Thanks for reaching out!
The Joule module is very sophisticated, and as any modern system on module it requires some very specific circumstances on boot time in order to boot. There are two documents that will help you get this task done, the module's datasheet and the expansion board design guide, they can be found at:
In specific, check section 3 of the datasheet and section 4 of the expansion board design guide. These documents will provide the information needed to boot the board as expected.
Now, since your board is not booting, I would firstly suggest you to place it on the standard expansion board to see if it is still functional. This is just to make sure that there wasn't any major issues on your custom board that could damage the board.
Anyhow, you might have already read the documents I shared above before posting this thread, so, you might be doing everything right but not getting any output from the debug port. If that is the case, did you make sure that the FTDI circuit was correctly placed? I suggest you to check section 8.1 of the expansion board in order to learn how this is circuit should be placed.
If you still have issues after checking the information above, I'd suggest you to check the schematics of the expansion board (Schematic for the Intel® Joule™ Expansion Board ) and based your board in its circuitry.
I hope this information helps you,
got a little further, after all, was a stupid thing. the DNX pin was up all the time, so the system was trying always to update
But, still not working correctly. Sometimes, it simply boots up for a second, shows the splash screen and falls asleep again (?).
As I'm not using the BQ25892 power controller, what is the correct pullup voltage on the PMIC_PWRBTN_N line?
The BQ25892 has an internal pull up, but the data sheet doesn't tell the voltage level ...
Thanks in advance!
Since you mentioned that you are not using the BQ25892, I assume that you are developing a custom board, am I right? If so, I must mention that this is not covered under the support scope, therefore, we might not be able to help you. However, we'll do our best effort to see what might be happening.
I usually recommend custom board designers to use Joule's expansion board schematic (http://www.intel.com/content/www/us/en/support/boards-and-kits/intel-joule-kits/000022329.html) as an example, this is usually the best approach for design issues.
Since your question is very specific (the voltage in PMIC_PWRBTN_N), I decided to check it myself. So, I checked the schematic for the signal PMIC_PWRBTN_N, it only appears three times. In sheet 8 you can see that it is connected to the BQ25892 as an input on the pin QON.
So, the first thing I thought was to check the BQ25892 datasheet (http://www.ti.com/lit/ds/symlink/bq25890.pdf) to see what this pin does and what its operational voltage is. According to the datasheet, the QON pin provides BATFET enable/reset control to exit low power ship mode or full system reset function and it is active on low (this can also be appreciated on the schematic).
In the datasheet I found that this pin does not work on a fixed voltage but instead a range, so the information on the BQ25892's datasheet is of no help in this specific circumstance. So, the next steps is to check the rest of the PMIC_PWRBTN_N labels on the schematic. One of them simply indicates the pin in which the signal arrives to the Joule module (9 of J2). The other label is more interesting, since the QON pin is active on low, the PMIC_PWRBTN_N signal should be most of the time on high and only on low when the signal is active.
In sheet 22, you can find the last PMIC_PWRBTN_N label. Here you will see that it comes from SW2 (the power button or PWR BTN as seen on the expansion board itself), in fact this is the only place that you'll see the signal PMIC_PWRBTN_N marked as output. Anyhow, there is something strange about this, according to the schematic, it is not connected to any "power source" that pulls up the signal.
If you compare the schematic of the general purpose button and the power button (both on sheet 22) you'll see what I mean. In the general purpose button +VDD1 (1.8v) pulls up the BTN_N signal, I believe PMIC_PWRBTN_N should have the same configuration, however, I can't assure that this is the case. There is a possibility that the pin QON on the BQ25892 can be left like this (not pulled-up). Nevertheless, I will have to investigate more about this to see if this is an error on the schematic (I believe it is) or if this in fact the way PMIC_PWRBTN_N works. I will let you know as soon as I have any useful updates.
We have an update for this case.
In section 3.1.1 of the Joule expansion board hardware guide (http://www.intel.com/content/www/us/en/support/boards-and-kits/intel-joule-kits/000023394.html) it is mentioned that when the button is pressed the PMIC_PWRBTN_N signal is pulled down. When the button is released, the PMIC_PWRBTN_N signal is pulled high to +VSYS.
In section 3.4 of the Joule's datasheet (http://www.intel.com/content/dam/support/us/en/documents/joule-products/intel-joule-module-datasheet.pdf) you can see that the system voltage rail specification. It mentions that +VSYS 3.6V min, 4V typically, 5.25V max.
According to these documents, the signal PMIC_PWRBTN_N would typically be 4V but it would vary accordingly to the state of the Joule as mentioned above.
Just a short update, the joule module seems to have a pullup on the module itself, so removing my "wrong" pullup to vdd1 did the trick, and the board now boots every time.
Just in case anybody looks for it ...
I understand, that makes sense. If the pull-up resistor is within the module itself it will not be shown on the schematic of the expansion board which is what originally confused me. I appreciate that you share this information with the community, I'm sure it will be of help for future users and it is of help for me.
If you ever have any other questions, issues or doubts, please come back to the community and we'll try our best to help you.