Crystalfontz 128x64 LCD shield vs example code on nRF52DK

[howard n2wx]

Has anyone modified the spi_master_using_nrf_spi_master example (or otherwise gotten) the Crystalfontz CFA211TFH 128x64 panel running w/o mods on the nRF52 DK? I have a panel that runs well on an NXP KE06 (this is a low end M0+ core with very few periphs) eval board but I'm having trouble getting the board up on the DK's example. The CFA211TFH has an ST7565P.

EDIT:

My post doesn't have a lot to go on; I'll try to fill in what I hope is helpful additional info.

The significant differences between the nrf example and my cow-worker's implementation on a KE06 are:

• KE06 implementation has a 128x64 rather than the 128x32 of the example
• My Crytalfontz panel has a ST7565P rather than the ST7565R of the example's expected Arduino shield
• KE06 implementation has CPOL=1 and CPHA=1 (mode '3' in spi_master terminology), rather than the mode 0 of the example, but I'm not sure this is significant
• Also probably not significant is that my panel's SCLK and MO are coming out on Arduino shield pins 13 and 11 respectively, and I changed these to match in the example code.

Here's a partial scope capture of the nRF52-DK sending two bytes 0xAA and 0x55. It's a little truncated off the end but the last byte finished out as it started:

Yellow is /cs, red sck, blue is MO, green is A0 (I'm sending it as data)

[howard n2wx]

Problem solved.   This LCD panel has two (count 'em) chip selects: /CS1 and CS2.  Where'd CS2 come from? In any event once I forced that one high all the time and fixed the socket it worked as expected.