OPA567: Battery simulator high ripple output

Vo Thai Nguyen

Part Number: OPA567
Other Parts Discussed in Thread: OPA2335, OPA335, BQ25121A

I have made a battery simulator based on OPA567 (see picture). However, the output voltage contains the high ripple voltage (see picture) when no load. I have checked voltage at tespoints VDAC, 1.8VREF but there are not any ripples. Could you teach me how to filter out these ripples? Thank you.

over 3 years ago

0 Thomas Kuehl over 3 years ago

TI__Guru* 93105 points

Hello Vo Thai,

Is the ripple frequency at 443 kHz or 452.9 kHz as seen on the DSO displays? If so, the ripple might actually be oscillation associated with the OPA2335/OPA567 loop design. What kind of load gets attached to VOUT? Does the ripple disappear under any particular output conditions?

Regards, Thomas

Precision Amplifiers Applications Engineering

0 Raymond Zhang1 over 3 years ago

TI__Mastermind 48056 points

Hi Vo,

What you observed is not ripple, it is oscillation due to a "Q" created from two op amp feedback loops, which OPA335 rolls off at -20dB/decade, and OPA567 rolls off at -20dB/decade. When you feedback output of OPA567, it creates the "Q" circled in red. The BWs of two op amps are very close, and the Q is created at the unity gain BW of the feedback loop in OPA567. To get around the issues, I recommend to do the following. Otherwise, it is going to be difficult to compensate the Q with a zero in the feedback loop.

If you have questions, please let us know.

I would prefer the top choice, see the attached simulation.

/cfs-file/__key/communityserver-discussions-components-files/14/opa567-Ripple-08182020A.TSC

Best,

Raymond

0 Vo Thai Nguyen over 3 years ago in reply to Thomas Kuehl

Prodigy 90 points

Hi Thomas,

Thank you.

Yes, your thoughts are right. I have just tried some kind of loads as following:

-Resistive load: 10Kohm@3.8V->ripple same as no load. 100ohm@3.8V-> no ripple. 100ohm@3V-> ripple with lower amplitude.

-BAT pin of a PMID (BQ25121A) with 2.2uF cap in parallel (PMID adopted in a DUT): no ripple.

-VIN pin of a buck-boost IC with high capacitor >300uF: low amplitude ripple.

So the ripple level is depending on the load.

0 Vo Thai Nguyen over 3 years ago in reply to Raymond Zhang1

Prodigy 90 points

Hi Raymond,

Using the OPAMP OPA2335 in the loop to increase open loop gain is my intention. I want to minimize the voltage output from rail in order to reach as close as to 5V at the output with the 5V power supply (see picture). Maybe I am wrong.

Thank you for sharing the file, but I am new to the TINA.

0 Raymond Zhang1 over 3 years ago in reply to Vo Thai Nguyen

TI__Mastermind 48056 points

Hi Vo,

I am not convinced that your issues are resolved --> maybe you do not observe the "ripple" as much.

Please perform a small or large step function or pulse at input of OPA2335, and observe the output OPA567's response. If the output does not observe the oscillation behavior, then your problem is resolved. We trust our simulation model, and it is shown that you will have oscillatory behavior at the output.

Regarding to RL=1kOhm, Acl > 100dB (closed loop Gain), that it is referenced to OPA567's output swing behavior from the rail at the load. The overall circuit gain = Gain_OPA335 * Gain_OPA567, but OPA567 will swing close to rail, if the its load is light (e.g. R_load = 1kOhm, Acl > 100dB).

Best,

Raymond

0 Vo Thai Nguyen over 3 years ago in reply to Raymond Zhang1

Prodigy 90 points

Hi Raymond,

I post the waveform pictures of 2 cases: original version and modified version (x2 gain stage uses OPA335). I think the issues are resolved. I also do simulation on the file you provided. The output square-wave waveform is clean after modifying the circuit.

Thank you for your useful help.

Modified version: