Can't run project.

Status
Not open for further replies.
Some of us were wrong.
I have 18 pins different.
Look at the picture please. Capture.PNG
 
Oh, I see - the wire is hidden by the silk screen.

Can you check with a voltmeter that pins 3 & 4 on the MKL04 really are getting 3.3V?
 
Yes. Checked.
There on both of them is 3.3V from the power stabilizer. Just like on 5 and 6 there is ground.
If in doubt, I can record a video with all the measurements that you will tell about.
Thank you Paul.
 
Let's look at pin 18 (Program pin) again. If you added a resistor, please remove it. Do you still see 0.18V on that pin? What happens to the voltage when you press the button?
 
Very strange situation.
Removed the pull-up resistor.
Removed the button, closed the wires of the multimeter.
Not closed (which is strange) voltage at pin 18 - floats from 0.806 to 0.954.
When closed to earth, 18 is naturally 0.

Connected oscillograph.
No pickup, ripple, just floating DC voltage.
 
Last edited:
Good morning.
Our next steps?
I can dismount the MKL04 and mount another one.
If there is a positive result, we will close this case.
Let's write everything down on the defective chip.
If not ?
Chips I do not solder hot air. He gently soldered the pin by pin with a soldering iron,
pressing the temperature sensor to the chip body with a finger, without heating it more than 50-60 degrees C (122-140 F).
 
Last edited:
I will start a new build.
There is something wrong with the assembly.
I'll start with power circuits and MKL04 without MK66.
I will check at every step that at 18 pin.
By the way, in the current assembly, the DC resistance at 18 pin is about 5 MΩ, that is, there is no contact with the ground.
 
As a sanity check, a couple days ago I did some tests with the MKL04 chips. When the chip is powered with 3.3V, and a 2M resistor is connected between pin 18 to GND, but nothing else is connected other than the 3.3V power and that resistor, pin 18 measures 3.25 volts.

I must admit, I do not understand how the chip could be failing to pull up pin 18 to 3V. It should always do that. Your layout looks like the 3.3V power is good, and you've verified it with a voltmeter.

The only other cause I can imagine would be pulling the MKL04's reset pin low (pin 31), so it doesn't run. But that pin is unconnected on your board.

Very mysterious problem...
 
Yes.
It's hard not to agree.
18 is completely untied at the moment from everything. It is simply soldered to the contact pad, which does not lead anywhere.
As I said earlier, the resistance of the pin and the pad relative to the ground is 5 MΩ.

Indeed it is strange.
I could still make problems with power. But power is supplied from the stabilizer and the track resistance from the control point KT1 to 3 and 4 pins is 0.01 ohms.
With the micro currents that the MKL04 consumes, this resistance can be considered zero, it doesn’t fall on anything, as shown by the voltage measurements on pins 3 and 4, it is equal to the supply voltage.

But I think it is worth pausing.
In addition to the assembly on a clean PCB, I will pre-check all the tracks not only for breaks but also for conductivity, especially at vias.
To begin, mount the power and only MKL04.
In the end, I think we'll figure it out and find the reason. I hope so.
Thank you.
I will definitely report the results.
 
Last edited:
Hello colleagues.
This morning I was engaged in experiments.
He simply repeated Paul's experiments both with MKL04 chips that were already soldered in the device, and with new ones not yet used.
Those chips that were soldered in the device are no longer serviceable. There were 2 of them. Both of them on pin 18 have 0.2-0.5 volts.
New ones taken out of the box, both previously sent and last sent, have 3.3 V on pin 18.
From here, make the sad conclusion, something with the PCB and everything else is wrong.
 
If you remove the MKL04 chip from your board and power it up, do any of the 32 unconnected pads measure more than 3.3V or a negative voltage?
 
Pad -> Volt
3, 4 -> 3.3
5, 6 -> 0
8, 25 -> 3.3
10 -> 3.3
15 -> 3.3
16 -> 0
17 -> 0.6
18 -> 0 or network interference and induced emf, no more than 40mV
all other (not connected) 0 or network interference and induced emf, no more than 40mV
All tested pins (pad) positive voltage.
Measurements made by the device with an input resistance of 10 MΩ
 
Last edited:
Without bootloader MKL04
USB detected the same as
idVendor: 15A2x0073
0x0409: "Freescale Semiconductor Inc."
0x0409: "Kinetis Bootloader"
 
0x0409: "Freescale Semiconductor Inc."
0x0409: "Kinetis Bootloader"

This is normal, if the K66 boots up. NXP ships those chips with a "flashloader" program present in the flash. It's documented in their reference manual, chapter 14 starting on page 277.

I have no idea why those MKL04 chips are getting damaged when on your board. Something must be wrong, but I'm having a hard time imagining what could do this, if all the pins are in the 0 to 3.3V range and the 3 programming pins (30, 31, 32) are unconnected.
 
I hope with your help we will finally figure it out.
In any case, everything will become apparent later.

Unfortunately, the new PCB with a slight refinement and bug fixes will come only on Monday.
I will report on progress, if there is not any new ideas.
Thanks.
 
There is one stupid idea.
Could this be due to the thick board?
Board 1mm. Parasitic capacitance tracks?
 
When you get the new PCB, maybe solder only the MKL04 chip and minimum parts needed for 3.3V power. Hopefully you will measure 3.3V on pin 18? Of so, then try adding more parts and keep checking pin 18 after each group of parts....
 
Yes.
It's hard not to agree.
18 is completely untied at the moment from everything. It is simply soldered to the contact pad, which does not lead anywhere.
As I said earlier, the resistance of the pin and the pad relative to the ground is 5 MΩ.

Indeed it is strange.
I could still make problems with power. But power is supplied from the stabilizer and the track resistance from the control point KT1 to 3 and 4 pins is 0.01 ohms.
With the micro currents that the MKL04 consumes, this resistance can be considered zero, it doesn’t fall on anything, as shown by the voltage measurements on pins 3 and 4, it is equal to the supply voltage.

But I think it is worth pausing.
In addition to the assembly on a clean PCB, I will pre-check all the tracks not only for breaks but also for conductivity, especially at vias.
To begin, mount the power and only MKL04.
In the end, I think we'll figure it out and find the reason. I hope so.
Thank you.
I will definitely report the results.
I'm just taking a wild stab at this from what you said in first sentence here. Pin 18 should be going to your program switch or button. I think you've done the same thing as I have on our first run. If you matched the pjrc schematic, it has the MKL02. The pinout for the MKL04 is different. Pin 10 is the program pin on the schematic. But Pin18 is the program pin for the MKL04. Hope this helps.
 
I'm just taking a wild stab at this from what you said in first sentence here. Pin 18 should be going to your program switch or button. I think you've done the same thing as I have on our first run. If you matched the pjrc schematic, it has the MKL02. The pinout for the MKL04 is different. Pin 10 is the program pin on the schematic. But Pin18 is the program pin for the MKL04. Hope this helps.

Thanks for the tip.
Please see the first post in this thread.
There I will attach the scheme by which the assembly was going and according to which it is being manufactured now.
There's pinout MKL04.
 
Good evening.
There is news, but sad.
It all happened again.
Soldered only the power circuit and MKL04.
At 18 - 0.2V and 3.0V if the foot 3 is suspended.
3 and 4 - 3.333V at power supply.
5 and 6 resistance to ground 0.001 Ohm.
8 and 25 - 3.332V
I'm at a loss.
 
Last edited:
Do you have an oscilloscope? Maybe capture the power supply voltage at startup? Perhaps there overshoot or some other problem that's damaging the chip?

Or do you have a known-good Teensy? And another bare PCB? Maybe solder only a brand new MKL04 chip and absolutely no other parts to the PCB, not even any decoupling capacitors, so the MKL04 is the only part. The connect the GND and 3.3V power to the known-good Teensy, so it can power up the MKL04 chip. Then measure the voltage at pin 18.
 
Status
Not open for further replies.
Back
Top