RAA229001 VRM Controller - programmable or not ???

[DynaxSC]

Hi electronic folks

Does anybody know if this controller is programmable or not ?

I have replaced this controller on a MSI MAG B560 TOMAHAWK board, as it had a short on VCCS pin (pin 41). I have ordered a few new on Aliexpress, they seem to be original, however they do not start to work. When I move same chip from a second working board to the one I have replaced the RAA229001, the VRM starts normally to work. It seems that the chip is OK, as it generates on the VCCS pin nearly same voltage as a chip on a second working board i have (VCCS probably it's some internal reference voltage, as only component connected to this pin is a 10uF cap).

Now either the chips I bought on Aliexpress are fake, or they need to be programmed somehow. There is only a RAA220001 driver datasheet available (inside is a diagram with RAA229001) but there is nothing about programming of the RAA229001 there. However i found some references on some pages with programmers, that they are compatible with the RAA229001, eg:
https://www.elnec.com/en/device/Rene...BQFN48-6x6%5D/
The device has an SM Bus interface, so maybe this is used for programming ?

If anyone has a more deep knowledge about this topic, I'd appreciate to share it. Also how to program this annoying controller would be nice to know.

Below a screenshot from the schematics part of the a.m. board with the VRM part. Maybe this will help others, as the Renesas datasheet is shit as it does not contain pin number information (have no idea why they do not disclose this information).

[lotas]

Re: RAA229001 VRM Controller - programmable or not ???

Usually, dots of colored paint are applied on programmed chips from the factory - this means that the chip has been programmed and has passed the test, so that it still needs to be programmed.

[DynaxSC]

Re: RAA229001 VRM Controller - programmable or not ???

You are right, the chip on the working motherboard has a dot. So need "only" to find out how to program this bloody chip, or better copy the configuration. Can be a nice riddle.

[lotas]

Re: RAA229001 VRM Controller - programmable or not ???

Nothing is known about this chip, but some microcontrollers of the same company "Renesas" have read protection, that is, when you connect a programmer and try to read the firmware data, this data is simply deleted and you get a non-working controller.

[DynaxSC]

Re: RAA229001 VRM Controller - programmable or not ???

Thanks for the warning. I already been suspecting that they protect the firmware.

I discovered that Renesas has some software for configuring and programming their VRM chips:

-> PowerNavigator - for generating of the VRM schemes and configs
-> ProductionConfigTool - this is probably for the programming of the config file (firmware) into the VRM chips

Theoretically having the software it should be possible to generate a config file from scrath, and then program it. But w/o any experience it might be difficult.

Another issue is that this software needs probably some specific hardware programmer.

Does anybody know what kind of programmer hardware is possible to use ?

[DynaxSC]

Re: RAA229001 VRM Controller - programmable or not ???

Discoverevd that for programming the RAA chip, as well as lot of ISL Renesas VRM chips one needs an USB to PM Bus Adapter called ISLUSBEVAL1Z.



This thing is not in stock at any of the known distributors, but it seems there is some probably clones available on Aliexpress. The above user manual has even a schematic diagram, but w/o the 8051 CPU code it's useless.

Also some of these ELNEC programmers can probably do the programming, they are however extremely expensive, so no option for a normal man. But it also seems there are some clones available, but opinions are not good, as they do not work reliably.

Does anybody has any experience with this chines devices, do they work ?

[DynaxSC]

Re: RAA229001 VRM Controller - programmable or not ???

I got now one of these china programming modules, and they do work with the respective Renesas software. I was able to attach the motherboard to the programmer and I can discover the VRM chip on the motherboard. It is possible to monitor the status of the vrm chip and send commands and program it in circuit, however the design of a working programming parameter set is not easy, as there are a lot parameters to be set, which must match the design of the board, eg the no of phases, addresses, the used SVID IMVP standard, and a plenty more. I was even able to remotely switch on and off the VRM chip using the PM/SMB/I2C bus. Naturally it would be better to read out the parameter set instead of reinventing it, haven't tried this yet, but will try. However this will be probably blocked at factory programming, so there will be no other way as to reinvent the parameter set. The issue is here, that the vrm chip has a limitation of number of programming steps, I found at the moment two limits for different areas, one is 30, the other is 10, so the number of tries is limited, but the number of programming options is almost unlimited. I hope here the the engineers at factories are lazy and do not tweak a lot of the parameters, and use basically the default settings, so it might be the chance that only the most important parameters must be set correctly. Will see what comes out. If anybody has some knowlege about this which could help to reinvent the parameter set, it would be greatly appreciated to share.