20volt bias voltage for sensor with a 10v voltage reference and Op-amp, is this circuit loos

 

I
am designing a 20V voltage bias for a MEMS Sensor. Not sure about all the
resistor and component value i selected. Please can some one verify and
confirm? 

Kris.... On the face of it your circuit looks OK.  BUT (and it's a big but) the following will trip you up:
As far as I can see, the maximum supply voltage for the MAX876A is 18V, you are using 24V.  Maybe use a 15V zener with a 1K series resistor from the 24V to supply it.The circuit will act to supply 20V at the emitter of Q1.  That means the base of Q1 will need 20.6 volts and the collector should have even higher voltage.  Yet it only has 20V.  I'd supply the collector of Q1 from at least 24V.The same goes for your opamp U6A.  Give it a bit more headroom, it has to supply 20.6V to Q1 Base.  The LT6274 can operate with up to 32V supply so it will be fine with 24V as well.You have a 200 Ohm resistor from the emitter of Q1 (which is your output, though not labelled as such) to ground.  This will take 100 mA even before you take any current from the output.  I'd make this 1K or even higher so you have more output capability, though you don't say what you need.Finally your gain equation is missing something, it should beVout = Vin *( RF/RI  + 1)In this case Vin = 10V (from the MAX876A) , RF and RI are R15 and R14 respectively, both 10K.So Vout = Vin * 2 = 20V.   Your resistors were right, but the equation was missing Vin.Hope this helps... Best // David

KrisI have some questions as to what you are trying to do and what you are trying to drive. You seem to have selected some relatively pricey components and I am not sure if they are necessary.  I am also puzzled as to where you manage to get 21.75V and 20V? However David has given you a great generic answer plus some significant corrections to your design. However I have some major questions about the power supply to U6A. I tried to relate this to the data sheet of the LT6274. The nearest I can find is the S5 package, but pins 2 and 5 on your schematic appear to be inverted: On the data sheet V- is pin 2 and V+ is pin 5. In addition pin 5 in your schematic (if it is a supply pin) does not connect to a supply at all.  It looks to me that C48 was supposed to be the decoupling capacitor to U6A.Forgive me if I have misinterpreted your intentions.

Thanks David for your valuable suggestions, indeed  very helpful for making the design more stable or usable!  I am fairly new to this arena and need to do more research to get this right! Thanks again for your help!Best regards.

Hi Aubrey, I am trying to design a precision, near zero current 20v bias circuit for a accel board. It's for density study so trying to save space on the board hence the parts are pricey but small. Do you have any better suggestion? Yes LT6274 pin connection was wrong and so does the bypass cap! thanks for checking it! as I want a single supply the V-(pin2)  could be grounded, correct? Thanks again for your help!Best regards

KrisYes pin2 could be grounded.I would also like to query the input connections to the op-amp.  I believe you have pins 3 and 4 inverted. You said "near zero current". Why do you need Q1? U6 can source/sink 15mA.

Hi Kris-Instead of the MAX876, you might consider using a Maxim REF01CSA+.  It allows up to 33 volts on the input.-Rick