Programmer mode of various Renesas MCUs

[gashtaan]

I'd like to know more (anything) about programmer mode of Renesas MCUs, that is vaguely described in various datasheets like:

7.6 Programmer Mode
In programmer mode, a PROM programmer can be used to perform programming/erasing via a socket adapter, just as a discrete flash memory. Use a PROM programmer that supports the MCU device type with the on-chip Renesas Technology 128-kbyte flash memory.
(citation from https://www.renesas.com/us/en/document/mah/h836049-group-hardware-manual, page. 106)

Anyone have any experience on this? How does it work internally? What is the mechanism of communication between MCU and such "PROM programmer"?

[peter-h]

You need to get the original data book. It's all in there.

If you mean the chips I think you mean, a number of the CPU pins become the pins of a traditional EPROM (usually non-erasable although quartz-topped versions did exist) or FLASH.

For example the H8/323 (HD6473238F10) programs as a Hitachi 27C256 EPROM. We still have a big stock of these and program them regularly on an ancient Data I/O Chiplab programmer.

[gashtaan]

You need to get the original data book. It's all in there.

If you mean the chips I think you mean, a number of the CPU pins become the pins of a traditional EPROM (usually non-erasable although quartz-topped versions did exist) or FLASH.

For example the H8/323 (HD6473238F10) programs as a Hitachi 27C256 EPROM. We still have a big stock of these and program them regularly on an ancient Data I/O Chiplab programmer.

So to utilize this feature on my own I need to find out what EPROM is inside the specific MCU and what pins it's exposed to if I boot it in the test mode. The datasheet unfortunately omits this information (at least these I read so far). What do you mean by "data book"? I've never heard such term.

[peter-h]

What do you mean by "data book"? I've never heard such term.

Shows how the world has changed



Fortunately most data books can be found on the internet.

[gashtaan]

Oh I see, I searched for Renesas data books. Sometimes it's hard to follow who was who at specific time

[nctnico]

Seeing that photo makes me feel old. The (old) times that you needed hundreds of kilograms of datasheet books and reference manuals to get electronic designs done.

[peter-h]

But you didn't actually need all those books. Because you designed all the hardware yourself, you used only a small range of parts. I have a huge pile of those books (tried to give them away to anybody who will pay for shipping - posted it all here) but only a few were ever relevant to any design I did.

If an LM358 was ok, you used that one, $0.10, covid or no covid, and job done. And a Z80 (or H8, etc) was used for everything at all possible. Whereas as we often see here on EEVBLOG people are using all kinds of different CPUs. Some will get burnt doing that as say the 32F4 range gets rationalised but by then they will be in the next job

On the topic of Hitachi chips, they ran most of them for 25 years which is pretty good going. And then you got a LTB option, to stock up for another 10 years I still sell a product with an H8/323 which came out c. 1988.

[gashtaan]

I'm trying to find at least some info about data book of H8/36049 group (H8/300H Tiny Series), but nothing comes up yet. Is it possible that this MCU type was created after the Hitachi sold this technology to Renesas and Renesas did not publish data books anymore?

[peter-h]

Farnell still stock that one
https://ie.farnell.com/renesas/hd64f3687gfpv/mcu-16bit-20mhz-lfqfp-64/dp/3213725
Awfully expensive!

The hardware manual is here
https://www.renesas.com/us/en/document/mah/h83687-group-hardware-manual

Page 134 of the PDF (page 100 of the document) has the details. It looks like a newer implementation which is self-programmed i.e. it does not emulate an old EPROM. It is not simple.

FWIW, I would absolutely not use this chip for any new design. Not at that price. Some arm32 like a STM32F4xx will outperform it by 10x and cost 1/4 of the price. Hitachi/Renesas did good stuff (I designed-in an H8/323 in 1994 and that product still sells, and it was a good platform for C code; better than a Z80 for accessing variables on the stack) but then got left behind. The software tools were also relatively poor for Hitachi; the GNU compiler was crap so I used a Hitech C compiler.

Here
https://www.renesas.com/us/en/products/microcontrollers-microprocessors/other-mcus-mpus/h8-family-mcus/h836064-microcontrollers-32-bit-h8300h-cpu-core-non-promotion
it is labelled as obsolete.

[gashtaan]

The hardware manual is here
https://www.renesas.com/us/en/document/mah/h83687-group-hardware-manual

Page 134 of the PDF (page 100 of the document) has the details. It looks like a newer implementation which is self-programmed i.e. it does not emulate an old EPROM. It is not simple.

I'm not interested in software method flashing you are referring to, this is well documented. I'm curious about accessing the flash directly, see the 140 (106) of the same datasheet. At least, there is mentioned what flash is used in the MCU (FZTAT64V5). There is something interesting at page 524 (490) - it seems the EEPROM in stacked on some corner of the main die, it probably has a serial interface according the bonding wires count... not sure how the picture is actually accurate though.

Anyway, this is all related to H8/3687 group, not to H8/36049 I'm interested to, but probably there are some similarities, so thanks for the direction The data book for H8/36049 would be very helpful, if any has been published...

I'm not going to use this MCUs in the new design, it would not be a wise thing to do for sure