Lithium battery for a PHEV?

[alec_t]

Here's your .asc file with a few changes. The 'k' comments are now coil coupling directives. TL431 is a reference voltage generator, here wired to give a 2.5V reference against which a divided-down Vout is compared by U1. R7 is deliberately set low so that the U1 operation can be seen relatively quickly.

 

[mading2018]

Thank you very much, just one question regarding the voltage regulator TL431. I need to add it to my library since I don't have that component.
Do you how to add new components for MAC? It differs a bit from windows considering the folder system.

I found these files:
http://ltwiki.org/index.php5?title=File:TL431_test.asc

and I read this post, but it was for windows.

 

[alec_t]

Sorry, I don't know how MAC does things. For simulation you could just replace the TL431 (precise reference) with a zener diode (less precise).

 

[alec_t]

Here's a way to make your own reference.

 

[mading2018]

But if Vout is compared by U1, how do I making sure that Vout is compared to 300 V? I want to obtain a stable voltage (around 3oo V) at amp_out.
Does it means I need to change the resistors? I tried to do that, but I think it has a little effect.
Cause what I understand, Vee and Vcc is either set to 5-15 V in order to supply the LT6016, and should not be changed to anything else?

Oh, so the file DIY-ref.asc can replacing the the zener diode, to make the reference more precise?

 

[alec_t]

how do I making sure that Vout is compared to 300 V?

You choose a reference voltage Vref and make a potential divider from two resistors so that you can tap off a fraction Vref/300V from Vout.

so the file DIY-ref.asc can replacing the the zener diode, to make the reference more precise?

Yes. Or, simpler still, just use a voltage source as the reference.

 

[mading2018]

Okay, I see. But it seems that whatever I do, I always get the same output voltage from the LT6016. I do not know why, it shows 15 V,
is it due to the supply voltages sources are connected in the wrong way to the LT6016? I dont' think so tough. Can you please analyze my circuit once more?

 

[alec_t]

The LT6016 isn't in that .asc file. I've attached a revised .asc
Is this what you're trying to do?

 

[mading2018]

Well, all I am trying to do is to get a quite stable 300-400 V (DC) output from the converter. So what I understand I need some kind of feedback in order to watch the output is really getting out 300-400 V (DC).

 

[mading2018]

The output voltage should be something similar to this graph (from the research paper), the yellow curve (the PFC voltage).

 

[mading2018]

The LT6016 isn't in that .asc file. I've attached a revised .asc
Is this what you're trying to do?
View attachment 111803

I am not sure if you saw my post, so I post it again. I am trying to do is to get a quite stable 300-400 V (DC) output from the converter. So what I understand I need some kind of feedback in order to watch the output is really getting out 300-400 V (DC).

 

[alec_t]

Yes, you need feedback. The comparator output (post #48) changes when the converter output goes above 300V and you use the comparator output as feedback to modify the switching behaviour of the converter, i.e you reduce the 'on' time of the MOSFETs.

 

[ChrisP58]

I am not sure if you saw my post, so I post it again. I am trying to do is to get a quite stable 300-400 V (DC) output from the converter. So what I understand I need some kind of feedback in order to watch the output is really getting out 300-400 V (DC).

Have you abandoned using a dedicated PFC controller like the LT1249, which includes the necessary feedback function, that you had running in this thread?
https://www.electro-tech-online.com...ontroller-properly-to-ac-dc-converter.153228/

 

[mading2018]

Have you abandoned using a dedicated PFC controller like the LT1249, which includes the necessary feedback function, that you had running in this thread?
https://www.electro-tech-online.com...ontroller-properly-to-ac-dc-converter.153228/

Yes, because I do not know how to implement the LT1249 to a interleaved (2-channel) boost PFC, do you know how to do it? I have looked everywhere and it seems there is no controller for interleaved boost PFC in LTspice.

I am also suppose to simulate a (1-channel) boost PFC, then I can use LT1249, so that should be okay.

 

[mading2018]

Yes, you need feedback. The comparator output (post #48) changes when the converter output goes above 300V and you use the comparator output as feedback to modify the switching behaviour of the converter, i.e you reduce the 'on' time of the MOSFETs.

Aha, I see. So you mean the comparator output should be connected to the MOSFET somehow? I am not sure how I should connect the output to the MOSFET correctly in order to reduce the 'on' time ? Can you please show me how?

PS. Oh, wait, you mean I can maybe just remove the voltage sources and connect the comparator output to the MOSFETs? DS.

 

[alec_t]

Oh, wait, you mean I can maybe just remove the voltage sources and connect the comparator output to the MOSFETs?

No. Although simply switching off the MOSFETs will stop the boost action and hence keep the output voltage in check, it will also prevent the PFC working correctly. I've no experience with PFC, but as I understand it (I could well be wrong) it's a separate issue from closely regulating the output voltage, involves monitoring supply/load currents and voltages, and is usually performed in a pre-regulator stage.

 

[ChrisP58]

PWM, or Pulse Width Modulation, is where the duty cycle of the mosfets is modulated to regulate the output voltage of a switch mode power supply. In simple terms, the on/off time of the signal driving the gates of the mosfets is constantly adjusted in response to real time measurement of the output voltage. The output voltage is measured, then compared to a reference. Then a decision is made to increase or decrease the on/off ratio of the mosfet.

It is similar to how you control the speed of your car. You look at the speedometer, compare it to what speed you want to travel at, then adjust the pressure on the gas pedal accordingly.

For Power Factor Correction, it is much more complex. The modulation scheme is intended to make the input current look like a sinewave. To do that, it has to look at the realtime input voltage waveform, the realtime measured current, as well as the output voltage. Then calculate what on/off ratio to drive the mosfets at for each pulse. Then measure and calculate it again for the next pulse. Again and again.

That's what the LT1249 and other active PFC control chips do. To do it manually, you'll need to emulate all of the internal functions of those control ICs.

 

[mading2018]

For Power Factor Correction, it is much more complex. The modulation scheme is intended to make the input current look like a sinewave. To do that, it has to look at the realtime input voltage waveform, the realtime measured current, as well as the output voltage. Then calculate what on/off ratio to drive the mosfets at for each pulse. Then measure and calculate it again for the next pulse. Again and again.

That's what the LT1249 and other active PFC control chips do. To do it manually, you'll need to emulate all of the internal functions of those control ICs.

I tried again with the LT1249, I think it looks okay. I got some what of constant output voltage with small ripple I think.
I have an 230 V (rms) and 4 A input, but do you think it looks okay do have both a current source and a output voltage?

 

[alec_t]

do you think it looks okay do have both a current source and a output voltage?

Your current source is doing nothing useful. It won't limit the input current, if that's what you're aiming for. The input current is determined mainly by the output load current and the ratio of the output voltage to the input voltage.

 

[ChrisP58]

I've got to agree with alec. The AC grid is a voltage source. Modeling is as a current source in a PFC simulation will give you false results.

The purpose of the PFC control circuit is to force the current to be drawn in phase with the input voltage. But the simulation environment needs to be as close to normal field conditions as possible for the results to be valid.