OPA1S2385: Transimpedance circuit design questions

yunsik kim

Prodigy 130 points

Part Number: OPA1S2385
Other Parts Discussed in Thread: OPA350, REF2030, , REF2033, REF2025, TIDA-00316

Hello, TI experts.

I'm designing a transimpedance sensor using MEMS capacitor.

I configured a circuit using TI's OPA1S2385, REF2030, and OPA350.

The circuit consists of transimpedance amplifier, all-pass filter, SAC ADC, voltage reference and added buffer.

Here is my schematic.

Here are my questions:

1. Without OPA350 buffer, can REF2030 drive OPA1S2385 and ADC?

2. If I connect the OPA1S2385 supply voltage pin to 3.3 V from LDO instead of BUFF OUT, is there any degradation of performance?

Best regards,

over 3 years ago

0 kai klaas69 over 3 years ago

Guru 140200 points

Hi Yunsik,

oops, the OPA1S2385 is very fast. Do you need this speedy OPAmp? What is the bandwidth of your input signal?

Can you tell a bit more about your application? What is the input signal? What is the MEMS capacitance? What do you need the allpass filter for?

Kai

0 kim yun sik over 3 years ago in reply to kai klaas69

Prodigy 20 points

Hi Kai,

This circuit senses vibration on MEMS capacitor Cs.

When the capacitor vibrates, the capacitance of Cs changes slightly (about 10pF).

The first amplifier A1 converts the capacitance change to voltage change (ΔVo1).

Because the voltage change is very small and transient, the second amplifier A2 enlarges the change and delays it.

Then the ADC measures the delayed voltage change (ΔVo2).

The gain of the second amplifier is 30.

The input signal bandwidth is 4 kHz.

Yunsik

0 Marcoo Z over 3 years ago in reply to kim yun sik

TI__Mastermind 19370 points

HI Yunsik,

While you work with Kai for the OPA1S2385. I can answer question 1.

The REF20xx family has been successfully used without a buffer in our TIDAs. TIDA-00316 uses the REF2025/REF2033 without a buffer to level shift a 16Khz signal. This is higher than the 4Khz from your circuit, so this should work.

0 Michael Steffes over 3 years ago in reply to kim yun sik

Guru 47485 points

Well that first stage is not going to show anything but a different integrated noise as the source C changes. Are you sure you don't want some kind of test signal on the V+ input to have it change amplitude with C changes. 4kHz is pretty low F, you are going to need a feedback C for stability, that is described in this article showing a range of decomp op amps that will give you lower noise,

https://www.planetanalog.com/applying-high-speed-de-compensated-vfas-hitting-performance-targets-while-tuning-phase-margin-insight-10/

Amplifiers

Amplifiers forum

OPA1S2385: Transimpedance circuit design questions