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.

Having problem in driving mosfet

Status
Not open for further replies.

hariecengg

Junior Member level 1
Joined
Mar 1, 2013
Messages
19
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
India
Activity points
1,505
Hi Geeks,
I am doing a 250V 50Hz 500W bipolar spwm inverter project using Atmel microocntroller .I successfully generated signals and using look up tables to control output voltage.Now the problem is when i apply spwm signals to the mosfet the mosfet does not turn fully on.I checked the gate voltage it is around 3.15v (Vavg) at spwm duty cycle 70%. The maximum current i am getting only 3A.I tried paralleling of 6 mosfets no improvement only increased 1A more.
So 12V x4A =48W
i am using irf540 logic mosfet and TLP250 driver stage before the mosfet.
How can i increase the current ? what i am doing wrong ?Googled a lot and read many forums but no help !
 

Attachments

  • Capture.JPG
    Capture.JPG
    38 KB · Views: 92

Hi,

It mainly depends on PCB layout, schematic and so on.
Please post it.

Klaus
 

Hi,

what topology are u using, single conversion, double conversion, what voltage are you inverting from. you have to say. Many times what you give is what you have. even if u parallel a 100 Mosfets if there is no where to source it you cant sink it. Thats one of the rule of thumb in Electronics. Enumerate further share your idea very well then people can know how to help you.

Tunde
 

Hi babatundeawe thanks for your reply.I am using single conversionand the voltage i am using to inverting is 12V DC .Converting 12v to 250V AC Sine wave.From microocontroller to driver then to mosfet and the outputs from drains are connected to (12-0-12)V/250V (600W) Transformer
 

Hi KlausST,
I am testing everything with breadboard

- - - Updated - - -

Hi babatundeawe thanks for your reply.I am using single conversionand the voltage i am using to inverting is 12V DC .Converting 12v to 250V AC Sine wave.From microocontroller to driver then to mosfet and the outputs from drains are connected to (12-0-12)V/250V (600W) Transformer


And when i checked with SG3525 a normal pwm waveform with 50% duty cycle the mosfets are fully on and a single mosfet giving current of about 10-15Amps of current as the drain output and on paralleing the 5 mosfets getting easily about 60-70A.
But the same mosfet is driven by spwm gives only 1.28v at the microocntrollers pin. This 1.28v is a average voltage not a logic high level voltage and after using mosfet driver i am getting 3.28 as Average voltage(Vavg).Everything i measured using Oscilloscope. How can i increase the average voltage.So that i can turn on the mosfet fully.The spwm frequency i generated is 3khz.At 100% duty cycle it gives only 4.10V(avg). The problem i am facing now is not able to get enough current from the drain of the mosfets.
I tried using IR2110 instead of TLP250 when i connected with the gate of the mosfet it IR2110 damaged in a second.Any one will give me the right schematic for IR2110.I used IR2110 as low side driver
 

@ Sunnyskyguy

Exactly its similiar to the schematic of yours.But gate to source i have used 10K Resistor and 47ohms for gate resistor.But i am not using SG3524/25 .I am using SPWM signals from microcontroller

The schematic you posted like i tested before and its working without any problem.I dont find any difficult in driving mosfets.The output from SG3524 is just a square waveform with complimentary outputs.So there is no Vavg problem here to drive the mosfets by varying the duty cycle of waveform we can simply adjust the output voltage.

But in spwm having Vavg problem.I am not able to get enough voltage to turn on the mosfet fully on.
Like babatundeawe mentioned in his post even if i paralled a 100's of mosfet there is no use because here the Vavg is very low.If i able to turn on the single mosfet to fully ON .Then it is easy to increase the current by paralleing mosfets. I am getting only 30% of current as output in drain..Please suggest me any ideA

Thanks
 
Last edited by a moderator:

I have no idea why you are always talking about average voltage. For a MOSFET switch, gate-source high voltage and low voltage matters, average value varies with duty cycle. The important point is to achieve at least 8, better 10 or 12 V Vgs high level and < 1 V low level. What do you see in your circuit?
 

Yes FVM as by your point how can i achieve 8v at the gate.The maximum voltage so far i got by TLP250 is 4.3V and by buffers after totem pole transistor configuration same result. Am i need to increase the power supply voltage for TLP250 so that can i achieve 8v ?
 

Hi,

I don't hink a breadboard is useful for driving mosfet circuits. The path Fet_drain, driver, Fet_gate is critical. Low impedance is essential here.

Then you say it is about the same schematic...
But driving a mosfet gate needs much more pulse current than a microcontroller can output. Also the voltage level is critical.
A microcontroller output is usually 3.3V or 5V...with no load.

You are talking about AC sine wave output, i think it is more likely DC output...

Klaus
 

Hi KlausST,
For testing purpose i have everything separately built in PCB. Microcontroller in development board and mosfets in PCB . So in btw to drive the mosfets which already i have in PCB i am only using Buffer IC and TLP250 in breadboard.

From ucontroller signals to ----> drivers in breadboard---->output to mosfets in PCB

So where i am doing wrong ?

And yes u are right without load the logic level of microocntroller signal is 3V to 5V and after the driver stage the logic level is 12v.

On measuring in Oscilloscope the signal level is 12V but Vavg is very low getting only 3 to 4v.

But i am confused as FVM said to Fully ON mosfet only the high level matters.Then why i am having trouble in getting current not more than 3amp with 90W load
 

Yes FVM as by your point how can i achieve 8v at the gate.The maximum voltage so far i got by TLP250 is 4.3V and by buffers after totem pole transistor configuration same result. Am i need to increase the power supply voltage for TLP250 so that can i achieve 8v ?
TLP250 is specified for minimum for minimum supply voltage of 10 V. You surely get more than 4.3 V output with 10 V supply, or something is very wrong with your circuit.
 

Hi,

By reading
But i am not using SG3524/25 .I am using SPWM signals from microcontroller
I thought you don't use any driver..

Any one will give me the right schematic for IR2110
Isn't it in the datasheet. If not you will find a lot of application notes on driving mosfets on the IR.com pages.

drivers in breadboard---->output to mosfets in PCB
Here high current pulses may cause problems when the lines are not short and wide enough. Low impedance.

On measuring in Oscilloscope the signal level is 12V but Vavg is very low getting only 3 to 4v.
As FvM said: Avg voltage makes no sense here.

Klaus
 

Hello , i know there are schematics in datasheet and in application notes.IR2110 gets hot and damaged in a fraction of secs when i connected to gate of the mosfet that is why still i am using TLP250. Because here IR2110 costs high thats why i am asking 100% tested working circuit

2.jpg

This is the circuit i tested and failed.
Does anyone know when should i have to go for IR2110 instead of using transistors or TLP250 or any mosfet driver ics.
Any advantages ??
 

Hi KlausST,
For testing purpose i have everything separately built in PCB. Microcontroller in development board and mosfets in PCB . So in btw to drive the mosfets which already i have in PCB i am only using Buffer IC and TLP250 in breadboard.

From ucontroller signals to ----> drivers in breadboard---->output to mosfets in PCB

So where i am doing wrong ?

And yes u are right without load the logic level of microocntroller signal is 3V to 5V and after the driver stage the logic level is 12v.

On measuring in Oscilloscope the signal level is 12V but Vavg is very low getting only 3 to 4v.

But i am confused as FVM said to Fully ON mosfet only the high level matters.Then why i am having trouble in getting current not more than 3amp with 90W load

Hi;

I guess you re not designing the inverter for NASA, or is it a top milltary secret? If its not put a schematic and or something and let people see what you are doing right or wrong. i dont think anyone can know what you dont let show. atleast sunnyskyguy put up a schematic for you to compare with yours.

put up something scheme, snap your PCB and let people see. you make things easier, rather than letting guys do the guess work of what you re doing or not do.

Tunde
 

This is the circuit i tested and failed.
I don't understand how the circuit is related to the original SPWM problem. IR2110 is a bootstrap driver and doesn't work well for a pure highside switch.
 

Hello Dude :-D Seriously i am not designing inverter for NASA :-D .Actually i dont have schematic just checking everything with development board .Only to generate spwm signal is used proteus for simulation . K i will draw and post the schematics soon :)
 

Hello , i know there are schematics in datasheet and in application notes.IR2110 gets hot and damaged in a fraction of secs when i connected to gate of the mosfet that is why still i am using TLP250. Because here IR2110 costs high thats why i am asking 100% tested working circuit

View attachment 117823

This is the circuit i tested and failed.
Does anyone know when should i have to go for IR2110 instead of using transistors or TLP250 or any mosfet driver ics.
Any advantages ??

Hi;

I dnt know what you re using the resistor for but what i can say is that IR2110 is a high and low side driver and for inverter you will need two. if using a fullbridge mode or just one for push-pull (and i dont think you ve specify the mode you re using).

attached is how i use to do mine and for the while it has been working good.

hope this helps

tunde
 

Attachments

  • Diver.pdf
    24.8 KB · Views: 119

To babatundeawe :-D i am not designing any inverter for NASA and it is also not a Top military sectret

Help me now, i am attaching the schematic here,..
Untitled.jpg

Any sugestions ? Dont know where i am doing mistake

- - - Updated - - -

Thanks dude..I will test ur schematics and tell about the results soon..And i am not using Full bridge mode. Using only Half bridge .

- - - Updated - - -

This is the output i am getting from transformerCapture.JPG
 

If you don't get at least 10V on the gates, there is something wrong with your design/layout as mentioned before.
The TLP250 should have no problems driving the gates provided you are following all the recommendations. Have you complied with note 5?
 

Attachments

  • note5.png
    note5.png
    47.4 KB · Views: 150

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top