Potential Error in LC Schematic?

Constantin

Constantin

Well-known member
Hi Paul,

I did something sacrilegious the other day, I 'extracted' a MKL02Z bootloader chip from a existing LC board to find a new home on a MKL26Z-based power measuring device I am trying to develop. I'm so excited re: trying out the LC chip in this capacity that I couldn't wait, sorry! :cool:

However, try as I may, the thing isn't responding when plugged into the USB port (i.e. it doesn't enumerate). I found no shorts with my multimeter, and then I went over my schematics twice and found no error there either (vs. the posted schematic).:confused:

Today, I thought I'd compare the posted pic of the Teensy LC vs. the schematic (thank you for posting that!) and I wonder if I may have found an error in the posted schematic vs. the actual board shown here.

Note how on the bootloader chip there is a connection at pin 10 (the top red arrow) but no apparent connection on pin 11 (the bottom arrow)
Teensy LC.jpg

I suspect that the PGM button is attached to pin 10 in real life and not pin 11 per the schematic (see below). On the off chance that I'm right, would you recommend simply jumpering the two pins with a blob of solder, or does the MKL02Z dislike that? (i.e. in case pin 11 is not high-Z by default like most of the pins on the MK20)
Teensy Schematic.png

I'm not home now to test the connections with a multi-meter and I am more than happy to be wrong... so please accept my apologies in advance if I am! But sight unseen, I'm willing to bet an ice cream sandwich that pin 10 is attached to PGM. Anyone want to take me up on that wager? :D
Or can someone use a multimeter to see if pin 10 of the bootloader is electrically attached to one side or the other of the PGM button? Or that pressing PGM causes pin 10 to go to GND?

Note: This potential error, if it is one, will not affect anyone except those who want to roll their own Teensy LC. Currently, you'd have to scavenge a LC boot loader off an existing board or wait until PJRC starts shipping them individually. Thus, even if this turns out to be an errata, it does not affect pre-assembled boards from PJRC.com. Buy a few and enjoy them - I know I do!
 
Last edited:
You are amazing and thanks! I tried deleting the thread and then re-posting in the correct forum before realizing that I couldn't delete my own thread! (oh well).

Thanks again!
 
Ok, based on preliminary measurements wirth a crude and rude fluke 87 DMM it indeed appears that it's pin 10 connected to the PGM tactile switch. I could successfully toggle between a high resistance and 0.3 ohms by pressing the button. Meanwhile, pin 11 seemed unimpressed, ie no connection to ground nor any change in resistance as a function of pressing the button. Looks like pin 11 is not connected to anything.
 
The teensy LC 'clone' works once pin 10 and 11 were joined electrically with a blob of solder. My conclusion hence is that a very small portion of the currently published schematic for the LC does not reflect reality.

Further, I need to update my preliminary Teensy LC library published here to reflect the incorrect pin assignment.

I was able to erase the MKL26 and upload programs such as blink, analog read, and so on. Looks like the incorrect pin assignment for the PGM prevented the initial erase and prepare for a USB upload.
 
Last edited:
I wonder if switching the Teensy 3.x to this boot loading chip would not enable a simpler solution to the D33 issue on the Teensy 3.x series by opening up two more MK20 pins to use as a substitute for D33. Pin 26 (currently D33) would then have no need to be connected to anything. Teensyduino would have to keep track of what pain assignments to use (legacy vs. current bootloader) but presumably it can track the version of the bootloader it encounters and hence choose different internal pin assignments from the boards.txt file accordingly.

The smaller boot loader chip used on the LC is surprisingly easy to touch up with a soldering iron, just like the 5x5mm version of the mini54. I suggest to anyone with a reflow oven to try out that version versus the pinned MINI54 as it is possible to achieve a via-less connection between the main MCU and the boot loader, even on a two layer board.

The boot loader on the LC does feature a strange (to me) reset signal pad arrangement where the pads are not adjacent and require a via to connect. I am sure there is a very good reason for these pad choices that I'm simply not privy to.

What I did find interesting however is that the default state for the LC MCU pins appears to be HIGH. One of the pins is attached to a LED and it is lit continuously unless you program the pin different. That worries me a bit because I hoped that the LC MCU would behave similarly to the MK20, ie most pins are high-Z on startup, unless programmed otherwise.

Hiwever, the LED may simply be very sensitive. This series from OSRAM is designed for 2ma operation.

Last but not least: the error regarding the pin assignment on the boot loader chip does not affect anyone buying a pre-assembled board from PJRC or any other outlet. This issue only affects those who extract a boot loading chip from an LC.
 
Last edited:
Yes, you're right, the schematic was incorrect. The pushbutton connects to pin 10 (PTB2). My mistake!

I've updated the schematic. Thanks for reporting it. :)
 
You must log in or register to reply here.
Back
Top