Sunday, September 7, 2014

Minima Controller Shield checkout

UPDATED:

My Minima controller Arduino shield checks out as functional.  With two Si570 devices on the board, powering them from the Arduino regulator is pushing things, so I have isolated the 3V3 rail from the Arduino and provided my own regulator.  However, with only a single Si570 on the board, I am within limits on the Arduino regulator.  Therefore, I jumpered the power input pin back to the Vin pin on the Arduino to supply power to the shield.  This is the brown jumper wire seen below.  The yellow wire is the ground to my scope probe.



Measuring the voltage off the regulator and the Si570 look good.  Probing the output of the Si570 as seen above, we have the following default frequency of 56.32 MHz output seen below.



For those following along, the schematic of the shield looks like this.  There really isn't anything to it.  Just a multiplexer in front of two Si570's, voltage regulation and breaking out some of the I/O pins to separate cable connections for the Minima.




Now to go figure out how to talk to the multiplexer and the Si570.

I removed the shield from the Arduino Uno and connected up a bus pirate to see if I can talk to the PCA9546.  When searching the I2C address space for devices, initially I got a "response" from all possible addresses.  Clearly this is wrong...

Probing the PCA9546 I found that the voltages were correct, but that the reset pin was held low.  As seen from the schematic above, this is connected to the reset button and the ICSP reset pin.  However, this pin on the shield is not pulled up to any voltage rail and currently relies on the connected UNO to supply this voltage.  So, rather than power the shield from the bus pirate, I plugged it back into the UNO and reinstalled my brown power jumper and just monitored SCL and SDA pins with the bus pirate.



Now interrogating the bus yielded the following results:

Bus Pirate v3
Firmware v4.2 Bootloader v4.2
DEVID:0x0447 REVID:0x3043 (B5)
http://dangerousprototypes.com


I2C>(1)
Searching 7bit I2C address space.
   Found devices at:
0xE0(0x70 W) 0xE1(0x70 R) 0xFE(0x7F W) 0xFF(0x7F R)

Checking the PCA9546 datasheet, we see the following:


If you look at the schematic above, you will notice that I connect A0-A2 to ground, so the device address should be 0x70 (this value is the 0xE0 shifted right one bit to remove the read/write bit) and the read/write addresses should be 0xE0 for write and 0xE1 for read.

So, it looks like the PCA9546 can be seen on the bus and so now it might be time to write some actual code.

No comments:

Post a Comment