4
\$\begingroup\$

I am using DAC to produce voltage from 0-4.096V which I feed into the op-amp. I am using ADA4522 which can be supplied with 55V at positive rail and gnd at negative rail. Using the non-inverting op-amp configuration, R1=12.43k & R2=1k, I do get extended range of 0 to 54-point-something volts at op-amp output. Happy with that! Picture is below. btw. are resistor values okay?

enter image description here

Now to the other one. I want to amplify the voltage from the DAC(0-4.096V again) to the 0 to -55V. I have planned to connect ground to the positive rail and -55V to the negative rail. I guessed I should use inverting op-amp topology. But that seems not to work. I must have some stupid, obvious mistake I can not see.

enter image description here

Available power supply on this actual project is +55V & -55V with common ground and 5V with common ground. Every result I have is from spice simulation, ada4522 spice model is downloaded from Analog website.

EDIT: Sorry for bad-drawn schematic. Here is better one, I hope. Now it is clean that I have available dual-supply of +-55V and there is 5V supply, too. Thanks to your comments, I have changed R1 on op-amp for positive output. That is OK now. But for negative output, still struggling. If I connect positive rail of that op-amp to the 5V, I am out of op-amp specs. Struggling how to fix that.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
6
  • \$\begingroup\$ Welcome to EE.SE. Please use a schematic that uses the op-amp symbol, not a generic 5-terminal yellow box. There is a schematic editor built in here on EE.SE. Also showing what's a power supply and what's the DAC would help a lot. Check over your work and consider also adding some of your simulation results. \$\endgroup\$ Commented Sep 7, 2016 at 22:20
  • 1
    \$\begingroup\$ Also, having some knowledge of at least what characteristic of your circuit is not working will help produce a useful answer. It's unclear what a reasonable solution is if you haven't started figuring what may be wrong with your circuit design. \$\endgroup\$ Commented Sep 7, 2016 at 22:22
  • \$\begingroup\$ This question seems to be "I'm lost with this design problem, please help" which usually isn't a good format for questions here. You might want to break this down into smaller more directed questions. \$\endgroup\$
    – Daniel
    Commented Sep 7, 2016 at 22:24
  • 1
    \$\begingroup\$ "seems not to work" is not a very precise statement of the problem. \$\endgroup\$
    – user16324
    Commented Sep 7, 2016 at 22:27
  • 1
    \$\begingroup\$ The question is PERFECTLY clear - what is problematic is the circuit diagram symbols which may be confusing a few brains. If the circuit was redrawn with standard symbols more people may be happier. Or not. \$\endgroup\$
    – Russell McMahon
    Commented Sep 8, 2016 at 1:04

2 Answers 2

3
\$\begingroup\$

If you look at the data sheet, on page 6 you'll find characterization at single supply, 55 volts. Input voltage is 0 to 53.5 volts. In other words, you can't have an input closer than 1.5 volts to the positive supply. Your inverter attempts to drive the inputs right at the V+, and this won't work.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ Look at where gnd is connected in 2nd diagram. His IC "hangs below ground" - which is OK EXCEPT that Vin is +ve and ground referenced so is above his Vdd. He needs a split supply of some sort AND to meet common mode spec as you say. \$\endgroup\$
    – Russell McMahon
    Commented Sep 8, 2016 at 1:02
  • \$\begingroup\$ Noted. Changed the R1 resistor value to the 11.3K, now max. output is about 50V. That's fine. @RussellMcMahon - when I change positive rail of op-amp to the 5V, while leaving the negative one on -55V as before, simulation looks okay. But now I am over specified max. voltage, right? \$\endgroup\$
    – tiho4597
    Commented Sep 8, 2016 at 11:47
  • \$\begingroup\$ @SlouchingtowardBethlehem - see my refs to your answer in my answer and associated comments. \$\endgroup\$
    – Russell McMahon
    Commented Sep 8, 2016 at 14:36
1
\$\begingroup\$

(1) You are operating the ADA4522 on a single +ve supply. You have grounded the +ve rail so the IC is "below" the rail. BUT V1 is "ground" = V+ referenced so is above ground - ie ABOVE Vhigh_max.

If you wish to get a negative output from +ve input relative to some reference point you must use a dual supply. These do not need to be balanced (eg you may use ed +10 / -45V if useful BUT the absolute max is 55 Volt total.

(2) An approximation to split supply can be achieved by returning U1_pin1 and V1- to a voltage divider a few volts below V+ with suitable decoupling capacitor. .


Your 1st diagram appears to be running the IC at >> max allowed voltage. The simulator may not mind but the real IC will.

Running the device even at full maximum voltage rating is liable to be "unwise" and should be done only if absolutely essential and if you are aware of the risks.

Here is a super-abominable markup of your circuit diagram.
Despite this it makes it MUCH clearer what is wrong.
The opamp "hangs below" ground but Vin drives above ground and there is no supply above ground level. U1 pin 1 is above max allowed Vcm.

Lowering U1 pin1 to about 2V below ground will allow input common mode requirement to (just) be met. V1_-ve also must be connected to this reference point. V1+ = Vin+ is allowed to be above the maximum common mode voltage as long as the circuit is arranged to drive U1-pin2 to below the max Vcm (as is the case hereas the circuit acts to drive U!-2 to the same voltage as U1-1. ).

enter image description here

\$\endgroup\$
2
  • \$\begingroup\$ Thank you for your answer. I am struggling to understand the point (1) where you say that I should use some reference point and dual power supply. Where it should be referenced? Could you check the edited first post for schematic? I understand that I can not put input higher than the positive voltage rail now. \$\endgroup\$
    – tiho4597
    Commented Sep 8, 2016 at 13:10
  • \$\begingroup\$ @tiho4597 My point was made more simply (and before I wrote it) by WhatRoughBeast - on page 6 of the datasheet (see his answer) it specifies a maximum "common mode voltage" which neither input can exceed during normal operation. At V+ = 55V. this is 1.5v below V+ = 53.5V. If either input is taken above 53.5V wrt groi\und / V- at V+ = 55V then the IC will notoperate correctly. You can either use a 1.5V positive supply (or ideally slightly more) above the "ground" at V1- and U1 oin 1 in 2nd figure, and lower V+ so total supply is still <= 55V, OR lower U1-1 and V1- to say -2V. Same result. \$\endgroup\$
    – Russell McMahon
    Commented Sep 8, 2016 at 14:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.