|
Hello Everyone,
I’m currently working on simulating a half-bridge circuit using GaN devices from GaN Systems in LTSpice. To generate gate pulses, I’m employing a voltage PWL source to create a double-pulse pattern with varying time durations.
At present, I’m using a single resistance for both the turn-on and turn-off phases of the gate drive, but I would like to model separate resistances for each. I understand that this can be accomplished using voltage-controlled switches to selectively route the gate signal through different resistors during the turn-on and turn-off phases.
However, I’m having trouble implementing this in LTSpice and would appreciate any guidance or example models that demonstrate how to achieve this configuration. Has anyone successfully set up a similar circuit and can share their approach?
Thank you in advance for your help!
|
On Tue, Nov 12, 2024 at 05:11 PM, <asadullahkk514@...> wrote:
However, I’m having trouble implementing this in LTSpice and would appreciate any guidance or example models that demonstrate how to achieve this configuration.
I think it should be fairly straight-forward. Can you show us what you tried, which apparently did not work?
Make sure that the drive signal controlling those voltage-controlled switches, swings to both sides of the switching threshold voltage, the parameter Vt. Sometimes people set Vt=0 and then drive the control input with a waveform that swings between 0 and some X volts. That is not right.
Andy
|
Please upload your schematic with the symbols and device models that you’re using so we can see what you’re trying to do.
A simple way to implement a switched resistance is with the R=(formula) syntax.
Instead of typing a number for the resistance, enter a formula.
For example, enter the following for the resistor’s value. Be sure to include the “r = “ part:
r = if(time<1u, 1k, 2k)
That implements a single resistor whose value changes at 1 microsecond. You can use nested if() statements to get more than 2 discrete values.
Another thing you can try is setting the resistance equal to a node voltage r=v(x). Then create a voltage waveform at node x with the shape (in volts) that you want your resistance to have (in ohms).
Remember that resistors can’t be zero ohms. (Make them 1u ohm to simulate a hard short.)
|
Thank you, Andy and Chris, for your response. Attached is my schematic. I am designing a gate driver circuit with a turn-on voltage of 6V and a turn-off voltage of -3V, using separate gate resistances for each state. Before connecting the voltage-controlled switches, I am able to observe a constant voltage. However, once connected, I am only seeing a square wave that ranges from 6V to 1.3V. My goal is to achieve a square wave signal transitioning from 6V to -3V. This circuit was developed with the assistance of the provided LTspice schematic from GaN Systems. While their reference design utilizes a single pulse, my test setup requires two pulses of different durations. Could you please review this and provide any insights or recommendations for achieving the correct voltage levels and pulse configuration?
find the document by following the link below https://drive.google.com/drive/folders/1VcXDgfPA_qCaWEb_K-kumVl6MoMtQjvg?usp=drive_link
|
No, the onus is on you to upload your schematic to the group's Files > Temp
location, as requested in the instructions on the Homepage.
Please don't unnecessarily direct us to 3rd party sites.
--
Regards,
Tony
On 13/11/2024 10:17, ASADULLAH KHAN via
groups.io wrote:
Attached is my schematic. I am designing a gate driver circuit
with a turn-on voltage of 6V and a turn-off voltage of -3V,
using separate gate resistances for each state. Before
connecting the voltage-controlled switches, I am able to observe
a constant voltage. However, once connected, I am only seeing a
square wave that ranges from 6V to 1.3V. My goal is to achieve a
square wave signal transitioning from 6V to -3V.
This circuit was developed with the assistance of the provided
LTspice schematic from GaN Systems. While their reference design
utilizes a single pulse, my test setup requires two pulses of
different durations.
Could you please review this and provide any insights or
recommendations for achieving the correct voltage levels and
pulse configuration?
|
I have uploaded LTspice schematics and pictures in the group with a description: Modeling Gate Driver Circuit for GaN Half-Bridge in LTSpice with Separate Turn-On/Off Resistances.
|
On Wed, Nov 13, 2024 at 04:17 AM, ASADULLAH KHAN wrote:
find the document by following the link below
https://drive.google.com/link_to_offsite_storage
No!
Don't ever do that. As it says on the group's main webpage:
Please do not point us to other ("third party") file storage websites, even if it is one where you have an account. Instead, upload your files to this group's website, and only into the Temp directory.
Also, uploads must be complete including all symbols (.ASY files) and SPICE models that did not come with LTspice itself. LTspice symbols do not "live" on the schematic; they are in separate symbol files. By not including tho symbol files for new devices that you either downloaded or created, your schematics become unusable by everyone else.
Please go back and read the group's instructions on its main webpage, then try again, and then we can help you better.
Andy
|
On Wed, Nov 13, 2024 at 05:08 AM, ASADULLAH KHAN wrote:
I have uploaded LTspice schematics and pictures in the group with a description: Modeling Gate Driver Circuit for GaN Half-Bridge in LTSpice with Separate Turn-On/Off Resistances.
Thank you - I see that you did now upload the schematics. However....
They are missing the symbols and SPICE models for your parts:
GaN_LTspice_GS665088T_L3V4P1.asy - file is missing!
SPICE model for GaN_LTspice_GS665088T_L3V4P1 - file is missing!
Please upload those.
Andy
|
ASADULLAH KHAN,
I'm going to ignore the GaN device on your schematics, because you forgot to upload it. Looking at your voltage-controlled switches:
On the schematic GaN_Gate_driver.asc, you drove S2 with a control signal that toggles between 0 V and -1 V. But the switch model has Von=1, so it will not turn ON until the control voltage reaches +1 V, which it never does. Switch S1 S2 never toggles on. That is the main problem. You need to fix that. I guess S2's model should use Voff=-1 and Von=0.
Did you want the high-drive resistance to be 12 ohms, and the low-drive resistance to be 7 ohms? That is what you have made them. Perhaps the switch parameter Ron should not be 2, but that is for you to decide. In fact you did not need to add external resistors R_on (10 ohms) and R_off (5 ohms) because those resistances could have been incorporated in the two switch models. But it might be easier to interpret the schematic if the resistors are external and the switch models have much smaller internal ON-resistances.
I glanced at the schematic GN008_GANSYS_Half_Bridge_Switching_Test_v2.asc, but it looks like it might have the same problem(s).
Andy
|
two options, diode and 2nd resistor, for fast off, or totem pole or transistor to ground with separate off gate, or center node of two in series of on path. If your drive is strong enough then only need to improve off conductance. asymmetric resistors don't solve everything, but do help.
I have uploaded LTspice schematics and pictures in the group with a description: Modeling Gate Driver Circuit for GaN Half-Bridge in LTSpice with Separate Turn-On/Off Resistances.
-- Ron Davison principal Efficient Electronic Solutions 760.505.4380 "What I cannot optimize, have not been given the opportunity, yet." Ron Davison “What I cannot build, I do not understand” Richard Feynman
|
Andy,
Thank you so much for your brief response. I resolved my problem, and thank you for the suggestions for R_on and R_off. Regards Asadullah Khan, Kaim Khani.
ASADULLAH KHAN,
I'm going to ignore the GaN device on your schematics, because you forgot to upload it. Looking at your voltage-controlled switches:
On the schematic GaN_Gate_driver.asc, you drove S2 with a control signal that toggles between 0 V and -1 V. But the switch model has Von=1, so it will not turn ON until the control voltage reaches +1 V, which it never does. Switch S1 never toggles on. That is the main problem. You need to fix that. I guess S2's model should use Voff=-1 and Von=0.
Did you want the high-drive resistance to be 12 ohms, and the low-drive resistance to be 7 ohms? That is what you have made them. Perhaps the switch parameter Ron should not be 2, but that is for you to decide. In fact you did not need to add external resistors R_on (10 ohms) and R_off (5 ohms) because those resistances could have been incorporated in the two switch models. But it might be easier to interpret the schematic if the resistors are external and the switch models have much smaller internal ON-resistances.
I glanced at the schematic GN008_GANSYS_Half_Bridge_Switching_Test_v2.asc, but it looks like it might have the same problem(s).
Andy
|
Correction of my typo:
I wrote:
... Switch S1 never toggles on. That is the main problem. ...
That should say: "Switch S2 never toggles on."
Andy
|
|
|
Tony Casey
