When using OPA4354, chip temperature is high.

Hi Eggsy,

what I noticed immediately is that the 1.65V reference voltage is not low ohmic enough. A simple 10k voltage divider is unsuited to generate a pseudo ground. The consequence is that the OPAmps "see" each other by the voltage drops of their currents flowing into this node. This can cause unwanted feedback and result in instability.

Moreover, the non-vanishing source resistance of pseudo ground (5k !) causes a heavy imbalance of differential amplifiers which heavily degrades their common mode rejection.

Kai

Hi Kai,

How to fit it?

Thanks!

That is true the bias setup fed into the diff stage will be part of that stages gain, not sure that would cause overheating (normally, we think of excessive power as being associated with oscillations). 

The other thing I wonder about is that extremely large 47uF feedback C - the op amp Zol might not like that, perhaps, if you are oscillating, try removing those temporarily. A clearer schematic would help, ideally in TINA. Your 2nd stage MFB filter R values are relatively low given the very high noise coming out of the first stage R values, 

I put your MFB RC values (as best I could read them) into a tool that extracts what is being implemented, here that is - not sure why you want a 100kHz + filter following your input stage that is bandlimiting (with that large feedback C) to about 0.3Hz (1/(10k2pi47uF)

Hi Michael ,

Maybe you look the value of feedback C which is wrong. It is 47pf rather than 47uf in my pic.

Thanks!

yes, very hard to read, can you improve the image?

Hi Michael,

Sorry about that.

I have only this one.

well you could redo it in a TINA file and attach, unnecessary hurdle to resolve without clear values. 

Hello Eggsy,

Can you clarify on the conditions when the amplifier heats up?

Is this when only power is supplied or are you providing an input? Is heat the only issue, is the amplifier behaving normally?

Best,

Hasan Babiker