my bipolar switching for SPWM single phase h-bridge.

[rockyou]

Hi everyone,happy new year 2010.


how come my circuit generate wrong output at R2?im expect it to be oscillate betweeen 2 polarity,but from the simulation,the pulse is between 8V-14V.
How do i fix this problem?

appreciate if someone can help me solve the problem.
Thank you.

 

[FvM]

Start with a known functional circuit from the forum.

 

[svhb]

Start with checking the output voltage of U1A. Probably the model of the 7416 will only generate 5V pulses.

 

[FvM]

Start with checking the output voltage of U1A.

That's one of many circuit faults. Other points are:

- Unsuitable drive voltage respective missing current limiting resistors for the low side switches.
- Unsufficient drive voltage for the high side switches
- Floating supply voltage of the power stage

 

[rockyou]

so basically my circuit is totally wrong?
i have to find other circuit in this forum?

 

[FvM]

Your intention isn't completely clear. It's very easy to build a working simulation circuit, because you can use floating sources
and other ideal elements. I doubt, that you're actually targeting to an application with kohm load and small signal output
transistors. Of course it's possible to build a bipolar H-bridge, either with complementary or NPN-only output transistors.
But the driver design would be based on the actual current and voltage level. It's also unclear, what's your motivation to
use BJT instead of MOSFET. Most edaboard H-bridge circuits are obviously using MOSFET.

 

[rockyou]

i not sure bjt or mosfet from the circuit in pdf.the symbol quite looks like a bjt,so i jz put it.im also refer to other type of the h-bridge circuit,most of them is using the mosfet,i agree with u.so right now i have to modify my current circuit,change the bjt to NPN mosfet.

my intention is to generate the pwm for bipolar and unipolar as shown in the pdf document.my amplitude sine wave voltage=2V,while amplitude for triangular wave voltage is 2.5.

 

[FvM]

The paper is showing IGBT switches, they are integrating a bipolar transistor and a MOS driver transistor on a chip. Their field of
application is a few hundred to multiple thousand V and a few ten up to thousand A. Apart from a different input capacitance,
they are driven like MOSFETs.

The paper is focussing on unipolar versus bipolar PWM scheme and not showing any driver details. Actually, the high-side
switches have to be driven by a level translator not shown in the principle schematic. IR2110 boostrap drivers are
an example of integrated drivers suited for MOSFET and small IGBT.

You'll find several dozen threads about H-bridge drivers when searching edaboard.

 

[rockyou]

Start with checking the output voltage of U1A. Probably the model of the 7416 will only generate 5V pulses.

so u mean the 7416 have to replace with other component?

Added after 13 minutes:

The paper is showing IGBT switches, they are integrating a bipolar transistor and a MOS driver transistor on a chip. Their field of
application is a few hundred to multiple thousand V and a few ten up to thousand A. Apart from a different input capacitance,
they are driven like MOSFETs.

The paper is focussing on unipolar versus bipolar PWM scheme and not showing any driver details. Actually, the high-side
switches have to be driven by a level translator not shown in the principle schematic. IR2110 boostrap drivers are
an example of integrated drivers suited for MOSFET and small IGBT.

You'll find several dozen threads about H-bridge drivers when searching edaboard.

im start searching since yesterday,but when i type h-bridge keyword,only 34 result match.most of them is not what i want,and the method is used microcontroller,dsp and chip to get the pwm.
the IR2110,cannot found in the pspice student version,so i decide to put IRF150 npn mosfet into my circuit,when somehow cannot work.the error keep pop out by stating that the node is floating for 4 mosfet.
when the thing come in by schematic,i found that it quite hard to simulate by to get the result like the theory shown in the book.

 

[FvM]

Yes, replacing the BJT by MOSFET's only removes one of many circuit errors mentioned above.

I don't understand, why you continue using a TTL inverter? Seems like you are somewhat resistant to advice.

 

[rockyou]

haha,no la.actually my final year project.the topic state need to use analog component 1,cannot use microcontroller or programming method to build the circuit.
so ur suggestion is not trying to use TTL inverter?

 

[karesz]

Hi,
TTL is so & so a problem!
If (logic)inverter; then CMOS, then TTl has logic levels =<3V--for most MOS isnt enough as Ugs(...
K.

P.S.:
If you have analogs to use, why not an inverting OP?

 

[FvM]

Apart from the TTL inverter problem, it's completely unclear what you are trying to achieve. The absurd "load impedance"
of 12k suggests some kind of principle simulation, so why not using controlled SPICE voltage sources to drive the
individual FETs?

If it's a step to design a functional circuit, then you should have gate drivers with sufficient current capability, made
from real components. IR2110 was just an example, you can use a few discrete components instead, example circuits
have been discussed at edaboard. Their performance is quite different, but you're invited to learn about these differences.
In any case, you should understand both the SPICE simulator and pratical H-bridge requirements. You can hardly
do a final years project without understanding the circuit basics.

 

[rockyou]

my inverting OP use to build my sinewave form circuit and triangle waveform circuit.
so ur suggestion inverting OP is replace the MOSFET?
i agree that TTL circuit is so many problem,i have no choices since most of the circuit i found is like that configuration.

 

[FvM]

Clarify the purpose of your circuit first.

 

[rockyou]

i had done the triangle circuit and sine wave circuit,which is the 1st phase for overall.the h-bridge is later phase,i had really read the pwm book borrow from library...after study it,i thought it is going to be easy.after that i searching schematic online for h bridge.I found the it,the file i attached at post number 7.haiz...i have to done this in 2month,kinda rush.so the alternative way is asking at forum.hope to get some feedback from member and let me think what the problem have to concern in this project.so so sorry about my silly question,i will further my investigation regarding h-bridge.if any problem,i will get back again.thank you.