Custom Teensy 4.1 - Device not recognized by PC (maybe wrong power-up sequence)

[Mic_n]

Hello everyone,
first time here on the forum, so my apologies if I'm doing something incorrectly.

I'm working on a project (a musical instrument based on teensy4.1) and decided to recreate the teensy circuitry instead of placing a teensy board on the back of my pcb.

I based my design on the teensy 4.1 schematics available on the PJRC website and omitted the the chips labeled as "optional" and the ethernet chip. I'm using MKL02Z32 bootloader chip and MIMXRT1062DVJ6B_BGA196 microcontroller purchased from the PJRC webstore.
Soldering process went fine and the pcb looks good, but when I connect the board to my laptop via USB the device is not recognized. The LED labeled as D4 on the teensy4.1 schematics start flashing approximately twice every 1.5/2 seconds.

According to the "Troubleshooting & Diagnostic Blink Codes" section of this page (https://www.pjrc.com/store/ic_mkl02_t4.html), two blinks mean that the NXP JTAG is not responding, so there's no communication between the bootloader chip and the microcontroller and this may be caused by either wrong/missing connections or a wrong power-up sequence.
I double-checked the connections on my schematics and they look fine, so, assuming the BGA chip is correctly soldered (I'm quite confident about that and I've already reflowed it), I thought the problem may be caused by the power up sequence.
I started measuring all voltages mentioned in the 8 steps of the "Power Up Sequence" section of the same page linked above and this is what I've found:
- my PMIC_ON_REQ pin is high with 3.1V and not 1.1V as mentioned on the page (I'm using this V.reg https://www.mouser.it/ProductDetail/595-TLV75733PDBVR, pinout is different from the package used on teensy4.1 board, but should be fine once you connect all the pins in the right way);
- the IMXRT DCDC buck converter does not produce any voltage: I have 0V on both sides of the 4.7uH inductor and the four caps! (This is the inductor I'm using https://www.mouser.it/ProductDetail/810-BCL322520RT4R7MD);
For what concerns the voltages mentioned in all the other steps, everything looks fine and I've a stable 3.3V exiting from the voltage regulator.

Any idea of what is causing this problem? Should I try an inductor with different specs?

I also realized I'm using a wrong flash memory chip: this page of the webiste (https://www.pjrc.com/store/ic_mkl02_t4.html) clearly says flash memory chip w25q64 must have an "M" at the end and w25q64-...-Q are not supported.
My chip has a "Q" at the end, just ordered the right one, and I'll replace it as soon as it'll be delivered. The "Troubleshooting & Diagnostic Blink Codes" section however says that a flash memory incorrect ID should be reported with 3 blinks and this is not my case (I'll probably have the 3 blinks error message once I'll solve the "buck converter" problem, I think).

Thanks to anyone

 

[PaulStoffregen]

I edited the web page info about PMIC_ON_REQ. It is indeed ~1.1V during the startup before the 3.3V regulator turns on, but then becomes 3.3V when 3.3V power is up. If the 3.3V regulator is working, you'd probably have to use an oscilloscope to actually observe it at 1.1V, since that state doesn't last long enough for a DC voltmeter to measure.

If you're not getting the buck converter to turn on, first check the signal at DCDC_PSWITCH. It's supposed to be logic high 3.3V to tell the DCDC converter to turn on.

Next check if you really have 3.3V at the DCDC_IN pins.

BCL322520RT-4R7M-D looks like it ought to work. If you're getting 0 volts at DCDC_LP pins, likely causes are no power at DCDC_IN or DCDC_PSWITCH didn't change from low to high, or something is shorted to GND.

 

[Mic_n]

Hi Paul,
thanks for the tips. I confirm I have 3.3V on both the DCDC_PSWITCH and DCDC_IN pins. None of the components (resistors and caps) look shorted to GND, I guess something went wrong when soldering the microcontroller, I'll try to desolder it and resolder in place and see what happens. I'll post updates on that.
The correct flash memory chip should be here within a couple of days, I'll see if something changes with the correct one even tho the chip ID shouldn't be related to this problem.

 

[slash2]

I realize this is a very old post, but just in case you are still out there Mic_n, did you ever resolve this problem, as I am having the exact same one. This is on all three boards we have built so far. The first board we built did get to 8 blinks, and then we realized we had the wrong processor. We change that out and it went to 2 blinks and no DCDC. Two more boards (with correct CPU) have never gone past 2 blinks and pass all the power steps except the CPU Power one.

BTW, Thanks Paul for the excellent descriptions and troubleshooting pages.