New Build: Not getting 5V or 3.3V


 

KevinL

New member
Hi

I recently completed my HM 4.3.4. I connect my Pi Zero W to it yesterday both with 12v and USB. It booted right up.

Today, I was starting to work with it again and it fried my Pi Zero. I then check the 5V points and I'm getting 10v. On my 3V points Im getting 6V.

Any help of where to start would be great!

Base board pictures: https://imgur.com/a/8mUI2yN
 
Well, all the parts look to be in the right place, apart from missing a 100k resistor by the pit probe jack. A picture of the solder side of the board too would be useful, unless I'm missing a way here to get to another image. I'm not sure how you could end up with 10V on the 5V unless something backfed 12V through the 5V regulator... which still wouldn't cause the 3.3V bits to be 6V (since that should be fine taking 10V down to 3.3V by itself). I think you're going to have to start pulling parts off until voltages start looking right on the PCB. The ATmega can come out (which also might be fried since its Maximum Operating Voltage is 6V), but that's probably not going to change anything since it doesn't drive any of the supply voltages.
 
Thanks for the quick reply.

Concerning, the 100k resistor - is the required even if using the thermoristor? The instructions made it sound like it was 10K for Thermoristor or 100k for thermocouple (not both)

Opps, looks like the solder side didn't upload, here it is: https://imgur.com/a/VYlE5iW Excuse the long cut offs - I need a new cutter

Guess what, I just took the ATmega off and reseated it. I'm getting the right voltage now. I'm not sure if my multimeter just gave a wrong reading or something.
 
Thermistor probes need both the 10k and 100k resistors. The 100k is part of the RC filter circuit and must be present for all probes, the 10K is the pullup resistor that is required for thermistor probes only (thermocouple does not need a pullup resistor)
 
Thanks for the info!

I'll test some more on my board about the voltage. I'm 99% sure this is just me being stupid. I likely had a bad reading from my multi-meter/wrong setting and I likely matted the Pi on the wrong pins which cause it to fry. I don't think there is anything wrong with my board after all.
 
Im at a lose, I fried another Pi Zero.

I tested the baseboard/LCD only on 12V and all my 5, 3 and 12V points. They all came back good.

I matted the Pi to the HM and powered the Pi via USB. The HM came up great, every probe worked great, and no issues reported in the GUI.

When I tried the 12V - the screen came on but the Pi wouldn't boot. Re-tested with the USB and the Pi seems to be dead.
 
Did you check the voltages on all of the RasPi header pins when the HM is powered up via 12v to ensure than only 5v is going to the 2 x 5v pins on the RasPi header?
 
Yes I tested the Pi two 5V and G25 (3.3) points and they came back normal.

For what it’s worth when powering the Pi via USB. The LCD lights up and works as expected.
 
Have you checked the HeaterMeter base board just by itself, powered with 12V and go through every one of the RPi header pins on it and check them for weird voltages? We should only be supplying 5V on those two bottom right pins, and then only the RXD, TXD, SCK, MIS, MOS, and G25 are connected. The rest should just be floating or GND. Nothing higher than 3.3V on those listed pins, apart from the 2x 5V pins.
 
Have you checked the HeaterMeter base board just by itself, powered with 12V and go through every one of the RPi header pins on it and check them for weird voltages? We should only be supplying 5V on those two bottom right pins, and then only the RXD, TXD, SCK, MIS, MOS, and G25 are connected. The rest should just be floating or GND. Nothing higher than 3.3V on those listed pins, apart from the 2x 5V pins.

With baseboard only on 12v - I did not see anything other than 5 and 3.3. (on the Pi connector)

Am I suppose to be seeing something on SCK and MIS? Or are you just saying that's the only PIN actually connected.
 
Last edited:
Huh, that's really strange then. You don't have a microdamper with the TRRS headphone jack cable do you? I know there can be issues if that cable is inserted while powered.

There shouldn't be anything on MIS, but SCK should toggle very quickly (at 2MHz) every time the LCD updates, around once a second. The multimeter may or may not indicate it, since it goes pretty fast.
 
I re-read the SCK at more fine grain setting on the multimeter and I do see if toggling at a very low voltage (.01-.02).

I also re-read everything on a fine grain setting:
-MIS = sometimes floating around .04, sometimes constant .01
-TXD = about .17
-RXD = 3.3
-SCK = toggling around .01
-5V = both reading 5

I've got an Adapt a Damp wired via CAT5 - haven't tested it yet and its not connected.
 
Last edited:
Those all seem right, I can't imagine what could happen to have it do something terrible to the Pi. Is that with a pit probe plugged in / with it trying to run the blower? I'm at a loss as to what could happen, and that soldering all looks good apart from some long tails as you've mentioned. If you want to mail it to me to have a look at it and see if I can figure it out, I'd be happy to hook it up to the scope and look for anything unusual.
 
Those all seem right, I can't imagine what could happen to have it do something terrible to the Pi. Is that with a pit probe plugged in / with it trying to run the blower? I'm at a loss as to what could happen, and that soldering all looks good apart from some long tails as you've mentioned. If you want to mail it to me to have a look at it and see if I can figure it out, I'd be happy to hook it up to the scope and look for anything unusual.

The readings above are from just the baseboard only on 12V.

I'm fairly certain the first Pi I matted incorrectly when not paying attention which fried it via 12V. That Pi gets quite warm on the main chip.

What's weird about the second Pi - it worked by powering the whole HM via the USB. Next day it didn't work on either USB or 12V - it doesn't get warm/hot or make any weird noises like the other one did. Almost wondering if this second Pi just happened to die.

Let me try another Pi and then mail it to you if that doesn't work (thanks for the offer thats amazing!). Lucky enough I have a Microcenter so Zeros are cheap to come by.
 
... Almost wondering if this second Pi just happened to die...

Or maybe the HM file system or SD card got screwed up and no longer boots? I would try a new card freshly imaged.
IDK if the zero w has any indicator LEDS? Power/Drive activity?
You could also try plugging the pi into a USB on a computer and see if it is recognized? It should try to install new hardware BCM2708 or something pretty close to that. If you have plugged the pi into this computer previously you may need to watch the device manager to see if it is recognized (new hardware prompt only shows the first time you plug in) I wouldn't do this with the pi you smoked though, just the one that is unresponsive and doesn't get hot.
 
Last edited:
I thought the same thing.

Installed the image on a new SD card and got no activity lights. Nothing booted.

Then as you said, I tired connecting the Pi via the USB to my computer (non power port) and the BCM device didn't show up.
 
Sorry you have no joy with your pi... If you solder in the headers yourself I suggest you only solder in the portion that is required for the HM (13 pairs starting on the SD slot end) This makes it harder to mate the pi improperly.
 
You could also try plugging the pi into a USB on a computer and see if it is recognized? It should try to install new hardware BCM2708 or something pretty close to that.
Wait, what? I was prepared to learn something new this morning since I'd never tried to plug a Pi into a computer since I assumed nothing would happen. I just tried two different models, a Zero and a Zero W and neither even made the windows Device Manager flicker (even back-powering it through the USB host port on the Pi). Are you sure this is the expected behavior? I know other SoCs do this, but I do not think the Pi does.

The LED indicator on the Zeroes is somewhat weird in that it takes almost 2 seconds before it lights at all. If the SD card has the wrong image on it, it blinks twice and turns off. If it can't read the SD card properly or if no SD is inserted, it won't light up at all which makes it look dead. In both of those "no LED" cases, it pulls about 41mA on the 5V input. In the case of using the wrong image it pulls about 110mA.
 
I plugged my pi zero-w into my computer (windows 10 PRO), the first time the "installing new hardware" balloon popped up, after that it showed on the device manager when plugged in. Here is a screen shot of the device manager showing the entry for the pi zero-w. (BCM2708 Boot)

PiZero_Win10.JPG


I plugged in two different Pi Zero's and they both showed up the same way. I also plugged in a Model B and B+ but didn't see anything happen in the device manager. I found this information (indicating a pi zero-w would show up as new hardware in windows) while searching for ways to tell if a pi zero-w is dead.

The LED's did light on the B and B+ but I did not see any LED's on the Zero W. Doing some research I find the LED's on the Zero W are a bit wonky, didn't work properly on the first release version but behaved better with more recent release versions. (not sure if this is due to hardware or firmware differences) No matter how long I left the zero-w plugged into the PC the LED's never came on (with no SD card installed). When I plugged the zero-w in with an SD card installed with a working copy of linkmeter one LED near the USB port did flash for a while, then went out. Ironically, after I plugged in the zero-w with the SD card attached it is no longer showing up in the device manager when plugged into the USB port near the edge of the board (the one I used while the SD card was attached), however, when I plug in the zero-w using the USB port toward the middle of the board (without SD card attached) it does show up on the Device Manager.... Goofy...
 
Last edited:

 

Back
Top