RM46L852: Maximum Bit Rate value allowed for the SPICLK

Francisco Vera

Part Number: RM46L852
Other Parts Discussed in Thread: HALCOGEN

Hi,

I have a question about the maximum bit rate value allowed for the SPI communication on the RM46L852 microcontroller. I want to use the MT29F4G01ABAFDWB_IT_F flash memory, this memory can work at a maximum bit rate value of 133 Mhz, but I would like to work to 110 Mhz or 55 MHz. However, on page 160 of the RM46L852 datasheet (https://www.ti.com/lit/ds/symlink/rm46l852.pdf?ts=1712518075244&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FRM46L852) appears that the minimum value of SPICLK cycle time is 40 ns -> 25 MHz.

However, the Halcogen tool allows getting a higher value of SPICLK frequency. Regarding this application, I could reach a value of 55 MHz.

This situation contradicts the information that appears in the datasheet. Therefore, what information would be correct?

Best regards,

Francisco.

9 months ago

0 Ralph Jacobi 9 months ago

TI__Guru*** 133515 points

Hello Francisco,

Our expert for this device is on holiday leave, so please expect a small delay in response from them.

Best Regards,

Ralph Jacobi

0 Francisco Vera 9 months ago in reply to Ralph Jacobi

Prodigy 120 points

Hi Ralph Jacobi,

Thanks for your reply. Ok, I look forward to receiving his response.

Best regards,

Francisco.

+1 jagadish gundavarapu 9 months ago in reply to Francisco Vera

TI__Guru 58895 points

Hello Francisco,

Even though HALCoGen allows you to configure more than 25Mhz, but the actual supported maximum frequency for SPI in Hercules devices is 25MHz only.

Datasheet values were correct.

You can also confirm the same from below old threads:

(+) RM46L852: Micro SDXC High Endurance for the RM46L852 - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

(+) TMS570LC4357: SPI Clock Questions - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

(+) TMS570LC4357: SPI Clocks - Arm-based microcontrollers forum - Arm-based microcontrollers - TI E2E support forums

--
Thanks & Regards,
Jagadish.

0 Francisco Vera 9 months ago in reply to jagadish gundavarapu

Prodigy 120 points

Hi Jagadish,

Could we say that Halcogen has a bug?

Best regards,

Francisco.

0 jagadish gundavarapu 9 months ago in reply to Francisco Vera

TI__Guru 58895 points

Francisco Vera said:
Could we say that Halcogen has a bug?

We can't say that because, HALCoGen is generic for all the Hercules devices and it won't follow any particular datasheet limits,

0 Francisco Vera 9 months ago in reply to jagadish gundavarapu

Prodigy 120 points

Hi Jagadish,

It is true that HALCoGen works for different devices, but I consider that it is a bug because the user must first choose the device before it can be configured.

Therefore, HALCoGEN has a bug for the MibSPI/SPI controller of the RM46L852PGE device. This is not an opinion but a fact. On the other hand, HALCoGEN also does not offer the possibility to configure the SPI Extended Prescale Register 1/2. So, Texas Instrument should fix these bugs.

An interesting situation is that if I configure the MibSPI controller (through HALCoGEN) to work at 55 MHz, it seems to work fine. Therefore, is there really a hardware limitation or am I causing that if I continue working with this configuration it may stop working correctly in the future?

Best regards,

Francisco.

+1 jagadish gundavarapu 9 months ago in reply to Francisco Vera

TI__Guru 58895 points

Hi Francisco,

Francisco Vera said:
It is true that HALCoGen works for different devices, but I consider that it is a bug because the user must first choose the device before it can be configured.

Yes, we can choose the device before configurations. But as far as my understand and usage HALCoGen doesn't provide limitations for configuration values.

And i agreed with you, it is a valid requirement. I will log this, and we will consider this for future HALCoGen release.

Francisco Vera said:
An interesting situation is that if I configure the MibSPI controller (through HALCoGEN) to work at 55 MHz, it seems to work fine. Therefore, is there really a hardware limitation or am I causing that if I continue working with this configuration it may stop working correctly in the future?

It might work but we are not testing with extreme conditions, right?

For example, it might work at room temperature for 55Mhz bug we can't expect to be work at extreme temperatures like -40/105 degrees. But the 25Mhz given datasheet is a value tested at all the extreme conditions.

So, anything operating more than 25Mhz cannot be guaranteed from TI side and it is user risk.

--
Thanks & regards,
Jagadish.

0 Francisco Vera 9 months ago in reply to jagadish gundavarapu

Prodigy 120 points

Hi Jagadish,

Thanks for your reply. Thanks for logging these bugs.

Yes, I have not tested these features in extreme conditions because I am not going to work in these conditions and I have not the resources or means to be able to do it. Therefore, I will analyze in detail if it is necessary for me to work for the 55 MHz frequency and I will consider working for the 25 MHz frequency.

Best regards,

Francisco.