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.

The Ground termination between comparator and the load

Status
Not open for further replies.

SimonPoon

Newbie level 6
Joined
Mar 4, 2016
Messages
11
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
110
Hi all, I have face a problem for measure the current load by comparator, below is the block diagram for our product, it's a battery product.

We have a Q1 to control the battery on off by MCU, when MCU is cutoff the Q1, the Load is not cut because the path is change, current flow through the op amp TLV8802 input to ground, when I cut the supply of TLV8802, the voltage flow to regulator from the battery through the load and TLV8802 input. Since we are operate very high current motor, change to P-FET to cut the power is not available (higher Rds)

Is there any idea to fulfill this?
p.s. Cut the power and current detection when motor stop operate or the user switch off the battery(hence no need to detect when cut the Q1 cause the Battery is slept)

Thanks for the help!

blockdiagram.jpg
 
Last edited:

The above circuit seems to always saturate from the positive input.
The open loop gain of the comparator is very high.
 

The above circuit seems to always saturate from the positive input.
The open loop gain of the comparator is very high.

When Q1 is turn on (common to ground) and the motor is start to load, there have a sense signal come from the shunt, and the op amp get the signal to trigger the MCU. Those operation is work normally.
However, when the Q1 is off, what I expect is the ground cutted, the load cannot operated, but the load is pass the op amp internally and short to ground, like the below picture.

blockdiagram-2.jpg
 

As a general rule, the shunt resistor is placed just above the ground, in order to avoid changing the negative reference, but I'm not sure if could be used an open loop topology as you did.
 

Review TLV8802 data sheet, absolute input voltage ratings to understand why your circuit can't work.

Another design flaw is the lack of non-zero comparator input threshold to detect a load current.

You either need a comparator with 30V common mode range, or should use current sense between MOSFET source and ground.
 

Thanks andre and FvM's advise, does it mean I can't use this op amp to achieve my design? or I need to use pure comparator instead of op amp better?
Beside, I had a brainstorm, how is the position is exchange between MOSFET and shunt? does it reasonable?
 

Hi,

A comparator output usually is either HIGH or LOW.
An OPAMP output usually is linear somewhere between both supply rails. Non saturating.
--> Can you tell us your idea how you can "measure" current with the use of a comparator?

There are many thousands of example circuits around to "measure" current. What is wrong with them?

Klaus
 

Hi,

A comparator output usually is either HIGH or LOW.
An OPAMP output usually is linear somewhere between both supply rails. Non saturating.
--> Can you tell us your idea how you can "measure" current with the use of a comparator?

There are many thousands of example circuits around to "measure" current. What is wrong with them?

Klaus

Thanks for your reply, maybe my explanation no good. Actually what my need of my product is: when the load(motor) start operate, the MCU wake up, that's it, no need to measure the current value, because save the battery, MCU is always sleep until the load is start operate, then show the led, voltage measure...etc. I have try to use transistor try to achieve it, but the result is no good.
 

Hi,

OK the function is much clearer now.
Motor ON --> current flow --> detect current --> wake up MCU (If I understood right)

But on thing confuses me:
The motor is switched OFF/ON with the MOSFET. And the MOSFET is controlled by the MCU. So doesn´t the MCU already know when the motor is ON?

***

* What is your desired threshold current. = At which current you want to detect that the motor operates? (minimum)
* What is the (expectable) peak current?

Klaus
 

Hi,

OK the function is much clearer now.
Motor ON --> current flow --> detect current --> wake up MCU (If I understood right)

But on thing confuses me:
The motor is switched OFF/ON with the MOSFET. And the MOSFET is controlled by the MCU. So doesn´t the MCU already know when the motor is ON?

***

* What is your desired threshold current. = At which current you want to detect that the motor operates? (minimum)
* What is the (expectable) peak current?

Klaus

Hi Klaus, correct, the FET on/off is used to let user control the battery on and off or any fault condition to cut the power, the battery will sleep to save battery when idle a few minutes, but this time, the battery need keep stand by, thus, the FET should keep ON, so MCU didn't know the situation when sleep.

The range of motor should be 100mA to 7A because the customer can use different type of motor

Thanks
 

Hi,

If the 7A really is the expectable peak current...then you need a low ohmic resistor to keep voltage drop and power dissipation low.
Let's say power dissipation shold be be about 1W at 7A:
P = I^2 × R --> R = P / I^2 = 1W / (7A× 7A) = about 20mOhms.
The voltage drop at 7A is I × R = about 140mVolts.

But to detect a low current motor, then you need to detect currents below 100mA.
To get some margin ... let's calculate with 50mA.
50mA × 20mOhms = 1mV.

This is too low to drive a comparator directely, therefore you need to amplify it.
For standard comparators a threshold voltage of 50mV should be sufficient.

To get 50mV output from 1mV input you need a gain of 50.
It seems you have single supply only (no negative supply)
Look for an Opamp:
* rail to rail output
* input common mode voltage should go down to GND
* input offset voltage < 100uV

Connect the both opamp circuit inputs (signal as well as GND reference) directely at the 20mOhms shunt with extra traces.

Klaus
 

Hi,

If the 7A really is the expectable peak current...then you need a low ohmic resistor to keep voltage drop and power dissipation low.
Let's say power dissipation shold be be about 1W at 7A:
P = I^2 × R --> R = P / I^2 = 1W / (7A× 7A) = about 20mOhms.
The voltage drop at 7A is I × R = about 140mVolts.

But to detect a low current motor, then you need to detect currents below 100mA.
To get some margin ... let's calculate with 50mA.
50mA × 20mOhms = 1mV.

This is too low to drive a comparator directely, therefore you need to amplify it.
For standard comparators a threshold voltage of 50mV should be sufficient.

To get 50mV output from 1mV input you need a gain of 50.
It seems you have single supply only (no negative supply)
Look for an Opamp:
* rail to rail output
* input common mode voltage should go down to GND
* input offset voltage < 100uV

Connect the both opamp circuit inputs (signal as well as GND reference) directely at the 20mOhms shunt with extra traces.

Klaus

Really much thanks for your help, I had learn so much for your reply, I will try to find a match op amp based on your advise to achieve the design. Thank you!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top