RM46L430 EMIF Switching Characteristics

Prad,

I believe these characteristics are based on simulation data not test data.

CL means the load capacitance (total).   It is a lumped model and the time for one edge (rise or fall).  So 2 edges would mean 2x this.  

This table is intended (IMO) to be used with slower peripherals like SPI or GIO and with higher loads.  EMIF isn't really  a good fit for using this simple model.  The better way to analyze EMIF is to use the IBIS model for the device.   However, we have not published a model yet for the RM46 part.

Hi Anthony,

Thank you very much.

Just in case our customer want's to know, with what "Load Capacitance" are
the EMIF timings(mentioned on datasheet) are tested or simulated? is it with "Zero" load capacitance?

It would be very helpful if there is any details of the circuit diagram of the TEST CONDITION..

Best Regards
Prad

Hi Prad,

I believe our characterization is done by looking at the knee of the transition - where the pin just starts to show that it is transitioning - so that the result is unaffected by the load capacitance.  So in that sense you can say yes.  It's not the same but the effect is to measure and avoid the effects of capacitance.

Hi Anthony,

Thank you,

Our customer is saying that the EMIF timing Switching Characteristics is very important
for their design, without knowing the correct timing it would be difficult to simulate the
timings of EMIF and its related operations.

We sincerely request you to provide us the IBIS model or simulation details of the EMIF module.
Just in case my mail address is pradeep_kumar@ktl-corp.co.jp

We can understand it would be difficult to provide us the simulation model,
if possible could you please provide us the brief details of the TEST CONDITION(Circuit?) of
EMIF timings on the datasheet.

I am sorry for this kind of request but it seems customer is having difficulties in
designing EMIF timings based on the datasheet.

Best Regards.
Prad

Prad,

Sunil Oak is working on putting together the IBIS model for this chip.   We have models for the IO but you have to combine them w. the model for the package and that's the step that he is working on now.

At the same time, the EMIF is not specified to run very fast on this part so I am actually a bit surprised that the timing is so critical.  Would you please let me know what your customer is trying to connect to?  If it is to SDRAM then I guess my concern is that most SDRAM today is spec'd to operate much much faster than the EMIF on the RM46 so there should be plenty of margin.   If it is some other memory or FPGA then they should know that the timing for the async interface supports programmable wait states for setup, strobe and hold times so it is very flexible.

I'm assigning this to Sunil though for the IBIS portion of the question.

Hi Anthony,

Thank you very much for considering the request.
If possible please let us know the time when would we receive the IBIS model.

Our customer seems to be connecting SDRAM and it seems this timing is crytical for their design.

According to the datasheet 3.9 Output Timings ->Table 3-5 and your first comment
we understood that 2.5ns (@CL=15pF) time is for low EMI pins or slower peripherals,
so we believe, Rise time&Fall times for EMIF would be faster(i.e.<2.5ns).

Meanwhile, I am very sorry to ask the same question again,
regarding the "EMIF timing Switching Characteristics" on the datasheet
as you mentioned above we understood that,the timings are based on simulation.
Can we assume those simulations are performed without any external memory devices(Zero load capacitance)?
(I think we may get the answer in the IBIS model)

We appreciate your kind support.

Best Regards
Prad

Hi Anthony,

Thank you very much.

Meanwhile, please let us know if there is any updates on the IBIS model release.

Best Regards
Prad

Hi Anthony,

I am sorry for bothering you again.

Could you please let us know the release schedule of the IBIS Model.
We shall ask our customer to wait till that time.

Best Regards
Prad

Hi Sunil,

Thank you for the updates.

Regards
Prad

Hi Sunil,

Please let me know when could we download the IBIS model.
As you know it was not updated last week.

Meanwhile, please let us know if there is any procedure
to find out EMIF timings using the IBIS model.

Regards
Prad