LM5114: Simulation in LTspice

Yoram,

Thanks for your interest in our driver.

The model does not support "floating" condition as the VSS will only recognize the GND reference node of your circuit.

I suggest you use one of our half-bridge drivers from the link below.

https://www.ti.com/power-management/gate-drivers/half-bridge-drivers/products.html#p3162max=20;120

You may try UCC27282 as a starting point.

Regards,

-Mamadou

Mamadou,

thanks but

this is not resolve my problem.

I need the LM5114 for is specific characteristics because using high speed high voltage (> 200V) Gan devices

Hi Yoram, 

I am helping over the weekend.

LM5114 is a low side gate driver, see image below. not half bridge driver. Could you please help advise what you are looking for?

is it a question about selecting a fast driver to drive the low side GaN like LM5114 or LMG1020. or a half bridge driver to drive a half bridge. For example: LMG1210, 200V half bridge driver.

or you have a technical question where the simulation doesn't work? for this case, can you please help send us a block level schematic diagram so that we can know what you are struggling with.

Thanks.

Many thanks for your help.

My system: ~1nF gate load, high current gate pulse  and ~600V full bridge

Do you have any suggestions of  device that cad run spice model in "float" mode? (not obligatory GND = VSS)

Otr if you please can TI provide new model for 5114 that can do float simulations?

per you suggestion:

lmg1020 does not indicate to drive  ~1nF load.

Regards to all

Yoram

Yoram,

The drivers from the link below should fit your application needs.

https://www.ti.com/power-management/gate-drivers/half-bridge-drivers/products.html#p3162max=121;700

Regards,

-Mamadou

Hi Mamadou,

I'm truly sorry but this is unacceptable conduct. avoiding direct answer.

We here at CMT Medical (We are part of Thales France Corporate) invested a lot in our current PCB product

according to the notion that TI is a reliable company.

We are in a problem that the Spice not working without any disclaimer that this GND=VSS issue in you web site.

Your answer is not satisfying according to any respectable conduct of a respectable corporate.

I ask you again to answer if you can provide us with new simulation model. If you can't , I would ask you to refer us to a higher TI entity

that can there kind of decisions.

Hi Yoram,

I am sorry that you found my answer unacceptable and I appreciate your sincere feedback.

From your application description, it looks like you're trying to simulate a full topology using LM5114 (which is a low-side driver suited for different applications, boost/interleaved PFC, SR, etc...) which is why we suggest our half-bridge families on the link above where you have the option of using LMG1210, UCC27710, UCC27712 and UCC27714 based on drive strength, power transistor, gate drive voltage, UVLO, and gate turn-on/off requirements.

These drivers will be suited for your 600V bridge topology as they all include high-side and low-side channel to drive heavy loads and high gate charge FETs. 

The drivers' models are available on the respective product folder pages shown below:

-UCC27710.

-UCC27712.

-UCC27714.

-LMG1210.

Also, if possible, kindly share your block diagram and/or schematic so that we have an understand of which IC is driving the LM5114 in your current setup in order to provide you with a workaround solution and/or alternative.

I will look forward to hearing from you.

Regards,

Hello Mamadou,

UCC27710, UCC27712 - great products. DT interlock too large . can't use it
UCC27714 MinimumHI/LION pulse too big
LMG1210 snom677.zip LMG1210 Unencrypted PSPICE Transient Model does not unzip
        in the far past we did try to get spice model but refused.
what next?

what is it will take to get a new model?

Yoram

Hi Yoram,

I have attached the LMG1210 unencrypted PSpice model below.

lmg1210.zip

Additionally, what are your application's targeted DT and min pulse width?

Regards,

-Mamadou

Hi Mamadou,

Thank you for the model. It means new PCB and we haven't given up yet on obtaining the "float:" model of LM5114

FYI: pulse width (freq > 1Mhz) and DT 60-80 nSec

Our PCB works but at some PWM conditions we encounter some problems due to our own specific architecture.

So we want to resolve those issues with more tight simulations, that now apparently is impossible due to the VSS=GND issue.

Regards

Yoram

Hi Yoram,

Given your switching frequency and depending on your choice of FETs and gate drive voltage range, you might also run into a thermal issue on the driver side. The switching frequency of the driver IC as turning on/off the gate of the transistors (assuming you're driving power FETs) requires the driver to dissipate power internally from P = Qg * VDD * Fsw which will induce junction temperature rise that will need to be limited to 125C.  

For those reasons, I suggest you consider LMG1205 (PSpice model attached below) and/or LMG1210 from the previous post as they're suited for >1MHz switching frequency. 

LMG1205 Unencrypted PSpice Model.zip

Regards,

-Mamadou

Great services for <200v

not suite for me

Yoram, 

I am looking into other possible workaround solutions, will get back to you before end of the week.

Regards,

-Mamadou

Sorry to use strong langue but:

Without simulation model, in case we do change our PCB it will not contain any of TI drivers.

Also I can't sign-off this E2E session as resolved.

Please see this issue hang as is, until you do send a model.

New further communication about this issue if futile

Yoram, my suggestion is to use the BJT buffer for your simulation since you're simply using the driver IC for the drive current capability of the IC.

As previously mentioned, we're going to use this information to evaluate the feasibility of our models as "floatable" for our latest and future IC's.

Regards,

-Mamadou