LSF0108: LSF0108RKSR series model selection

Hi Dylan,

thanks for your help

there is another question I want to ask

is that universal about"Select the model that corresponds with the planned VrefB voltage, so in your case 3.3 V." in write and read?

Hi Dylan,

Sorry to sking another question 

after simulating the level shift between 1.8V to 3.3V(please see the pictiure below)

Red line is output by using series model: LSF0101_IO_18_S

Green line is otput by using series model: LSF0101_IO_33_S

as the simulation result can see

it's shows using  LSF0101_IO_18_S may can level shift 1.8V to 3.3V

it's different that you told me "Select the model that corresponds with the planned VrefB voltage, so in your case 3.3 V."

is there have any comment or reson you can tell me why?

thanks

Tommy 

Hey Tommy,

Can you share your circuit with me?

Hi Dylan,

sure,please the picture below

the simulation point is U8

the different is ones is use series model: LSF0101_IO_33_S and the other is series model: LSF0101_IO_18_S

the result is the picture i send it yesterday.

thanks

Tommy

Hi Dylan,

sorry to disturb you again

I'm afraid of some misunderstanding between us

I want to share my Hyperlynx file to you

do you guys have  Hyperlynx lincense?

please help to setup LSF0108 properly for us

lsf0108_DEFULT.ibs

thanks

Tommy

Hi Dylan,

and this is the circuit I set 

add another point : when you use this file , please delete the file name .IBS first

lsf0108_set up.7z.IBS

THANKS 

Tommy

Hey Tommy,

Thanks for added info. I'll pass this to our modeling expert to dig into, please give them 24 hours to ramp up on the issue and provide feedback.

Hi Tommy,

I am looking into this to see what is the issue. I am waiting on the hyperlynx license as I cannot open the file otherwise.

Are you able to use TINA-TI or PSPICE to simulate? We have working models in that case.

2514.LSF0108_Example.TSC

1581.LSF0108_Behavioral.cir

LSF0108_Behavioral.lib

Regards,

Karan

Hi Karan

No, we don't have PSPICE to simulate it, but we do download the TINA-TI.

From 1581.LSF0108_Behavioral.cir, we can know what's the corresponding volt for each pin/model.

Looks to me like the series model connects to EN pin, and EN is the same as VREFB.

So, I think we should use 3.3V series model in simulation when we do level shift from1.8V to 3.3V.

But, the simulation results are not what we think.

BTW, are LSF0108 using the same series model( pass FET architecture) as PCA9306?

Hi Klaus,

Thats correct, you should be using the 3.3V series model but something in the model is not working.

Yes the data may be different but the PCA9306 has a similar model architecture. Are you able to use the PCA9306 IBIS model?

Regards,

Karan

Hi Karan

Yes, from your schematic, we should use 3.3V, since EN connects to VEFB, but just like what you said, something in the model is not working.

We've compared LSF0108 and PCA9306 series model looks like both behaviors are the same, so I don't think PCA9306 will work.

Klaus

Hi Karan

For IBIS file, do you have any suggestions? or can we use another one?

Klaus

Hi Klaus,

Unfortunately I am not able to figure out why the IBIS model is not working. It looks like the data is there but it is giving opposite values.

I will suggest you try to use the TINA model as it is fairly accurate.

Regards,

Karan

Hi Karan

I think we still need to figure out why the series model is not working

Since we've verified HyperLynx tool and we also used many IBIS files in HyperLynx, such as i2c switch, level shifting, etc.

For those components I mentioned above, it always has the series model in the IBIS file.

Could you get the HyperLynx license? so that we can compare apple-to-apple.

Thanks

Klaus

Klaus,

Yes, let me try to get the license.

Regards,

Karan

Hi  Karan

Regarding the fast-strong model, does it mean the strongly driven strength of the driver? like CPU.

Both read and write, is the same? I mean using the 1.8V series model for read and write.

Thanks

Klaus

Hey Klaus,

Today is a U.S. Holiday, please give Karan until Monday to reply. Until then, I recommend watching these training videos on LSF to understand how it functions. This may better help you to understand some of the things discussed.

Hi Dyan

Sure, I watched it before, but it's incorrect when using 3.3V series model in the simulation, so that's why we would like to confirm with Karan.

Looks like we could use 1.8V series model in the series model, but we're not sure is it ok for both read and write.

Thanks

Klaus