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.

how to match calculated and measured current of a brush DC motor ?

Status
Not open for further replies.

anotherbrick

Full Member level 4
Joined
Jan 10, 2009
Messages
217
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,298
Location
Istanbul , Turkey
Activity points
3,142
hello dear forum,

I have a full bridge driving a 24 V DC brush motor

I have measured the max current of H bridge with ACS sensor with STM32 uC
and wrote down for four different duty of PWM
I also measured the Vsource

I measured the L and R of the motor with LCR meter

I locked the rotor of the motor

however the calculated I max and measured I max are very different from each other

I have attached an excel file for calculating the I max of RL circuit from power electronics book of Muhammad Rashid

I tried with diferent R and L however still not a match

what am I doing wrong ?

please advise

thank you
 

Attachments

  • RL circuit current meas 2.rar
    3 KB · Views: 61

Hi,

I assume you just don´t take waveform into account.

Don´t use "RMS" measurement here, neither for voltage nor for current
Don´t use "max" measurement method.
--> use "average" measurement method.


Klaus
 
hello ,

Where can I find the formulas for calculating the average current ?

my final goal is to find the speed of the motor from current measurement

thank you
 

Difference between peak and average current isn't very large in this circuit, the observed differences between calculation and measured currents can't be explained this way.

There are however unclear points:

- what's the H-bridge switching pattern?
- is it correct to neglect switch voltage drops?
- does the H-bridge driver involve dead time respectively asymmetrical delay?

Without an actual motor voltage waveform, it's mostly guessing.

my final goal is to find the speed of the motor from current measurement
Speed is primarily set by applied voltage. Current measurement may be used to correct for the torque related speed reduction.
 
hello ,

gate signals of one leg ( the other leg Low MOS always ON )

gate1.JPG

IR2183 driver has 500 ns deadtime

motor voltage and current waveforms (ACS sensor output)

motor.JPG

MOSFET data
STP100NF04
N-channel 40V - 0.0043Ω - 120A - TO-220 - D2PAK
STripFET™ II Power MOSFET

how to calculate the current and speed of motor correctly ?
which formulas are valid ?

20170613_175030.jpg
20170613_175115.jpg

thank you for your help

- - - Updated - - -

this is the current signal when motor locked - other waveforms same
current.JPG
 

You are apparently showing the D=0.35 case, calculated and measured peak current is only 5% apart here. If you are asking why the difference is so big for D=0.14, you should show the respective measurements.

As for your general questions, it's useful to determine the average current over a PWM period, average current is proportional to motor torque. Speed is in a first order proportional to motor voltage, or more exactly motor emf, terminal voltage - voltage drop at winding resistance.
 
I locked the rotor and give different duty and measured average , max and min motor current
I have 750 W transformer 220 / 24 V which I rectify and apply to circuit

I took 30 measurements in 100 uSec ( PWM frequency is 10 Khz )
I quicksort algorithm the measured data
I took average of 30 measuremets for average
took average of 3 most high measurements for I max
took average of 3 most low measurements for I min

as mentioned above I calculated R of motor from average measurements
the R is not constant ( I need constant R value for calculating speed when motor is running - isnt?)

Resistor.JPG

I dont understand - I am stuck
neither average calculated R is constant
nor I max calculation from formulas of Muhammad Rashid power electronics book
gives same values as measured data
 

0.141 duty

gates
gate 014 duty.JPG

motor Voltage and current
current 014 duty.JPG

I have one question ; should not the current waveform be like this ?

freewheel diode.JPG

thank you
 

Hi,

are you sure you gave the system enough time - before each measurement - for the values to stabilize.

Mind:
the first PWM period´s average current will be low.. then it will increase...after a certain time it will not increase (noticably) anymore. Then it´s time to start the measurement.


Mind:
that the coil (copper) will become warm... the warmer the higher the copper resistance.

Klaus

- - - Updated - - -

Hi,

I have one question ; should not the current waveform be like this ?

coil waveform is different form driver waveform
* coil: triangle shape
* driver: like your hand drawn picture.

Klaus
 
should not the current waveform be like this ?
No. The motor current should be triangular (integral of the rectangular motor voltage). The current through an inductor can only change continuously. The shown waveform clarifies that the motor doesn't behave purely inductive.

My first question is if there are additional components between H bridge and motor, e.g. filters, clamping diodes, saturated inductors. What's the exact motor type, does it involve eddy currents?
 
@Klaus: I wait 3 seconds until circuit stabilize
the ACS circuit is at the input to the H bridge

@FvM:
there is an electrolytic capacitor parallel to H bridge C18 1000uF
U10 is ACS current sensor
D37 D38 are not present
component values are not real
other than that it is a standart H bridge

sema.JPG

the motor label
the motor is standart DC brush motor with 24 V brake
Its current can be 50 Amperes when the wheelchair is against an obstacle

etiket.JPG

thank you
 

O.K. that's unexpected. You are measuring H-bridge DC link current rather than motor current. The deviations from expected waveform in post #9 can be mainly explained by capacitor C18.

If you are interested in motor current, I would measure it directly. Peak currents will be always different, motor average current can be probably calculated by putting in bridge duty cycle.

Imot,avg = IDC,avg/D
 
@FvM
thank you for your answer

//// Imot,avg = IDC,avg/D \\\\\

is there a webpage or book, where I can find this equation and theoretical explanation of this ?
 

Hi,

a paper, a pencil, drawing both current waveforms, estimating the average of both currents, estimating the relationship of both average currents....
--> maybe you can safe the cost for the book.

Klaus
 
@Klaus @FvM

hello thank you for your advise:smile:
however I didnot take pencil and paper but I opened MS Excel :idea:
and calculated R_motor with FvM's formula
and it worked really

Resistor 2.JPG

now I have another question
The IR2183 gate driver has 500 ns dead time
the motor current can reach 50 Amperes max at 24 V

my question ; is this deadtime enough ?
or what value does it have to have ?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top