I returned to this circuit and I think it's not suitable for low duty cycle. I wanted have DC voltage when duty cycle is about 1-5%. Maybe I should create true buck converter and use relay to change polarisation?

 

I returned to this circuit and I think it's not suitable for low duty cycle. I wanted have DC voltage when duty cycle is about 1-5%. Maybe I should create true buck converter and use relay to change polarisation?

Hi,

Just curious, why do you need a 1 percent duty cycle?

 

I returned to this circuit and I think it's not suitable for low duty cycle. I wanted have DC voltage when duty cycle is about 1-5%. Maybe I should create true buck converter and use relay to change polarisation?

It will work exactly the same.

 

Hi,
this circuits drivers Peltier module. The duty cycle is being set by PID. When Peltier reaches temperature, duty cycle is about 0-5%.

 

Hi,
this circuits drivers Peltier module. The duty cycle is being set by PID. When Peltier reaches temperature, duty cycle is about 0-5%.

Hello again,

Oh ok, that makes sense.

A regular buck circuit should be able to do that, but since your H bridge circuit was going to be driven like a reversable buck circuit, I don't see why it would not work just like a buck.

What led you to believe that the H bridge circuit would not work for this application?

 

I suggested the simulation results:

During low duty cycle, there are voltage spikes, not stable voltage.

 

I suggested the simulation results:
View attachment 300910
During low duty cycle, there are voltage spikes, not stable voltage.

Hi,

Which circuit are you using for those simulations? Even the 50 percent test doesn't look very good.

 

Hi,
I'm using this circuit in simulation:

 

Hi,
I'm using this circuit in simulation:
View attachment 301082

Hi,

Where is your filter capacitor?

 

See how you can filter it on 31.4 kHz:
____
ADDED:

I would feel more comfortable with 50uH or even 100uH, but that of course would be more expensive.

Especially for you: inductor S14100036, 56 uH, 17 A, Digikey, $148.00 for 16 pieces. https://mm.digikey.com/Volume0/opasdata/d220001/medias/docus/1284/S14100036.pdf

Or inductor IHB6BV101K-ND, 100 uH, 16.4 A, Digikey, $477.57 for 16 pieces. https://www.vishay.com/docs/34015/ihb.pdf

 

See how you can filter it on 31.4 kHz:
View attachment 301125

Hi there,

That looks interesting. I was just wondering, did you happen to check the ripple current through C1?
I ask because the cap has to handle that as long as the circuit is running. I would feel more comfortable with 50uH or even 100uH, but that of course would be more expensive.

 

Nice, thanks.
Sorry, I should have told that the load changed to 50W only.
The capacitor does the thing. The only matter that worries me in your circuit @Danko is unpolar operating.

 

Nice, thanks.
Sorry, I should have told that the load changed to 50W only.
The capacitor does the thing. The only matter that worries me in your circuit @Danko is unpolar operating.

Hi,

Be sure to get a capacitor that has a ripple current rating higher than you expect to see though that cap. That should do it i think.

 

The only matter that worries me in your circuit @Danko is unpolar operating.

Do not worry:
___
ADDED:
Voltage on load at PWM=5%