LTspice AC/DC convertor

[mading2018]

Your full wave bridge is not working. Remove the ground with the red X.
I do not know why R1.
View attachment 112031
It is not working right. It looks like the frequency response of the error amp is not right. C2,4, R3.
I don't have time to fix it today.

For the C2, C4 and R3 are standard values from a demo for the LT1249, so I haven't change them yet. Yes, it would be great if you have some time to fix it as soonest you have time. Thank you very much, appreciate all your help.

 

[mading2018]

I updated the circuit a bit, I added the feedback that I was recommended to do earlier in order to get a more constant output voltage.

 

[alec_t]

Feedback is provided by R8, R7.
What is the purpose of R1, R9, R10, V5 and U3?

 

[mading2018]

Feedback is provided by R8, R7.
What is the purpose of R1, R9, R10, V5 and U3?

Aha, I see. I have made a doubled feedback, in that case I remove R1, R9, R10, V5 and U3.
But do you think I can improve it in someway? Like to reduce the ripple and to make it more stable for a DC output?

 

[ronsimpson]

I made some changes. PFC circuits start up slowly. It takes about 250mS for it to get up and running stably.

 

[mading2018]

Hmm, I see, it looks good. But is it strange the values that I calculated is not so similar to the simulated values?
I think I want to have the simulation values close as possible to the calculated ones, if possible.

Ps. Is it possible to change the switching frequency internally of the LT1249? In the data sheet it says it is only giving 100 kHz, I would like to double it to 200 kHz, if possible. Ds.

 

[alec_t]

But is it strange the values that I calculated is not so similar to the simulated values?

Tell us how you calculated the values.
You can't change the LT1249 switching frequency.
Increasing C1 will reduce ripple.
Note that values of certain components connected to the IC, which you have copied from the example ciruit in the datasheet, were chosen for a supply input frequency of 600Hz, not 50Hz, so may need to be changed to suit a 50Hz input frequency.

 

[mading2018]

I followed a standard procedure for the calculated components for AC/DC converter with boost PFC, according to this source.
where I obtain the values: Cin=48 nF Lboost= 263 uH, Cout=187 uF. But this assumed to be for 200 kHz. However, I am not sure if the LT1249 can handel that frequency.

 

[mading2018]

Tell us how you calculated the values.
You can't change the LT1249 switching frequency.
Increasing C1 will reduce ripple.
Note that values of certain components connected to the IC, which you have copied from the example ciruit in the datasheet, were chosen for a supply input frequency of 600Hz, not 50Hz, so may need to be changed to suit a 50Hz input frequency.

Hmm, but is it possible only to use voltage source as PWM to the switch, so I can provide 200 kHz,
but in that case I need that feedback again...I guess.. or do you know If can implement 200 kHz externally to the PFC circuit?

 

[alec_t]

do you know If can implement 200 kHz externally to the PFC circuit?

If you do that, the LT1249 would be doing nothing useful and you would have to replicate its internal workings to generate the correct PWM modulation for the 200kHz.

 

[ronsimpson]

But this assumed to be for 200 kHz. However, I am not sure if the LT1249 can handel that frequency.

The frequency is set inside the IC. You can not change it. Why do you have to have 200khz. Try a different part.
Part of the problem I saw was you had a very wrong current sense resistor.

 

[mading2018]

I thought if I had a high switching frequency, I could shrink the size and value of the components, which is mainly my task in this.
The components should be so small as possible for this charger. But maybe, there are other way to do shrink them.
Otherwise, I can go back to use 100 kHz, which is the maximum switching frequency for the LT1249.

 

[mading2018]

I considering to use the LT8132 instead, cause it can handle higher switching frequency up to 400 kHz, and the components that I calculated for is for 200 kHz, so it should be okay. The output looks okay I think, but I used demo version. I added an EMI filter, but the simulation is running not smooth with it. Maybe the values in the filter is wrong, or the design of the EMI filter is incorrectly. I attached the file here, so you can see.

According to the formula I can calculate the V_out, but I am not sure what V_FB is? How do I determine V_FB?

 

[alec_t]

From the datasheet table, the FB voltage is 1.25V, with a tolerance of ±0.3V.

 

[ronsimpson]

I have said this before but I am not clear. I will try again.
Circuit top left has good Power Factor. Power line and resistor.
Circuit top right has good Power Factor. Power line, 4-diodes, load.
Next circuit with a capacitor has bad PF!!! You can not have a large capacitor after the diodes!
Next circuit has the filter on the power line. Before the diodes. (not cap after the diodes)
Last circuit is yours. Capacitors after the diodes. AND You made a LC resonate circuit.

The voltage right after the diodes must be like the top right picture.
Please read and tell me why PF is important.
Please tell me why there can not be a capacitor after the diodes.

(load can be a PFC IC or a resistor)

 

[mading2018]

Sorry I made a misstake about the EMI filter, I know it should be before the diode bridge, and I got the same waveform as you mention above. See image.
PF is important since the current and output voltage should be in phase with each other, and to utilize the whole energy avoid too much losses.
But what I understand it should be a capacitor after the diodes in order to smoothing out the waveforms?

However, I have som problem when I simulating the PFC Lt8312 circuit. How do I avoid "time steep to small" error?
This is the first time it occurred for me.

PS. I noticed that I can not simulate when I include the EMI filter, why is that? DS.

 

[mading2018]

From the datasheet table, the FB voltage is 1.25V, with a tolerance of ±0.3V.

Thank you for the information alec_t. If you have some time, could you please check my circuit in#56? I do not know how to remove
the "time steep to small" error from the simulation.

 

[eTech]

Thank you for the information alec_t. If you have some time, could you please check my circuit in#56? I do not know how to remove
the "time steep to small" error from the simulation.

Try setting initial conditions for the EMI filter inductors by placing a directive on the schematic:

.IC I(L3)=0 I(L4)=0

add 100m serial resistance to voltage source V1

and add some serial resistance to the EMI capacitors C11 and C12.

eT

 

[mading2018]

Try setting initial conditions for the EMI filter inductors by placing a directive on the schematic:

.IC I(L3)=0 I(L4)=0

add 100m serial resistance to voltage source V1

and add some serial resistance to the EMI capacitors C11 and C12.

eT

Ah, thank you. It seems to work now. But, I got overshoot in the beginning of the simulation, is it because to reduce that overshoot? May it be because I have design the resistors R6 and R5 to be 400 voltage out?

Do you know how to set the current input to 4 A? I have tried to ask it before here on this forum, and I got the recommendation to set the
current load to a suitable value so I obtain 4 A at the input. That's what I understand.
For an example: R= 325.79 V /4 A = 81.4475 ohm. Is that the right way to do?

 

[alec_t]

325V @ 4A = 1300W.
1300W / 400V = 3.25 A.
400V / 3.25A = 123 Ω.