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.

Register Log in

[SOLVED] Floating DC voltage measurement fault

Status
Not open for further replies.
Mithun_K_Das

Mithun_K_Das

Advanced Member level 3
Advanced Member level 3
Joined
Apr 24, 2010
Messages
899
Helped
24
Reputation
48
Reaction score
26
Trophy points
1,318
Location
Dhaka, Bangladesh, Bangladesh
Visit site
Activity points
8,254
Capture.PNG

I was trying to measure the voltage across R1 which is floating. In simulation it works fine. But in practical, the output is always high. I tried reducing resistor values of R5 & R7. But there is always a offset voltage of 0.7V in the output of Op-Amp. Is this op-amp fault?

which type of op-amp I need to use in this case?
 

The picture shows that there's no 0.7V offset, the output is correct for 0.8 A output current. More likely you see a minimal output voltage (saturation), not an offset.

Two OP parameters must be considered in this case, common mode input range and output range respectively saturation voltage. LM358 common mode voltage is met with 20V supply, output saturation voltage is around 0.7 V for sink currents > 50 µA, that's what you see.

Best solution would be a dedicated high side sense amplifier (like TI INA180), next best solution a rail-to-rail OP.

If you stay with LM358, increasing the 10k/100k resistors by a factor of 10 would reduce the saturation voltage to several 10 mV.

Another option is to implement an intentional offset of e.g. 1V by connecting R5 to a positive reference voltage.
 

Hi,

In your circuit the voltages at resistor R1 are not floating. They are fixed at positive voltage supply of the OPAMP.
The common mode input voltage is about 18V..

* power supply voltage is OK
* input common mode voltage is OK

Either there is a wiring error, or
your resistor matching is too bad.

Klaus
 

FvM said:
The picture shows that there's no 0.7V offset, the output is correct for 0.8 A output current. More likely you see a minimal output voltage (saturation), not an offset.

Two OP parameters must be considered in this case, common mode input range and output range respectively saturation voltage. LM358 common mode voltage is met with 20V supply, output saturation voltage is around 0.7 V for sink currents > 50 µA, that's what you see.

Best solution would be a dedicated high side sense amplifier (like TI INA180), next best solution a rail-to-rail OP.

If you stay with LM358, increasing the 10k/100k resistors by a factor of 10 would reduce the saturation voltage to several 10 mV.

Another option is to implement an intentional offset of e.g. 1V by connecting R5 to a positive reference voltage.
Click to expand...

I double checked the hardware, and it was as like this diagram. Simulation result is totally ok. But in real hardware I'm always getting 4.1V at output. Is this due to OP-Amp? If I use uA741 which is available at local market will it solve the problem? Because Ti INA180 is not available at our local market.

- - - Updated - - -

KlausST said:
Hi,

In your circuit the voltages at resistor R1 are not floating. They are fixed at positive voltage supply of the OPAMP.
The common mode input voltage is about 18V..

* power supply voltage is OK
* input common mode voltage is OK

Either there is a wiring error, or
your resistor matching is too bad.

Klaus
Click to expand...

Is there any guideline to select the resistors for this type of circuits?
 

Hi,

I agree with FvM, the 0.7V you see is output saturation.

Is there any guideline to select the resistors for this type of circuits?
Click to expand...
Ohm's law is enough.

Let's assume R4 is 1% too low, all others are perfect.
Then the voltage at noninverting input becomes 18.3335V.
This means the output will become +1.668V instead of 0.0V.

This makes about 1.7V per 1% resistor deviation.

Klaus

Added.
Now you say 4.1V.
This measns the resitor deviation from one to the other is about 3%.
Or you have leakage currents, Opamp error (maybe caused by ESD).

Can you upload a photo of your circuit..where we can see details?

Klaus
 

When I cut the input lines, I mean disconnecting R4 & R6 from R1 then output voltage of Op-Amp is 0.7V. But when connected, no matter current is changes, op-amp output is 4.1V. I'll try after changing several Op-Amps and another design I'm doing with uA741.
 

LM or UA makes no difference, they are just different manufacturer prefixes. The '741' is an old (> 40 years) design with relatively poor performance, you need a more modern design and the ones suggested are specifically for use where the input pins are at elevated voltages or draw very small input current. Forget the 741, it should have been killed off decades ago, it's a mystery why it is still available.

Brian.
 

Mithun_K_Das said:
View attachment 141270


which type of op-amp I need to use in this case?
Click to expand...

You need an opamp whose input common mode range includes its own positive supply level or even higher. The 741, as Brian mentions, is an antique opamp does not meet that requirement by a wide margin.

However, if that is all what you can get, then increase the supply voltage to the opamp about 5 volts higher than the voltage where the sense resistor is.

Lastly, the output of a 741 will never go down to zero.
Zero meaning its negative supply rail.

Brian, I ask that same question myself many times. There must be an answer, and the answer is that someone, somewhere is still buying it!!
 

Brian, I ask that same question myself many times. There must be an answer, and the answer is that someone, somewhere is still buying it!!
Click to expand...
I just checked and I have 10 unused ones still in a pack here - but their date code is 1986 so it isn't me :lol:

Brian.
 

Capture1.PNG
Capture2.PNG

today I tried with this circuit. It works in simulation but in real hardware it works randomly. very very disappointing.

- - - Updated - - -

In real field:

input ------ output

22V ----> 1.445V
23V ----> 1.445V
24V ----> 1.623V
25V ----> 3.667V
26V ----> 3.550V
27V ----> 3.02V
29V ----> 0.23V
30V ----> 0.23V

I still don't understand what is the problem with it.

- - - Updated - - -

Capture.PNG

I'm using this floating voltage measurement circuit to measure the voltage described in this image. Is there any other efficient way to measure that voltage?

- - - Updated - - -

If I use 4N35 opto-coupler will be be solved? I tried with 4N35 and it is almost linear.
 

Hi,

The problem is, that you are measuring DC voltage at a switching node.
You never mentioned this before. That is why we always ask for complete informations.

You say "floating", but I call floating as "high impedance, not defined voltage offset"
But in your case it is defined somehow by the switching circuit.
You can exactely say / calculate / simulate the complete voltage waveforms.

Sadly you still don't give the the voltage ranges of both batteries.

In your case there is a relatively simple solution:
Use a low pass filter to suppress the switching AC and focus on the battery's DC.
Do this with both differential inputs and reference both signal to the OPAMP's GND.

Klaus
 

It's annoying that you come with a completely different circuit specification after one week.

The basic problem is still the same, exceeding the common mode range of the amplifier circuit. But now you are exceeding it in the negative direction.

Your simulation circuit is wrong because it misses a common mode voltage source.
 

Hi,

I still don't understand what is the problem with it.
Click to expand...
You have a simulation tool.
Create a circuit that equals your reality.
Use the scope to see the OPAMP input signal.
Use the OPAMP datasheets and read about their input specifications.

Klaus
 

FvM said:
It's annoying that you come with a completely different circuit specification after one week.

The basic problem is still the same, exceeding the common mode range of the amplifier circuit. But now you are exceeding it in the negative direction.

Your simulation circuit is wrong because it misses a common mode voltage source.
Click to expand...

I'm not clear with this type of circuits. So it is really making me swimming in the middle of a sea. Anyway can you guide me to get a complete understanding about this type of differential amplifier circuits? I'm tired trying one by one. That is why you are seeing different circuits.

- - - Updated - - -

I'm really not clear with common mode voltage, common mode ratio. Do you have any application note or pdf that I can read and get a clear idea? TIA.
 

Make the differential amplifier a little bit different, with built-in voltage divider

vsense.jpg
 

Attachments

  • diff_voltagesense.zip
    1.7 KB · Views: 123
Hi,

Anyway can you guide me to get a complete understanding about this type of differential amplifier circuits? I'm tired trying one by one. That is why you are seeing different circuits.
Click to expand...
It's not the problem of the amplifiercircuit you show....
The problem is that you dud not show your requirement:
* you want to measure DC batrery voltage
* where the battery is on a switched (with reference to the amplifier GND) node. Introducing AC with negative voltages.

So initially there is no amplifier circuit....there are batteries and a switching circuit....this is what we need to see in the first post.

Do you have any application note or pdf that I can read and get a clear idea?
Click to expand...
Any basic Opamp tutorial will do. Many thousands in the internet.
Or you do a search in this forum...I assume there is enough information.

Klaus
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top