CSA,

Can you explain why there is no batteries inside the UCSBH1A controller
of MARK VIe system for storing the program during power fail? I have studied somewhere that the RAM used is NVRAM.

Can you elaborate??

 

Chiranjeevi,

When Product Code and Application Code are downloaded to the UCSx card (and I'm talking about initial programming, as well as any downloads made after a Major Difference resulted from some change), they are downloaded to a flash memory card installed in the UCSx card (usually in a slot in the top of the module). This flash memory card is non-volatile memory, and can be used in cameras and recorders to store photographs and videos. And, when those cards are removed from the camera/recorder they don't lose the information stored on them.

Non-volatile memory does not require any battery to retain the information. Some control system manufacturers use battery-backed RAM (BRAM) to store programming/configuration information, and if the battery fails or is disconnected the contents of the memory "module" are lost and gone.

But, GE stopped that practice back in the 1980s, and started using EPROMs (Electronically Programmable Read-Only Memory) and EEPROMs (Electronically Erasible Programmable Read-Only Memory) --and now non-volatile flash memory cards--to store programming/configuration information in the event of loss of power to the controller.

The controller actually uses volatile RAM (just like most personal computers, but in a different form) when controlling and protecting the turbine and auxiliaries. Volatile RAM loses it contents when power is lost, however, volatile RAM is, I believe, faster than non-volatile RAM for fast-changing data and reads and writes (and remember, many turbines run their application code ("sequencing") at a 40 msec rate (25 Hz). So, it's more suitable for high-speed control and protection (and fast PC programs)--but the contents are lost when power is lost.

When the controller is powered-up it goes through a boot-up process--and one of the things that happens is that the controller looks at the RAM and realizes that it's empty, and opens a communication channel to the flash memory card to get the programming/configuration it needs to operate. Once the information is downloaded from the flash memory card, that communication channel is closed, and the controller only uses the information it has in RAM for controlling and protecting the turbine and auxiliaries.

When someone makes a change in ToolboxST that results in a Minor Difference, and does a download to the Mark VIe, one is downloading directly to RAM and changing the contents of RAM. This can be done when the unit is running, but it should be done with an understanding of what might happen. It is possible to trip the unit, but not likely--it would take a special set of circumstances, and changes, for this to happen. So, not EVERY Minor Change download is guaranteed to not cause a turbine trip.

When someone makes a change in ToolboxST that results in a Major Difference, it's necessary to download the change to the flash memory card. And, to get the change to go to RAM for the controller to use one has to re-boot the controller--or, cycle the power to the controller to get it to go to the flash memory to get the new programming/configuration information to use to control and protect the turbine and auxiliaries.

I also believe there is a small non-volatile RAM memory section in each UCSx card, called NVRAM or NOVRAM, where data such as fired hours and the numbers of trips and starts are stored and protected from loss if power is lost to the UCSx. I also believe that on a periodic basis the contents of NOVRAM may be written to the flash memory card for further safekeeping. I haven't researched this too much, but based on how GE did this on previous Mark* control systems it makes sense they would do this.

Hope this helps!