lpcware

LPC2214/01 - Possible Silicon Bug?

Discussion created by lpcware Employee on Jun 15, 2016
Content originally posted in LPCWare by dunno.. on Mon Mar 16 13:40:34 MST 2015
Hi all,

i got redirected here by hints of Forum-Users at http://www.mikrocontroller.net/topic/361466 - This original post is in german, so im going to sum it up here:

today, i was able to find a bug where a LPC2214/01 run at 60Mhz showed irrational behavior in his SSP-Interface. I was trying to get a 30MHz SPI working with a Peripheral IC. (Design is a 4 layer layout, short traces, decoupling etc. with all the bells and whistles)

The signals looked quite ok on the scope, but the SSP-Interface read only gibberish, or even just zeroes (8 Bit Transfer).

At 15Mhz or lower, all worked just fine.

I got a hint by a colleague, who told me he had read about a issue with an LPC2103 in the dedicated errata sheet, about the devices not working properly when consuming much energy at high frequencies on the 1V8 Rail. (http://www.nxp.com/documents/errata_sheet/ES_LPC2103.pdf
-> 3.7 Page 9

In this Errata, NXP suggests to increase the 1V8Rail to 1.95V or thereabout.
I have found this Workaround (Increase 1V8 to 1.96V and disabling unused peripherals) to fix my Problem with the device.
Even though i cant tell for sure we havent screwed something up with the layout, i consider this beeing worth a note.
Maybe someone can even confirm or reject this, so we may get a hint for further users in the errata.
Next, i will be checking my designs with LPC 2124 for this.

regards
dunno..

Outcomes