MCF8316C-Q1: Migration from MCF8316A to MCF8316C

Hi David,

David Gustavson1 said:

What is the default i2c address of the MCF8316C?

It must be 0x00 according to the datasheet.

David Gustavson1 said:

Can I use the same .json file to program the MCF8316C?

You will have to reprogram the values David, there are minor changes. Alternatively you can just import the MCF8316A .json file and change any settings that may have changed from A to C due to register map changes. 

David Gustavson1 said:

Are the internal pullups for the FG and nFAULT on by default?  If not, how do I avoid the i2c communication problems that plagued the MCF8316A if I don't have pullups on the PCB?

There are no internal pull ups, but most of the Erratas in A were fixed in C. Could you give me more details on the problem you faced in A?

Regards,

Sachin S

Hi Sachin,

Thanks for responding.

How well do you know this chip and its history?  No internal pull ups?

The "A" version would give random connectivity errors:

I was told that without the pull-ups on FG and nFAULT, the device could enter some sort of test mode that would cause this problem.  Is that still the case?  It is very important that I get accurate information on this point.

Regarding the register compatibility, are you saying in need to compare the 2 datasheets register by register to look for differences?  That is very hostile.  Strike one, don't make the registers backward compatible, strike two, don't provide a cheat sheet of how to migrate from one to the other.

I2C address:

The "A" version datasheet showed the Target ID:

Target ID and R/W Bit: The first byte includes the 7-bit I2C target ID (0x01 by default, but can be modified
by setting I2C_TARGET_ADDR), followed by the read/write command bit. Every packet in MCF8316A the
communication protocol starts with writing a 24-bit control word and hence the R/W bit is always 0.

The "C" version datasheet deliberately removed that:

Target ID and R/W Bit: The first byte includes the 7-bit I2C target ID, followed by the read/write command bit.
Every packet in MCF8316C-Q1 the communication protocol starts with writing a 24-bit control word and hence
the R/W bit is always 0.

Where are you seeing that the Target ID is now 0x00?

Thanks,

Dave Gustavson

Hi Dave,

Apologies, I was mistaken. I have not used the internal pull up myself on the MCF8316C, so I did not double check to confirm if the internal pull ups existed. My bad, sorry for the misinformation.

David Gustavson1 said:

Regarding the register compatibility, are you saying in need to compare the 2 datasheets register by register to look for differences?  That is very hostile.  Strike one, don't make the registers backward compatible, strike two, don't provide a cheat sheet of how to migrate from one to the other.

We are in the process of building such cheat sheets, currently have a reg-map conversion tool from MCF8316C to MCF8316D, we will look to add a similar one for MCF8316A to C too.

David Gustavson1 said:

I was told that without the pull-ups on FG and nFAULT, the device could enter some sort of test mode that would cause this problem.  Is that still the case? 

I will check on this with the team and get back to you. I was not aware of this issue.

David Gustavson1 said:

Where are you seeing that the Target ID is now 0x00?

Again I will double confirm the values.

We are phasing out of the MCF8316A and suggest our newer parts - MCF8316C and MCF8316D, where most of the erratas are fixed and we have a better support system. But I do understand that you have a solution in MCF8316A and we would be happy to support you to transition here Dave. 

I will talk to the team and provide you details soon (mostly by Monday)

Thanks and regards,
Sachin S

Any updates?

Hi David,

Sachin S said:

I was told that without the pull-ups on FG and nFAULT, the device could enter some sort of test mode that would cause this problem.  Is that still the case? 

This is an errata in MCF831xA devices, should not be an issue with the C devices.

Regarding the default address, I was on business holiday till yesterday. Will talk to the design team and confirm soon.

Regards,

Sachin S