Defective Teensy 3.2's ?

Status
Not open for further replies.
S

skillHoarder

Member
Hi all,

I've been using Teensies for a couple years no and have had zero issues with the dozens of boards I've bought for various projects until just now.

Here's the issue:
I ordered 8 boards from Amazon (fulfilled by Sports Internet Solutions - a pjrc approved reseller, so I don't think they're counterfeit).

EDIT: To clarify, several of these boards would not boot up with their factory preloaded blinky sketch just after being taken out of their packaging... The others had various sketches loaded to them, or attempted to load to them before failing. There is no specific code or wiring diagram I can share...

When plugging in these boards for the first time, right off the bat 4 of them wouldn't establish USB connections or power on with blinky running...
1 of the others took a couple tries plugging and unplugging USB before it would boot up with blinky. It loaded it's sketch ok, but then didn't reboot.
The remaining boards worked ok for a little while and then died.

When I say died, I mean they don't power on. Hooking them up through a USB tester, I can see that there is 0 voltage running through them when plugged in. This can also be seen with a multimeter on vcc and gnd. Well, they actually read between 3mv-80mv... Still bad.

I should point out that I do have some good Teensy 3.2s running perfectly using the same USB port, cable, tester, etc which I bought directly from PJRC a little while ago. None of these seem affected and will load sketches and run beautifully all day.

Luckily, Sports Internet Solutions accepted them all back as defective. I ordered 8 more through the pjrc Amazon store, and this time fulfillment was through Karlsson Robotics (also a pjrc approved reseller). Some of these boards have been working fine for the past couple days, but 2 of these new ones just failed today in the same way...

I guess my questions are:

1) Has anyone else had issues like these with their Teensies? As I said, I've used dozens of other teensies in the past couple years with zero issues like this, so I'm pretty surprised...

2) Could Amazon be supplying faulty or counterfeit boards? I think I've mostly ordered direct from PJRC website in the past. Is the PJRC site using the same manufacturer as the amazon sellers?

3) Any ideas what's actually wrong with these boards? Maybe a batch got shipped from the CM with defective voltage regulators or something?

So confused...
 
Last edited:
without code or wiring diagrams we cant diagnose your issue
we cant guess what killed your teensies, simply telling us plugging it in and loading blink to it is not enough
nobody has nothing connected to a teensy for no reason for a thread without even mentioning anything about connections or code involved

"my car stopped working, what could it be?"
 
Paul or Robin will probably have feedback on this . . .

I see Amazon Sports I.S. units for $24.45 without pins? If you have others from PJRC you should be able to see they are identical? They should run right first time and every time out of the package.

Karlsson Robotics sells direct at PJRC price

My first came from SparkFun to round up to free shipping - the rest are PJRC or OSH Park supplied Purple but PJRC certified and perhaps discounted.
 
@tonton81
I should have clarified that several of them had these issues directly out of the packaging and wouldn't even boot up with their factory preloaded blinky programs. The ones that did work for a time each had different sketches loaded onto them, so there is no specific code to share. The failures mostly occurred before wiring them up to anything.

Perhaps my questions could be better summarized as "has anyone else seen these out of the box failures with their teensies, and might it be supplier specific?"

@defragster
I've looked them over for differences but don't really see anything other than slight differences in the solder masking and slightly lighter color/opaqueness of the PCB. These characteristics could just be batch to batch variation though.

Like I said, I've bought dozens of these, and this is the first time I've had any issues. Weird that so many in a row are having this issue. Seems like either they are from a bad batch, or I'm doing something to them. Not sure what I can be doing that would make them die immediately upon being plugged in for the first time though... All while having older 3.2s that are still working perfectly on the same USB port with the same sketches...

I just ordered another batch of 8 directly from the pjrc site though. Hopefully that helps determine something.
 
From previous discussion, I believe PJRC load up the blink sketches by hand as the last step of manufacturing (and of course, in doing so, they test it).
Possibly some mixed in returns or similar...

You say "Hooking them up through a USB tester, I can see that there is 0 voltage running through them when plugged in. ". You should have five volts being kicked out by your USB ports with no teensy or anything plugged in - is that the case? And the voltage on your 'tester' drops to 0v when the teensy is connected? Please ensure you're correctly referring to voltage and not current. :)
 
Indeed @Cosford - Paul has explained in KS and elsewhere that every Teensy is Jig tested for full functionality then removed and programming over USB leaving a Blinky Sketch ( non Serial USB aware ) to see it blink before it is called good. {edited per post 7 }

@skillHoarder : Did you get 'Welcome to Teensy' cards with each one? ( though when ones from OSH PARK seemed to come one card per order ).

Paul and Robin again can comment on the Silk Screen and finish - I'd expect all to be identical if they are PJRC supplied as I've not heard they got any rework done.
 
Last edited:
Every brand new genuine Teensy is supposed to blink when you plug it in. We test every board, first on the bed of nails which verifies all the pins actually work, and then by plugging in a USB cable and programming the LED blink over the USB.
 
@PaulStoffregen - It looks like the sketch is loaded, but there's an issue with the board powering on properly. The ones that work out of the package blink as expected. The ones that didn't work don't. The ones that worked for a time and then failed had blinky running on first powering up.
Also, is it possible that there are counterfeit boards being sold through your Amazon resellers somehow?

@Cosford - Sorry I was a bit vague about the USB tester too. The tester is one of these inline units that tell you how many volts are on the USB port, how many volts are on D+ and D-, and the ground reference of the USB port. It also does amperage and watt hours, temperature, etc. I'm getting 5v from the USB port as expected. In a working Teensy, I'll see 3v on D+ while it's plugged into the USB cable. In the failed boards, I'll see just a few millivolts.

And as of today, things get weirder. I started filming a quick video to show what the difference is between the working boards and the failed boards. I plugged in one of the failed boards, and as expected, it didn't do anything. I left it plugged in while I went to do something else, and a couple minutes later I hear the computer recognize it as a USB device, then load the last program I had tested on another board (a modified blinky), without me doing anything...
I'll upload that video a bit later, as I can also show the differences in the silkscreen. I'll also try leaving a couple of the other failed boards plugged into USB for a while to see if they are also randomly resurrected...

the tester:
inline USB tester.jpg
 
Ok it gets weirder. I was just able to replicate the "dead" board coming back to life after being plugged into the computer for a few minutes. It not only comes back to life, but it loads the "dead" board with the last sketch that was loaded onto another board... ?? I'm going to try it again with another "dead" board and get it on video this time.
 
And now it's dead again. Just left it plugged in running, and randomly it disconnected from USB (computer gave audio bleep of USB disconnect), and now there's no voltage on D+ again. Other teensies still work fine on the same cable...
 
@Frank B - I thought maybe that too at first, but they also won't power on directly from the Vcc and Gnd pins. And the same USB port and cable on the same computer still works perfectly with other Teensies. For all I can tell it looks like a bad batch of Teensy 3.2's

I've left this one on the USB cable for the past hour or so now, and it randomly reboots, disconnects USB, reconnects USB, etc. My older Teensies on the same setup run fine...
 
Last edited:
Do you have a resitor (some k-ohm) at hand ? You could check if the reset-line is the problem and add a pullup. And perhaps, do the same with the program-pin.
- just as experiment. At least, it will tell us if everything is ok with these signals.
I've had - rare - problems with reset (on a T 3.6)

Then, you can check if the defective Teensys work if you power them with the VIN pin (don't connect USB)
 
There was an article in a german computer magazine: Amazon sent INTEL-cpus instead of AMD cpus - in original AMD-packaging.
A buyer replaced the chips and sent them back... amazon sent them as "new" to the next buyer, who received the old Celerons (AMD packaging) instead of Ryzens..

Maybe a buyer killed some Teensies and sent them back ?
 
Last edited:
skillHoarder said:
Also, is it possible that there are counterfeit boards being sold through your Amazon resellers somehow?
Click to expand...

I sure hope nobody on the official distributor list is selling counterfeits.

Other sellers on Amazon have sold fakes and counterfeits, so I don't recommend Amazon or Ebay unless it's one of the known distributors.

But I would agree with Frank, everything you've described sounds like some sort of very strange USB hardware issue.
 
Status
Not open for further replies.
Back
Top