Need help from pros (emergency)(desperate)

[ricopt]

Any pro out there, pls pls pls help me with my design
I need to wrap up my design tomorrow.Need to settle with this problem
I am currently designing shunt feedback transimpedance with cascode gain stage and source follower.

I am able to get the gain at 525.6 for Rf=1k

but when I simulating Ac analysis with magnitude 1A , my Vout went to large negative value .and my biasing is correct.I can't find any solution to it.I also duno why it happen.Can someone guide me ?







These two are the info regarding the transistors



Transimpedance gain



Ac analysis circuit



bandwidth

 

[erikl]

Break your feedback loop!

If you want to run an AC analysis in order to measure gain & phase, you must break the feedback loop, which you didn't do in your simulation. Hence your (negative) transimpedance of -526Ω is superimposed by a positive one of +1kΩ (your feedback resistor), which probably causes this weird gain magnitude result.

You might want to study this thread, and especially my explanation attempt - which was wrong for that thread -, but might give you some enlightenment for your problem.

So break your feedback loop - either by the LC method shown, or - even much easier - by using the STB (stability) analysis, breaking the feedback loop by an iprobe.

 

[ricopt]

don't quite understand with your explanation....sorry with my noob skill
I really need more detail explanation....like how do i simulate bandwidth for shunt-feedback transimpedance amplifier ?

 

[sjamil02]

You should break the loop at 'in2' node. As what erikl suggested, using LC method, you can put a big inductance (1kH) and big capacitor (100uF) and injected your ac test signal.

 

[ricopt]

Sorry to trouble you, but can you do the LC method based on my circuit ?Like draw for me ?
I still dun get your meaning
how exactly break the in2 node ?

 

[ricopt]

Re: Break your feedback loop!

If you want to run an AC analysis in order to measure gain & phase, you must break the feedback loop, which you didn't do in your simulation. Hence your (negative) transimpedance of -526Ω is superimposed by a positive one of +1kΩ (your feedback resistor), which probably causes this weird gain magnitude result.

You might want to study this thread, and especially my explanation attempt - which was wrong for that thread -, but might give you some enlightenment for your problem.

So break your feedback loop - either by the LC method shown, or - even much easier - by using the STB (stability) analysis, breaking the feedback loop by an iprobe.

bro, i have seen your reply , but i dun understand...how i implement the LC method into my circuit ?
can draw for me ?or sketch for me ? pls ?

 

[erikl]

Re: Break your feedback loop!

i dun understand...how i implement the LC method into my circuit ?
can draw for me ?or sketch for me ? pls ?

I've already given you a very concrete hint in and use this LC connection (bottom left corner; the inductance L0 ≥ 1H in series with your 1kΩ feedback resistor, capacitor C2 ≥ 1F, instead of the voltage source V23 your unit (1A) current source.

HTH! erikl

 

[ricopt]

What current source should i replace with ?
I understand with the LC component, but what about the source ?
DC voltage source ?or AC voltage source ?Ac current source ?DC current source ?
How much value should i set for my source ?

 

[erikl]

Anyone who can read performs better

DC voltage source ?or AC voltage source ?Ac current source ?DC current source ?
How much value should i set for my source ?

So (re)read the last sentence of my former posting!! AC of course for an ac analysis; the DC bias will be provided by your feedback.

 

[ricopt]

instead of the voltage source V23 your unit (1A) current source ??????????

what you mean is it replace the V23 with my current 1A source ?

 

[erikl]

transimpedance circuit!

Hey! Is it a transimpedance circuit - or not?
A transimpedance amp converts input current into output voltage, doesn't it?

 

[ricopt]

oh ya, sorry
thx glad to have you all to help