From: Davis, Mark

Sent: Friday, September 16, 2011 11:17 AM

To: Usher, Nathan

Cc: Drewyor, Brian

Subject: FW: LLRF2 schematics

NOTE:  Minor correction to my suggestion regarding the Rabbit’s connection to the SPI2 bus:  The Rabbit should connected ONLY when BOTH /DSP_RESET (PD6) and PROG_B (PD7) are low.  It needs to have PROG_B high and /DSP_RESET low to restart the FPGA, during which the FPGA must be the master, so it must be disconnected from the SPI2 bus under these conditions.

From: Davis, Mark
Sent: Friday, September 16, 2011 8:39 AM
To: Usher, Nathan
Cc: Drewyor, Brian
Subject: RE: LLRF2 schematics

Nathan (cc Brian),

Regarding the LLRF2 schematics:

Don’t know if it is you or Brian that needs to do this, but don’t forget about the corrections for Rab_SPI2.MOSI and Rab_SPI2.MISO.  The schematic still shows these as being connected to pins 21 and 22 (PC6 and PC7) which is serial port A, which is reserved for the Diagnostic/Programming port.  The SPI2 xmit/rcv signals have to be connected to serial port C, pins 17 and 18 (PC2 and PC3).

Also a reminder that the Rabbit cannot set its xmit signal for any of the serial ports to input mode or do anything else to set them to a Hi-Z state when it is not using one of the SPI busses.  What this means is that something will need to be added on the LLRF2 board to disconnect the Rabbit’s serial xmit signals whenever they might interfere with any other use of the SPI busses.

As we discussed, this is definitely an issue for the SPI2 bus, so the Rabbit will need to be disconnected from the Rab_SPI2.MOSI signal when it is not using the SPI2 bus.  The easiest would probably be to use /DSP_RESET (PB6) and  PROG_B (PB7) for this.  The Rabbit will not try to use the SPI2 bus without setting at least one of these to a low level (both for all circumstances I know of at present), so if the Rabbit was disconnected from the SPI2 bus when both were high, then that should do the job.  NOTE that the Rabbit CAN set the pin it uses for the SPI2 clock to an input (HI-Z) state when not using it, and I don’t think it will be a problem if the Rab_SPI2.MISO signal is always connected to the Rabbit, so only its connection to the Rab_SPI2.MOSI signal is an issue as far as the SPI2 bus is concerned.

Also, as I understand it, the only master on the SPI1 bus is the Rabbit, so there shouldn’t be a problem with having it always connected to the Rab_SPI1_MOSI signal.

I hope to have some code for you to try later today or sometime Monday, but of course the prototype board will have to have these changes done before it will function.

Mark

From: Usher, Nathan
Sent: Thursday, September 15, 2011 2:26 PM
To: Davis, Mark
Subject: Re: LLRF2 schematics

Oops, I forgot that was a per-file option, not per-project.  I redid them, and they should all be correct in the attached zip file.

Nathan