Error in boost circuit simulation in Ltspice

Thread Starter

gokulpradeep010

Joined Jun 2, 2021
42
Hi all,
I tried to building a boost convertor with input 24V, and 180V .1A output. But when I simulate I am getting some strange waveform. Please help me to fix it. Hopefully the calculation part is right, as I have taken the circuit directly from TI website.
 

Attachments

Papabravo

Joined Feb 24, 2006
21,094
What is strange about them?
Can you give us the URL?

EDIT: There are a coupe of problems:
  1. You did not provide a symbol for the part, and the one that can be autogenerated from the subcircuit file does not match your schematic.
  2. There is no simulation command on your schematic so I seriously doubt that you have run the simulation or produced any waveforms that can be seen.
  3. Why would you include that huge text file on the schematic? It makes the schematic impossible to navigate.
 
Last edited:

Thread Starter

gokulpradeep010

Joined Jun 2, 2021
42
What is strange about them?
Can you give us the URL?

EDIT: There are a coupe of problems:
  1. You did not provide a symbol for the part, and the one that can be autogenerated from the subcircuit file does not match your schematic.
  2. There is no simulation command on your schematic so I seriously doubt that you have run the simulation or produced any waveforms that can be seen.
  3. Why would you include that huge text file on the schematic? It makes the schematic impossible to navigate.
https://webench.ti.com/power-designer/switching-regulator/simulate/5
 

Papabravo

Joined Feb 24, 2006
21,094
I'm mystified because I cannot find the .tran command on the schematic, but I can see that you set the value to 10 milliseconds. I agree with @crutschow , that 8 or 10 msec. is not long enough to observe any meaningful behavior. How have you defined the power on behavior of the input voltage source?
 

Papabravo

Joined Feb 24, 2006
21,094
That doesn't help -- it wants me to login and I don't have the credential to do that. You're not providing the meaningful information we need to help you.
Did you copy a reference design, or did you make that schematic up all on your own. It sure doesn't look like what I think a boost converter should look like. In particular the input voltage source is not connected to the inductor, but since I don't have the correct symbol for the part it is hard to tell what is going on. Send me the !@#$%^&*() symbol - please.
 

Papabravo

Joined Feb 24, 2006
21,094
Here is another wrinkle. The autogenerated symbol has 11 pins as does the subcircuit, but the IC only has 10 pins. This is a regular WTF moment.
EDIT: OK I understand now SS is not a pin, it is a parameter.
 

Thread Starter

gokulpradeep010

Joined Jun 2, 2021
42
That doesn't help -- it wants me to login and I don't have the credential to do that. You're not providing the meaningful information we need to help you.
Did you copy a reference design, or did you make that schematic up all on your own. It sure doesn't look like what I think a boost converter should look like. In particular the input voltage source is not connected to the inductor, but since I don't have the correct symbol for the part it is hard to tell what is going on. Send me the !@#$%^&*() symbol - please.
 

Attachments

Papabravo

Joined Feb 24, 2006
21,094
Thanks

First problem is that you have connected the voltage source V1 to the COMP(compensation ) pin and what looks like the compensation network to the VDD pin.

I would go back over your design with a more careful fine tooth comb to eliminate such obvious mistakes. Come back after reviewing the connections and revising the design. It cannot work the way you have drawn it.

EDIT:
As I have already mentioned, SS is a parameter to the subcircuit, NOT a pin. The device per the datasheet has only 10 (count 'em) pins - not 11.

You must have read the datasheet too quickly to have noticed this minor detail.

EDIT2: You said in your original post that you took the design from the TI website. Can you show us the original drawing so we can compare it to the schematic that you have prepared. A screenshot will be fine it that is all you have.
 
Last edited:

Thread Starter

gokulpradeep010

Joined Jun 2, 2021
42
How can it be the same, as the voltage is rising in the simulation results you show?
Post the "result".
Thanks

First problem is that you have connected the voltage source V1 to the COMP(compensation ) pin and what looks like the compensation network to the VDD pin.

I would go back over your design with a more careful fine tooth comb to eliminate such obvious mistakes. Come back after reviewing the connections and revising the design. It cannot work the way you have drawn it.

EDIT:
As I have already mentioned, SS is a parameter to the subcircuit, NOT a pin. The device per the datasheet has only 10 (count 'em) pins - not 11.

You must have read the datasheet too quickly to have noticed this minor detail.

EDIT2: You said in your original post that you took the design from the TI website. Can you show us the original drawing so we can compare it to the schematic that you have prepared. A screenshot will be fine it that is all you have.
SS is a pin, whereas EP is not a pin,
This is the orginal design, which I got from TI
 

Attachments

Papabravo

Joined Feb 24, 2006
21,094
Last problem: Like may who are new to simulation you have hard coded a file location into your symbol. This is a bad bad thing. It means the design and the symbol are not portable. LTspice is perfectly capable of finding the subcircuit file in one or more standard and convenient locations with just the file name. On top of that there is absolutely no need to have the subcircuit file in the schematic. It serves no useful purpose there.
 

Papabravo

Joined Feb 24, 2006
21,094
SS is a pin, whereas EP is not a pin,
This is the orginal design, which I got from TI
OK, then the subcircuit is misleading because it clearly is treated as a parameter. So if EP is not a pin, what is it doing in the subcircuit netlist?

There is still a disconnect because the schematic from TI has 11 pins on the part symbol.

EDIT: the problem appears to be with the package which has an exposed thermal pad underneath the package which according to the schematic is also connected to GND. The designation EP could be a shorthand for "Exposed Pad".

So I think I can fix the symbol so it matches the symbol in the TI schematic. That should be helpful in comparing the TI schematic to your schematic.
 
Last edited:

Thread Starter

gokulpradeep010

Joined Jun 2, 2021
42
OK, then the subcircuit is misleading because it clearly is treated as a parameter. So if EP is not a pin, what is it doing in the subcircuit netlist?

There is still a disconnect because the schematic from TI has 11 pins on the part symbol.
Yes, that's right but still it does not have much effect on the circuit I think. As you said I have corrected the design
 

Attachments

Papabravo

Joined Feb 24, 2006
21,094
OK, then the subcircuit is misleading because it clearly is treated as a parameter. So if EP is not a pin, what is it doing in the subcircuit netlist?

There is still a disconnect because the schematic from TI has 11 pins on the part symbol.
Yes, that's right but still it does not have much effect on the circuit I think. As you said I have corrected the design
OK here is my best attempt. It runs and produces an output, but I don't know if it is the one you want. Put all three files in the same fresh sub directory and see if you agree. I don't have time to double check the schematic right now, but I will get back to it later, unless you tell me it is not necessary
 

Attachments

Thread Starter

gokulpradeep010

Joined Jun 2, 2021
42
OK here is my best attempt. It runs and produces an output, but I don't know if it is the one you want. Put all three files in the same fresh sub directory and see if you agree. I don't have time to double check the schematic right now, but I will get back to it later, unless you tell me it is not necessary
What was the output it showed?
Was it 19.5V?
 
Top