[SOLVED] ir2110 output problem

[fedaye]

hi guys
i have a circuit like the pictures below. one of output of the ir2110 (LO) is ok but the another one (HO) is not ok. there is a problem i could not understand. outputs of sg3525 are ok (i observed via a scope)

 

[Anna Conda]

until you run the low side fet, there is nothing to charge the upper boot strap cap so the upper side of the 2110 will not work - no Vcc...!

By the way the 1N4007 may not be a good choice for the boot strap diode....!

 

[fedaye]

sorry, i could not understand you, can you say me what should i do?
also the diodes are uf4007
is this situation normal? because both of the outputs are not connected any mosfet, any load but one of them is ok

 

[Orson Cart]

you have to actually connect up the bottom mosfet and run both hi and lo in order for the upper gate drive to work......!

 

[red_alert]

For testing purpose only, you can connect the Vs terminal of IR2110 to COM (ground) then you can measure the high-side output, too. Are you driving a half-bridge? SG3525 is not suitable for that (its outputs are not complementary).

 

[fedaye]

For testing purpose only, you can connect the Vs terminal of IR2110 to COM (ground) then you can measure the high-side output, too. Are you driving a half-bridge? SG3525 is not suitable for that (its outputs are not complementary).

yes, but are you sure? i observed it and the outputs of sg3525 seemed like 2 square wave that was in a period. there was not any intersection for any duty ratio or frequency
i ran it with a resistor as load and one of the mosfets was very hot and the another one was lesser hot there was an explicit difference.

 

[red_alert]

the outputs of sg3525 seemed like 2 square wave that was in a period. there was not any intersection

There are no intersections but they're not complementary either. Both outputs have the same duty cycle (0 to 49%) but they are turning on alternately.

By example, for a duty cycle of 20%, first output will generate a pulse (ON state) with a duration of 1us followed by a pause (OFF state) of 4us. Next, the second output will generate the same pulse timming while the first output remains in OFF state. Thus you've got a 4us (80%) deadtime between every alternately output pulses.

is it convenient to connect the outputs of sg3525 each other and then using an inverter like 74hc04 for this.

There were some recent discussions on this area about driving a half-bridge using a SG3525. Summing the outputs it's not enough as you also need some deadtimes between low-side/high-side signals transitions.

 

[fedaye]

There are no intersections but they're not complementary either. Both outputs have the same duty cycle (0 to 49%) but they are turning on alternately.

does complementary mean two waves with %100 duty cycle (50+50)
what do you recommend to me? this is a stuff about my graduation project i have to do this somehow :bang:

 

[red_alert]

Complementary means that during a period, when one output is HIGH the other one it's LOW and vice versa.

What exactly it's your graduation project about?

 

[fedaye]

Complementary means that during a period, when one output is HIGH the other one it's LOW and vice versa.

What exactly it's your graduation project about?

it is a solid state tesla coil. i planned to use sg3525 and ir2110 with a half bridge (200 kHz) but you are saying this is not a good choice

- - - Updated - - -

is it possible to generate complementary waves without microchips?

 

[red_alert]

I've posted some examples (posts #11 and #13) in this thread:

https://www.edaboard.com/threads/330586/

For your switching frequency (200kHz), a simple reverse diode across the gate series resistor could be enough to generate some deadtime.

 

[fedaye]

same situation is still valid. i did them and connected a load but there is no change. what is the reason of this? one of the Vgs is ok but high side one is fixed a value like the pictures (#1). i believe this is not normal. :bang:

 

[Orson Cart]

200kHz is too fast for the 2110 chip (it won't work or will die after a while) and 400kHz is too high a clock speed for the PWM chip you are using 3525, the output will be all dead time and not much actual power pulse. You need to start again without these chips if you really want 200kHz switching...
We have a discrete circuit we have developed that runs at 130kHz, might go to 200kHz...

 

[Benn Ng]

hi, any solution to the mention topic? . able to further explain how do i connect the lower FET to test the HO pin 7 of the ir2110? i got the same output waveform mentioned by TS. please advise

 

[fedaye]

hi, any solution to the mention topic? . able to further explain how do i connect the lower FET to test the HO pin 7 of the ir2110? i got the same output waveform mentioned by TS. please advise

check your capacitors' connections, maybe you have made a mistake about + and - connections of the half bridge's capacitors.
my simple mistake was this.

 

[fedaye]

you can use sg3525 and ir2110 for a solid state tesla coil.
be careful, high voltage is dangerous.

 

[vinnu]

hi,
I am also facing the same problem. I am getting pulse at LO but not at HO. I even connected a FET to LO side and it works properly. but when HO was measured at the same time there was no pulse. Pls help me. I have attached the circuit that I have used. I am using 110 nF bootstrap capacitor. also tried increasing the value no effect.
Regards,

Vishnu

 

[FvM]

Bootstrap doesn't work without the low side switch connected + activated alternatingly with the high side.

You don't show the complete circuit or control waveform, so we have to assume that it's no valid bootstrap setup.

You can short Vs to COM to power the highside driver for test.