Electrical Engineering

Electrical Engineering

  • George Hedley Rokos added an answer:
    How do I drive parallel connected mosfets without using each driver for each mosfet?

    I will have to drive 25 Mosfets in parallel. Using each driver for each device is one way. Other than that, is there any other method to drive all parallel connected MOSFETS without having any time delay between them so that all devices turn on and off at the same time?

    George Hedley Rokos

    As multiply stated above, there will always be different delays to the different MOSFETs.  The use of separate drivers local to the MOSFETs will reduce the capacitive load on drive line, and this in turn will reduce the delays.  However, it is also possible to control the routing to the MOSFET gates so that the series inductance of each path is very similar (wide tracks are better here, because they reduce the individual inductance).  The delay difference will then be dominated by the mismatch between the input capacitance of the MOSFETs, but this mismatch can be quite small if the MOSFETs come from the same batch.  Note that the series inductance on the outputs can in some circumstances be equally important - and separate drivers do not help here.
    If you don't need fast edges on the switching you can also use a resistor at the driver to slow down the transition time beyond the delay mismatch.  (You may in any case wish to control the transition time to avoid generating high-frequency electrical interference).  However, the transition current will still not match between the different devices due to threshold mismatch; the mainproblem here would be hot carrier injection in the first device to switch on, though this too is less of an issue of the MOSFETs come from the same batch.  The final issue to consider when lengthening the transition time is the effect of dissipation during the transient.
    Regardless of whether you use separate drivers, simulations that include board parasitics are really helpful here, and (naturally) the MOSFET manufacturer's data sheets should be consulted for illumination on all these matters.

  • Stephen N Smith added an answer:
    Can we make a Linear Polarization Measurement device experimentally?

    I am on a research which deals with service life prediction due to chloride laden corrosion. Within my lab experiments, I need to measure the corrosion current density. For that I intend to use the Linear Polarization Resistance method. 

    But, my lab facility lacks linear polarization device, nor a potentiostat equipment. In that case, would I be able to make one to measure the polarization resistance of the corroding rebar? 

    I want to measure the current for different polarization values to obtain the Polarization resistance. Please follow the attached plot. 

    Stephen N Smith

    Yes, but it depends upon the selection of the op amp IC.  If you get the right chip and modify the circuit just a bit, it is possible to design one that can measure pA.  But to do that, you probably need to consult with an electronics engineer or tech,

  • K. Sreenivas added an answer:
    What is the best sensor to measure range in small precision?

    I want to find the distance of an object ahead of my device. That thing will approach my device and stopped. When that object is stopped, I want to know how far that object will be but it just about just millimeters or smaller away. What is the best sensor to measure that distance? It will be fine if the sensor is contact less or not. I already have 3 ideas:

    1. Using an infrared range finder. But as long as I now the precision is not very good.

    2. Using a piezoelectric sensor. This sensor is usually used to measure force from deforming the material. Can I use the deforming properties to measure the distance of the object from my device?

    3. Using visual measurement from a camera. I can get the picture of the object ahead my device, then calculate the real distance from the picture. I am not very sure about this idea because to get picture in small object like this will need very good camera and it will be very costly.

    What do you think about these ideas? Or there is any better idea?

    Thank you very much, sorry for my English.

    K. Sreenivas

    Piezoelectric micro-positioners are available commercially for a fine movement for different ranges, and are used quite commonly for fine movements of optical components. You can consult the web on linear displacement sensors and probably see if they are useful for you.

  • Vasile Surducan added an answer:
    Could a reflected wave damage a laboratory function generator?

    Recently one of our function generators (UTG9005C) started smoking and got slightly damaged (a couple burned transistors and resistors). 

    The function generator has 50 Ohms output impedance and was generating a square waveform with voltage 20 Vpp (this is pretty much the maximal output voltage) to a 50 m long coaxial cable with characteristic impedance 75 Ohms. The coaxial cable was terminated with an open circuit so there was a reflected wave.

    The coaxial cable was attenuating the reflected wave to less than 50% of the generated one (according to the oscilloscope), so the amplitude of the reflected wave must have been about 7-10 Vpp.

    Could this reflected wave be the cause of damage of the function generator? If yes then how would you recommend to protect it?

    I suppose one option would be to connect a resistor in series between the coaxial and the generator but i would like to hear other suggestions.

    Vasile Surducan

    It's difficult to say that reflected wave was the only cause of your problem. Try to repeat the experiment someday with the generator opened and see if the transistor  is heating during the procedure, this is common behavior on the final stage on reflected waves. Usually a reflected wave does not kill the transistor instantaneously...

    Have you used the DC offset voltage during the procedure?

    If the series resistor is protecting the output, you will not have the 2W in the load anymore... So, if your everyday purpose is to check cables (in the lab with the students...:)) , perhaps adding a well designed repeater at the function generator output will help.

  • Bhupendra Desai added an answer:
    What are the techniques to avoid overloading of distribution transformer?

    In rural India, some of the distribution transformer gets overloaded due to extensive use for agricultural purposes, especially for water pumping. This happens because the free electricity for the agricultural purpose is available only for a short duration on a particular day. Due to the stress, the transformer life reduces and premature failure occurs. One way to over come this is to have a overload relay to transformer protection than using a fuse which people will wire without following the specification.

    What are the other possible solution for this problem ? 

    Bhupendra Desai

    Refer my paper," Micro processor( or single chip microcontroller) based load sheading controller". It would give you scheduled hourly, priority based daily full load, control of week, without overloading, and would be monitored .....


  • Imran Khan Yousufzai added an answer:
    How we can select the sliding surface in Sliding Mode Controller?

    One of the robust and stable nonlinear controllers is sliding mode controller. This controller has two parts:

    1) Control mode: based on the sliding surface and a function to tracking follow for example in robot manipulator.

    2) Equivalent part: based on system's dynamic.

    So, what is the best method to select the sliding surface part? 

    Imran Khan Yousufzai

    The core consideration in choosing a sliding manifold is to check whether it supports the existence of "SLIDING MODES" in the given system's state space. Also you must verify whether the sliding mode dynamics does perform according to your application specifications.


  • Talita Conte Granado added an answer:
    PEDOT MEA - Why there is sudden increase in impedance for all electrodes?

    Hi, I bought a PEDOT 8x8 MEA form MultiChannel system two months ago and last month when I measured the impedance with their impedance measuring unit, I found substantial increase in impedance for all electrodes. In the past, the impedance was about 15kOhm for each electrode, and it increased to about 300 kOhm.

    I have always cleaned the electrodes with Terg-A-Zyme straight after experiment (retinal tissue) for 1 day and then store it in DI water. 1 day before experiment I do 2 mins oxygen plasma cleaning.

    The increase in impedance was sudden, i.e., the measurement done on 30th July showed most electrodes had impedance < 50 kOhm, and on the 18th August all electrodes had > 200 kOhm impedance.

    I do electrical stimulation with the MEA and sometimes I gave pulses that might be over the charge injection limit. However, it was only for a few electrode, so the increase in impedance should be limited to those stimulating ones, not all electrodes.

    Anyone know what is the cause? I am suspecting that there is loss of PEDOT coating. If so, what might be the cause for the loss of PEDOT coating?

    Talita Conte Granado

    I suggest SEM, or AFM. SEM will be a better option, because you can measure the thickness of the passivation layer and also see if you are having pin holes on it.

  • Sudipta Sarkar added an answer:
    Can we extract parasitic capacitances of a layout without an LVS check?

    I have a new device and I have drawn the layout for it. It doesn't have an associated schematic. Would it be possible to extract the parasitic caps for it?

    Sudipta Sarkar

    extraction can run without an LVS run for some tools.

  • Hao Ding added an answer:
    What is the difference between a microgrid and a smartgrid?
    Can anybody help me to know what the difference is between a microgrid and a smartgrid?
    Hao Ding

    microgrid considers the utilization of distributed energy systems in order to imrove the reliability and flexibility of the electricity. smart grid aims to combine intelligence technologies with electricity to imrpove the convenience and reliability of the whole electricity. micro grid is design for a small scale usually for a certain community while the smart grid is designed for the whole electricity system. They do not have any difference on the energy sources utilized.

  • Ezomar Uzcategui added an answer:
    Can someone provide a reference of Simulation of Medium Voltages (35kV) Cables in PSCAD?

    In my Job we are experiencing faults related to "cable termination damage" specifically in the Power Transformers entrance. I am seeking about models to simulate the over voltage in the "cable termination". Thanks a lot. EZOMARU

    Ezomar Uzcategui

    Thanks a lot sir. Feel free of ask or keep in touch. Greetings

  • Marco Bazzotti added an answer:
    How can I estimate the E-field from end-fed dipole antenna as described in the attached doc?

    There are two cylinders with different radius connected to each other. The excitation signal is located at the far-end of  the bigger cylinder and there is no ground plane in this structure?

    Marco Bazzotti

    I should use a (simple) simulation (e.g. with CST or similar software), being difficult to calculate directly.

  • Armin Joachim Gaul added an answer:
    How is Supply and Demand balancing done in decentralized markets?

    Utility System Operators in a decentralized energy markets accepts bids from Independent Power Producers IPP to meet the Demand in a timely basis. This is a known story, but interesting will be what algorithm they use or the logic they follow to reach a cost effective as well as considering carbon footprints emission restrictions. I would like to know whether there exists any such algorithm for utility system operators? Would be great, Thank you.

    Armin Joachim Gaul

    Maybe also these results from the project Smart Operator might be interesting for you-

  • Sujit K. Biswas added an answer:
    Is there a relation between ripples from a 3 phase bridge rectifier output and the harmonics due to the change in the load?

    i want to design a dc link capacitor for my 3 phase rectifier circuit.

    Also i want to do a harmonics study on 3 phase rectifier circuit 

    what is the load i have to keep at the output of rectifier circuit

    Sujit K. Biswas

    Assume diode rectifier and capacitive filter on DC. The ac source creates a 6th harmonic ripple voltage across the dc bus capacitor, which will be at 300Hz for ac 50Hz. At no load, the capacitor will charge upto the peak of the line-to-line voltage without being practically discharged, hence ripple will be negligible. As load current on dc side is increased through a resistive load, the capacitor tends to discharge from the peak voltage downwards towards 0.866times the peak value. It cannot discharge further down as the next phase voltage comes in to charge the capacitor up again after 60deg from the peak voltage. Hence, the dc voltage ripple magnitude (not frequency) will be dependent on the dc load current to some extent. For study, best is to determine the load experimentally. If the load is another switching converter, then both voltage ripple magnitude and frequency will be affected due to load.

  • André Roger Rodrigues asked a question:
    Does anyone have already worked with the Disruptive Effect Model to determine the empirical constants using the method of minimum variances?

    I have used MODELS to investigate how to determine the empirical constants (DE, Uo e K) from the mathematical equation of DE method. I am using method of minimum variance to determine the values of the constants aim to provide the best match between experimental and predicted V-t curves.

  • Barrie Gilbert added an answer:
    How to make 1st year electrical engineering student understand voltage, current, resistor?

    How to make 1st year engineering student understand voltage, current, resistor?

    Barrie Gilbert


    A)    Dear Fellow Travelers:


    B)   I agree wholeheartedly with Erik 


    C)   Understanding comes first from DOING


    D)   Young hands learn quicker than slow brains


    E)   The Velleman Projects looks like a great resource


    F)    I learned radio by disassembling WWII surplus equipment


    G)   The "THEORY" of wireless transmission largely came after Marconi


    H)    The best possible way to avoid the use of numbered items is to use letters


  • Xiangguo Li added an answer:
    How can I get the z-transform for the sequence a^|n| (a>0)?

    Hi all,

    the question is as the title, thank you in advance for giving an answer.

    I tried to decompose it into 2 sequences, anu[n] and a(-n)u[-n-1] (or should I have it as a(-n-1)u[-n-1] ), is this correct?


    Xiangguo Li

    Don't forget to determine and include the region of convergence (ROC).

  • Vasile Surducan added an answer:
    What are your thoughts on renewable ocean energy, such as wave, wind and tidal energy?
    For example, Marine Renewables off the coast of Ireland has the potential to produce over 70GW of energy, more than 14 times the country’s current energy demand. There are many such examples speaking about the viability of ocean energy. A floating windfarm off Coos Bay in Oregon was approved! Are there such attempts in your country, maybe realized projects?
    Vasile Surducan

    Perhaps the question can be somehow related to desalination of the sea water. I see in the future a complex plant on the sea using tide, wind, wave and solar energy, producing electricity, potable water and high quality salt.

  • Samvel Ter-Antonyan added an answer:
    How can I calculate the amount of needed voltage for electrostatically suspending a mass?

    Hi dears

    I'm going to design an electrostatically suspended mass system and I need to know the kinetics for calculating the amount of voltage that needs to apply on electrodes on top and below of the mass to create an electric field and suspending the mass.

    please help me as soon as possible 

    Samvel Ter-Antonyan

    Shahab, the micro size of suspended object (m,r) decreases the required voltage across the capacitor according to the formula above but creates big problems with a charge (q) of micro object.

    However, your sentence: "electrostatic suspended MEMS structure applicable for biomedicine" sounds odd for me, because a charged object can be suspended in the electrostatic field only for zero resultant force acting on it. The counteracting force in case of Albert and my examples above was gravitation force which, I guess, is not so effective in implantable biomedicine.

  • Adam Abramowicz added an answer:
    Why are resonances peaks spaced in the same range (whispering gallery mode spectrum in disk) ?

    I simulate whispering gallery mode by FDTD(Lumerical).

    I saw the dissertation that whispering gallery mode result by experiment

    and the resonances peaks are spaced same range. 

    why resonances peaks are spaced same range and  the radius of disk or thickness of disk have any impact on the peaks spacing

    Adam Abramowicz

    Peaks spacing depends on disk permittivity.

  • Amin Vahdat added an answer:
    Is it possible to tell comsol: only store specific parameters (like S11 in rf model)?

    i have performed a parametric sweep to evaluate S11 parameter with changing the width of my device, the simulation needs lots of memory and usually ends in error: out of memory! is it possible to tell comsol: only store specific parameters (like S11 in rf model)? and don't store the other parameter like the quantity of electric and magnetic fields in all nods and meshes in space?? because sometimes my simulation starts without error bur after some iteration the amount of needed memory increases and my PC can not provide the needed amount.

    Amin Vahdat

    Dear John 

    thank you for your suggestions

    actually the solver needs other parameters to compute S11 in each sweep step, but i am looking for a way to tell comsol when each step in parameter sweep finshes only store S11 and delete other field parameters and continue to the next step of sweep. so finally only S11 parameters are stored in whole sweep.

  • van-tuan Doan added an answer:
    What is the purpose of using 1.2 here in sine pwm in this simulink. In space vector pwm also I saw source voltage divided by 1.8. Why is it used?

    Please explain with sufficient theory

    van-tuan Doan

    scaling factor is for adjusting the modulation index of that rectifier. The amplitude of carrier is fixed = 1. The amplitude of modulation waveform is adjusted by that scaling factor. maybe 1.2 or 1.8 is a given value within the range

  • Bekka Nassim added an answer:
    How is sizing of electric motors done for a motor running at variable load to achieve maximum efficiency?

    Motors frequently need to be operated under varying load conditions due to process requirements. Mostly a motor is selected based on the highest anticipated load. But the highest load can occur only for a short period of time. Instead a motor can be selected with a rating slightly lower than the peak anticipated load which can operate at overload for a short period of time. Example: For a particular load cycle, the average load is around 69kW with peak load(180kW) occurring for a few seconds. For this load cycle, it won't be cost efficient if we select a motor with rated power around 180kW. What is the appropriate way to select a proper sized motor that would operate at its highest efficiency for maximum time?

    Bekka Nassim

    You can choose the RMS value of torque and speed during the cycle

  • E. A. Ogujor added an answer:
    What will be the change in the Line Data parameters in the IEEE standard Bus system if any line is considered as outage?

    What will be the change in the Line Data parameters in the IEEE standard Bus system if any line is considered as outage?

    E. A. Ogujor

    There will be a change in the bus impedance and admittances to which the line was connected and hence change in the power flow

  • Mahdi Salimi added an answer:
    How can I increase the gain of a DC-DC Converter?

    I have designed a Boost Converter, now I want to increase the converter gain without changing the converter parameters. Is it possible any way?

    Mahdi Salimi

    if you increase the converter duty cycle, voltage gain of the converter will be increased.

    if you have designed the parameters, you must improve elements structure and design.

    for example, if equivalent series resistances of the inductor and capacitor is reduced, the maximum accessible gain will be increased.

    Also, you can employ low voltage switches and diodes.

    however, in the standard boost dc-dc converter, maximum voltage gain is limited to 4-5.

    if more voltage gain is required, you can use other dc-dc boost topologies like Switched Inductor DC-DC boost converter.

    good luck

  • Zol Bahri Razali added an answer:
    When is the vertical vector potential due to horizontal electric dipole negligible?

    Dear Colleagues, speaking about electric dipoles, we know VED produce only vertical vector potential while HED produce both horizontal and vertical vector potentials.

    My question is related to the HED vertical vector potentials.

    In my calculations, it seems that HED vertical vector potential is often negligible.

    I know solenoidal HED does not produce vertical vector potentials and probably at low frequency aerial or insulated souces can be considered solenoidal.

    Non insulated buried sources are not solenoidal but, if I consider an horizontal conductor divided in short elements, it seem that vertical vector potentials effects of each element are quite deleted from the adjacent elements and only the ends elements produce vertical vector potentials.

    I very appreciate your suggestions.

    Best regards Roberto Andolfato

  • Negin Kananizadeh added an answer:
    What is the Comsol Error Unknown function or operator; Name: intop1 - Feature: Time-Dependent Solver 1 (sol3/t1)?


    I'm defining a voltage in Component 1 > Definitions > Integration as

    Vip = intop1((comp1.mf.Ex*ix+comp1.mf.Ey*iy+comp1.mf.Ez*iz)*Np/area)

    I also tried defining Vip in global variables as


    But I get the errors:

    Unknown function or operator.
    - Name: intop1
    - Feature: Time-Dependent Solver 1 (sol3/t1)


    Internal geometry error.
    Error in multiphysics compilation.

    I have attached the file,

    Thanks a lot

    Negin Kananizadeh

    Hi Robert, I am glad that you have figured it out. 

  • Jitendra Nath Rai added an answer:
    How can I solve five non linear equation roots with five equations with newton raphson method?

    where x1,x2,x3,x4,x5 are angles in degree..

    How to solve it with Newton-Raphson ? How to write these functions in matlab coding?

    Jitendra Nath Rai

    You can use the least-square method as suggested byGraham W Pulford.

  • Srikanth Reddy added an answer:
    How can wind-pumped storage coordination increase dispatchability in a day ahead programming,whereas the wind curtailment is not definite yet?

    How can wind-pumped storage coordination increase dispatchability in a day ahead programming,whereas the wind curtailment is not definite yet one day before real time?

    In other word, available wind power is not definite when ISO will determine commitment of day ahead program. on the other hand, wind curtailment is the difference between available and forecasted wind power.

    Srikanth Reddy

    Curtailment is more often defined as a state of wind generators not being able to transmit/dispatch the wind generation due to limitations of transmission network or the load is being already met by the committed generation other than wind. But, the context of pumped hydro storage is more related to managing variation in predicted and actual generation (higher or lower). Therefore, curtailment is more of an issue related to infrastructure/transmission/connectivity of network. Whereas, storage is related to capitalization of wind energy in case of higher actual generation and improvement in dispatchability in case of lower actual generation than the predictability. In case of transmission congestion, dispatchability becomes the job of transmission system operator who takes over the day ahead market operator to alleviate the congestion through real time locational marginal prices. 

  • Alex Van den Bossche added an answer:
    How does the operating frequency influence the leakage inductance between two coils?

    We know that leakage inductance depends on the distance between the two coils and coil/size ratio. If we consider two loosely coupled coils, what does the frequency vs leakage inductance curve look like?

    Alex Van den Bossche

    For low frequencies, the current is homogeneous in a conductor. The inductance contribution in the conductor itself is it is mu*1/(4*pi)*length. In conductors far away from other conductors this contribution vanishes at high  frequency. This is an example of transmission lines at low frequency, one observes an inductance in coaxial wires even if inner touches outer.:


    One observes an inductance even if wires touch. If conductors are close, the eddy currents tend to let the current flow on the surface side close to the other conductor, so on sides and concentrated, so the leakage gets the difference low-high frequency is more than mu*1/(4*pi)*length. In a coil, wires screen somewhat the field and the inductance is reduced.

    Bessel type equations give the impedance for a stand alone wire. The imaginary part contains the inductive part. In a conductor in transverse field, one can get also other Bessel equations. I deal with this in chapter 5 of the book "Inductors and transformers for power electronics", but I did not emphasize on inductance change as it has a minor effect on losses.

    Impedance functions are analytic functions, so real and imaginary part are linked. So if one knows the change of the resistive part of the impedance, one could predict the inductance change.

    Accurate finite elements can model it, but one needs a fine mesh at the conductor surface to model well the eddy currents. If there are many wires and strands finite elements get problems.

  • Sounik Kiran added an answer:
    What is the difference between Returnloss(dB) and S11(dB) ?

    Actually some research papers have written Returnloss(dB) and some others have written S11(dB). Couple months back I had seen there was an article in IEEE which has given the clarification. If any one have any information kindly tell me.

    Sounik Kiran

    Thank You Xavier Le Polozec, Abdelhalim Zekry, Sandra Cruz-Pol  for your suggestive answers.

Topic followers (178,943) See all