Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Choosing common mode voltage for inamp (ad623)

Status
Not open for further replies.

Antor

Member level 2
Joined
Mar 1, 2019
Messages
51
Helped
0
Reputation
0
Reaction score
0
Trophy points
6
Activity points
501
Hello there,

I want to measure line-neutral voltage with single supply AD623.

Grid Line-Neutral voltage 220Vrms
Input signal for inamp swings between -1V and 1V (Zero crossing at 0V).
I want my output signal Vpp=2V, signal swings between 1.5V and 3.5V (zero crossing at 2.5V.)
The circuit on attachment works perfectly.

Now my first question is about choosing best Vcm.
Should I choose Vcm as 1V ? (to shift the negative alternance peak to 0)
or should I choose Vcm more like 2V ? (more than needed)

Second question
Which one is best to get references for Vcm and Vref ?
Voltage divider with resistors > opamp buffer > follower > to where needed
Using a voltage regulator (TLV431 like advised here before) > opamp buffer > follower > to where needed

Option 1 disadvantage = because I'm using 7805 as 5Vdc supply. And I really don't trust it.
Option 2 disadvantage = choosing and finding correct resistors in market are truely painfull. I can't always get exact reference as I wanted.

Which option should I go for ?


0.jpg
Rg = inf (gain=1)
Figure 1 Vcm=0 V (Connected to DC neutral)
Figure 2 Vcm=1.1 V
 

You want to center Vcm within the common mode range, about 1.65 V. 1 V will give reduced output swing with 5V single supply, review datasheet figure 22.

You should also consider that the input voltage isn't necessarily symmetrical to circuit ground. In this case, you won't achieve the intended voltage swing. Higher divider ratio and gain > 1 are necessary to increase the AC common mode range.

For best accuracy, Vref should be derived from the same source as the ADC reference.
 
  • Like
Reactions: Antor

    Antor

    Points: 2
    Helpful Answer Positive Rating
Thanks FvM.
My last problem still remains because my MCU has no ADC reference.

And I'm about to follow Klauss's advice on current measuring. Seems like LEM fluxgate transducers are a bit problematic.

Do you have any suggestion with 0-10A current measurements ? I have 15mR shunt resistors.
 

Wait I'm confused about what you're calling Vcm. I assumed it was the amplifier output ref pin but your diagram labels it as the the input network ref pin. Are you asking about one, the other or both?

The input would ideally be referenced to the input_cm_range/2. And I'd want the output referenced to adc_ref/2. Use a 50/50 divider to generate adc_ref/2, bypass it and buffer it. If adc_ref/2 is close to input_cm_range then use it for both (and it probably is).

Agree that you need to consider input cm range carefully. Assume your ground could be equal to either AC input or anywhere in between.

If you want good matched resistors search "ACAS" on digikey.

I'll add that I don't think this necessarily calls for an inst. amp. Inst amps are good if high input impedance or economical matched resistors are needed. But you benefit from neither in this case. See the below where that could be any decent dual opamp:
AC line Mon.PNG
 
  • Like
Reactions: Antor

    Antor

    Points: 2
    Helpful Answer Positive Rating
Vcm is for both inputs. It pushes the signal from both sides.

Speaking regardless of the input/output demans of picture I shared; Is this what I should understood ?

We have an input signal Vi(t)=2.cos(wt)
Vcm set to 2V and Vcc=5V, gain set to 1
Output is now Vo(t)=Vi(t)+2
Signal should swing between 0-4V but since Vcm is for both inputs, output positive alternance clips/saturated at 3V (Vcc-Vcm)

To be able to get rid of negative alternance clipping I need to increase Vcm but now positive alternance may get clipped.
The solution is increasing the voltage divider (decreasing input for inamp) so I will not be affected by positive alternance clipping. Because my peak will be way under it. Putting a gain solves the decreased input.

I'll add that I don't think this necessarily calls for an inst. amp. Inst amps are good if high input impedance or economical matched resistors are needed. But you benefit from neither in this case. See the below where that could be any decent dual opamp:

LM741 dual supplied +-20Vdc output can be +-16V at tops. Slew rate is 0.7V/uS
AD623 single supplied 5Vdc output can be 0.2 to (Vdc-0.15)V. Slew rate is 0.3V/uS

I've choosen AD623 because it is fast and (output swing/supply voltage) is good. I'm sincerely open to any opamp suggestions.

If you want good matched resistors search "ACAS" on digikey.
Because of the region I live buying from digikey costs me a month or more. Also limit for tax free shopping set to zero a few days ago and shipping cost is included.
Lets say I wanted to import 2$ product with 10$ shipping fee, total cost is 12$ and it's tax is 70% and it costs me 20$ (20% + 20% +20% 3 taxes added to each other).
I need to buy from local stores and they really don't have much options.

asdf44, FvM I really thank you both. And I'm in a need of shunt resistor current measurement methods.
 

OPA196 is my current favorite low cost 'all-purpose' opamp though it sounds like you'd have a hard time getting it. I have no issues with the inst amp I'm just pointing out that 'regular' opamps work just as well in this particular case. If you have limited local supply this might be particularly useful.

Did you have another thread on shunt sensing? There are a lot of ways to do it and the nuances depend on the exact requirements.

One initial suggestion would be to look for integrated shunt amplifiers. TI's INA series has many. If you can't readily get these then a standard differential amplifier topology works as well. The inst amp might be usable here - you don't benefit from the high impedance but the internally matched resistors help the common mode rejection at least (not gain since that depends on an external resistor which will have limited match to the internal ones).
 

my MCU has no ADC reference
Most MCUs are using AVdd as reference, in this case it's well accessible. If the ADC has an internal reference without external connection, you would use an additional precision reference for AD623 Vref pin.

Vcm level must not be particularly accurate, variations are suppressed by instrumentation amplifiers common mode rejection.
 

Hi,

If global VCC is the ADC VRef, then the overall performance is limited.
Thus the whole design may be relaxed regarding precision.

Klaus
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top