Why Are Half-Bridge Capacitor Voltages Not Vs/2 in IR2153 LTspice Simulation?

423 9

Best answers

Why are the two half-bridge capacitors in my IR2153 LTspice simulation not holding half the supply voltage, and what is causing the midpoint to shift?

The capacitor voltages are not staying at Vs/2 because the half-bridge midpoint only remains centered when both switches transfer identical charge over time; any mismatch in switch ON time or duty cycle makes the midpoint walk toward one rail. A tiny imbalance is enough to do it, so if one switch conducts slightly longer than the other, one capacitor slowly charges while the other discharges [#21671372] [#21671374] The poster was also advised that the IR2153 datasheet drives a series LC load against circuit common, while this schematic drives a capacitive divider, so the circuit should be tested in a simpler form first or driven against circuit common / a half-supply source [#21671378] One reply also noted the circuit is resonant at about 1.7 kHz with Q≈3, so the switcher should be operated at the intended resonant frequency and the circuit checked without the capacitive midpoint before troubleshooting further [#21671380]

AI summary based on the discussion. May contain errors.

Report a violation of the law

Reply | New topic

Questioning why half-bridge capacitors are not Vs/2
#1 21671371 22 Apr 2014 09:06

ak min ak min

Anonymous

Questioning why half-bridge capacitors are not Vs/2
Post #1 21671371 22 Apr 2014 09:06

Hi

Im doing a project based on half bridge converter of IR2153 which could produce a high output voltage using the LTspice simulation by means of half bridge topology. Well, im just wondering as the volatge across the two capacitors that is pointed in the image shown wasnt give half the volatge source as stated in the theory of half bridge voltage source inverter. Can someone point me out and explain why the volatge across these two capacitor was not Vs/2? Does it something to do with some interference produced by the transformer in the cicuit? Thanks in advance.
ADVERTISEMENT
#2 21671372 22 Apr 2014 14:20

DAVID CUTHBERT DAVID CUTHBERT

Anonymous

Post #2
21671372 22 Apr 2014 14:20

For the center of the half-bridge to remain centered the charge through each switch must be identical over time. Otherwise the center point will walk to one rail.
ADVERTISEMENT
#3 21671373 23 Apr 2014 08:30

ak min ak min

Anonymous

Post #3
21671373 23 Apr 2014 08:30

What do you mean by the charge through each switch must be identical? As far as i know, when Im changing the value of resistor R3, for the Rt pin in the driver IR2153, the voltage across the two capacitors would also change.
ADVERTISEMENT
Duty-cycle mismatch shifts half-bridge capacitor midpoint
#4 21671374 23 Apr 2014 09:28

DAVID CUTHBERT DAVID CUTHBERT

Anonymous

Duty-cycle mismatch shifts half-bridge capacitor midpoint
Post #4 21671374 23 Apr 2014 09:28

If the charge thru each switch differs one or the other capacitor will slowly charge as the other discharges. Any difference in switch ON time will do this.

This is the big problem with the half-bridge and some designs incorporate circuitry to slightly adjust the ON time of one switch to keep the capacitor junction at the midpoint.

You can experiment with this by removing the IC and driving the FETs with pulsed voltage sources. Slightly change the duty cycle of one switch (I.E. 20% for one and 21% for the other) and you will observe the capacitor midpoint walk toward one rail.
ADVERTISEMENT
#5 21671375 23 Apr 2014 11:34

ak min ak min

Anonymous

Post #5
21671375 23 Apr 2014 11:34

By saying that, the capacitor did resonating but did not obey the voltage source inverter. What factor that disobey this voltage source inverter again because i am still a bit confused.
#6 21671376 23 Apr 2014 11:43

DAVID CUTHBERT DAVID CUTHBERT

Anonymous

Post #6
21671376 23 Apr 2014 11:43

Can you post a larger schematic? I cannot read the component values.
#7 21671377 23 Apr 2014 13:18

ak min ak min

Anonymous

Post #7
21671377 23 Apr 2014 13:18

This is the schematic diagram of my half bridge power supply.
#8 21671378 23 Apr 2014 15:29

DAVID CUTHBERT DAVID CUTHBERT

Anonymous

Post #8
21671378 23 Apr 2014 15:29

The IR2153 datasheet shows the driver driving a series LC against circuit common. You have it driving against the capacitive divider. Have you tried driving against circuit common? Or against a voltage source equal to one-half the supply?
Transformer loading shifts half-bridge capacitor voltages
#9 21671379 23 Apr 2014 16:02

ak min ak min

Anonymous

Transformer loading shifts half-bridge capacitor voltages
Post #9 21671379 23 Apr 2014 16:02

As I know, i altered the swtiching frequency by changing the resistor connected to Rt pin in order to get higher freqeuncy in the circuit. Having said that, what should be the cause of the problem in my simulation? As i measured the frequency resonant, it differs from what i obtained when i plot a graph of output volatge against frequency. And yes, when the circuit is not connected to the transformer, the voltage source equal one half the supply.
#10 21671380 25 Apr 2014 07:32

DAVID CUTHBERT DAVID CUTHBERT

Anonymous

Post #10
21671380 25 Apr 2014 07:32

Your circuit is resonant at 1.7 kHz with a Q of about 3. Are you operating the switcher at that frequency?

I would get the circuit working without the capacitive midpoint and proceed from there.

But note that unless the charge transferred by each switch is equal there will be a net DC charge injected into the capacitors and they will charge. The midpoint will "walk" toward one voltage rail.
Create an account, log in here. You will receive points by participating in discussions.
Join this discussion.

Install Elektroda application

Didn't find an answer? Ask Artificial Intelligence (AI)

*I agree to send the question to OpenAI, Google LLC - owners of language models in order to prepare the best response. The companies may monitor and log information entered into the form.

*I agree to publicly display my question and answer. The question and answer will be publicly available to everyone. The process may take a few minutes. Upon completion, you will be redirected to the page with the answer.

Wait...(2min)

Reply | New topic

Report a violation of the law

Topic summary

The discussion addresses why the voltages across the two capacitors in an IR2153-based half-bridge converter simulation in LTspice do not equal half the supply voltage (Vs/2) as theory predicts. The key issue is that for the midpoint voltage to remain stable at Vs/2, the charge transferred through each switch must be identical over time. Any imbalance in switch ON times causes one capacitor to charge while the other discharges, resulting in the midpoint voltage drifting toward one rail. This phenomenon is exacerbated by resonant effects and the presence of the transformer, which can alter the expected voltage distribution. Adjusting the switching frequency via the resistor on the Rt pin affects the capacitor voltages, and the circuit's resonant frequency and quality factor (Q) influence operation stability. Some designs include circuitry to compensate for ON-time differences to maintain the midpoint voltage. It is also suggested to test the circuit without the capacitive midpoint or transformer to isolate the cause. Driving the IR2153 against circuit common rather than the capacitive divider may also impact the voltage balance.
AI summary based on the discussion. May contain errors.

Why Are Half-Bridge Capacitor Voltages Not Vs/2 in IR2153 LTspice Simulation?

Why are the two half-bridge capacitors in my IR2153 LTspice simulation not holding half the supply voltage, and what is causing the midpoint to shift?

Didn't find an answer? Ask Artificial Intelligence (AI)

Topic summary