Question about voltage follower

[EDA_hg81]

voltage follower

I have configured an op-amp as a voltage follower.

The problem is when the input is 150mV the output is 880mV.

Is this problem related with input offset voltage, input bias current or input offset current?

The voltage follower doesn’t have any gain? How this can happen?

How I can fix this problem.

Thanks.

 

[FvM]

voltage follower stability

Hard to decide without knowing supply voltages and OP type, but the OP may be operated outside it's designed input or output voltage range.

 

[EDA_hg81]

voltage follower configuration

It is a voltage follower with 5V single power supply.

The op-amp is LT1806/7 rail to rail op-amp.

 

[saro_k_82]

how to check stability of source follower

Hey this opamp specifies 5uA bias current (out of opamp input) at 150mV common mode voltage in typical case. So if you have a resistance of 150K driving the input, you gain 750mV. So you shouldnt be surprised if you see 900mV at the output. With this opamp, the source impedances(on both the inputs) need to be very low.

 

[EDA_hg81]

voltage follower bias opamp

What you said is right.

In voltage follower configuration.

The inverting input is connected to opamp output and noninverting input is directly connected to the output of source.

The output impendence of source should be very low. I don’t it is the issue with input bias current.

When input is 3.5V the op amp output is correct, it is 3.5V.

But when input is 150mv, the op amp output is 850mV.

Do you have any ideas how to cancel this error voltage?

Thanks.

 

[saro_k_82]

inverting voltage follower

With this opamp., you cant get away with this error. The offset current is of the order of bias current so even if you try to match the resistances on both arms., you'll still observe the error. You need to reduce the driving impedance or change to an opamp that has bias current in the pA range.

 

[Borber]

stability source follower

Use osciloscope to check stability even OPA is declared as stable at unity gain.

 

[FvM]

voltage follower oscillates input bias current

Yes, oscillations may be reason for the DC offset. Generally, this kind of high speed amplifiers can't be operated without effective bypassing directly at the part. Also, a capacitive load or longer traces shouldn't be connected without a series termination (e. g. 50 ohms) at the amplifier. Oscillations would usually show by increased current consumption, you may burn your fingers when checking the device temperature.

 

[LvW]

stability voltage follower op amp

................
The inverting input is connected to opamp output and noninverting input is directly connected to the output of source.
The output impendence of source should be very low. I don’t it is the issue with input bias current.
When input is 3.5V the op amp output is correct, it is 3.5V.
But when input is 150mv, the op amp output is 850mV.
Do you have any ideas how to cancel this error voltage?
Thanks.

Hi EDA_hg81,

excuse my (probably) silly question, but did you bias the noninverting terminal correctly ? (There is no mentioning about it). You have a single supply of 5 volts and you get 3.5 volts at the output ? What do you mean - peak value or rms or even dc ???

 

[EDA_hg81]

single supply op amp voltage-follower

Thank you all.

The input to this op amp is DC coupling.

The DC input range is from 3.5V to 150mV.

When input is 3.5V, the opamp output is 3.5V.

But when input is 150mV, the output is 880mV.

The noninverting input is directly connected to the output of sensor and also connected to the ground by a 47K resistor in parallel connection according to reference design.

But opamp LT1807 I used is different from reference design.

LvW
Thank you for your question.
bias the noninverting terminal correctly.
I really have no idea about how to bias the voltage follower, please let me know.

the input is from 150mv to 3.5V, Should we need to bias noninverting pin in a single power supply op amp?

The PCB trace is very short(10mm), and clock is running at 250KHZ.

 

[FvM]

voltage follower voltage input offset op amp

Without additional attributes, DC measurement must be assumed, as all contributors did.

 

[Borber]

how do you cancel a follower

First check output voltage when input is 2V. If output is not 2V then you introduced some offset and attenuation to the "follower". When follower is working correctly up to certain input level and then goes wrong then check stability at this level.

 

[EDA_hg81]

op-amp voltage follower configuration

I think "saro_k_82" is right.

I will try another kind of op amp with much lower input bias current.

I also will check the stability of this circuit.

Thank you all so much.

 

[EDA_hg81]

voltage follower op amp single source

The figure shows the reference design I have referred to.

The op amp they are using almost has the same input bias current as LT1806/7.

What is the purpose of C32 and R15?

Do they want to use R15 and C32 to lower the internal impedance?

Thanks.