AS7262 6-Channel Visible Light/Color Sensor Breakout, ID 3779, Invoice 3021535 ordered 2/20/23
Cannot get either to work with unmodified AS7262_test code. Always get "could not connect to sensor! Please check your wiring." A multimeter says there is 4.25 V getting to Vin pin; there is 3.3 V on the 3V3 pin, The IND LED blinks. The other LED on the board does not blink but is OK when current limited voltage applied.
I believe there is no problem with the I2C clock and data line as on this same STEMTera breadboard (like a 16 MHz UNO), the 2 AS7341 boards I bought on same invoice do work. The only wiring change required between the two boards is the Vin. The AS7341 board continuously reports (at a different baud than the 7262)
F1 415nm : 0.00
F2 445nm : 0.02
F3 480nm : 0.02
F4 515nm : 0.02
F5 555nm : 0.03
F6 590nm : 0.03
F7 630nm : 0.03
F8 680nm : 0.03
Clear : 0.04
NIR : 0.01
So I think the breadboard works, the I2C lines work, the power works. The AS7262 sensor is an unmarked version. I cannot see any problem with my soldering on the header pins to any of the boards.
AS7262 Problem
Moderators: adafruit_support_bill, adafruit
Please be positive and constructive with your questions and comments.
- robrlstn
- Posts: 24
- Joined: Thu Dec 02, 2010 12:16 am
Re: AS7262 Problem
Today I ran an I2C scanner against all 4 boards, two AS7341 and two AS7262.
Results for the AS7341 scanning:
I2C device found at address 0x39 !
done
0x39 is correct according to the ams AS7341 data sheet page 21
Results for the AS7262:
Scanning...
No I2C devices found Should be 0x49 according to page 18 of the ams AS7262 data sheet
I cannot see any soldering problems on either AS7262 board. With the sensor unmarked, I guess pin 1 is known due to the slight offset of the aperture hole in the sensor.
Neither AS7262 board has the UART select jumpered so the I2C should be active.
The ams AS7262 data sheet says that the LED_IND will flash to indicate programming progress, page 16, and it does continually blink, maybe once per second, faster than the noted 500ms pulses during programming.
Both boards are buffered for 5 volt connections
It seems odd to me that both AS7262 boards do not work so I am assuming I'm still doing something wrong but I cannot determine what.
Results for the AS7341 scanning:
I2C device found at address 0x39 !
done
0x39 is correct according to the ams AS7341 data sheet page 21
Results for the AS7262:
Scanning...
No I2C devices found Should be 0x49 according to page 18 of the ams AS7262 data sheet
I cannot see any soldering problems on either AS7262 board. With the sensor unmarked, I guess pin 1 is known due to the slight offset of the aperture hole in the sensor.
Neither AS7262 board has the UART select jumpered so the I2C should be active.
The ams AS7262 data sheet says that the LED_IND will flash to indicate programming progress, page 16, and it does continually blink, maybe once per second, faster than the noted 500ms pulses during programming.
Both boards are buffered for 5 volt connections
It seems odd to me that both AS7262 boards do not work so I am assuming I'm still doing something wrong but I cannot determine what.
- robrlstn
- Posts: 24
- Joined: Thu Dec 02, 2010 12:16 am
Re: AS7262 Problem
I'm convinced that both of my AS7262 boards are BAD (ID 3779) and both AS7341 boards work, 2 of each purchased.
All 4 boards have been tested using minimal hookup thanks to level shifters on board: +5V, GND, SCL, and SDA.
Tested with 3 different Arduino boards (STEMTera, Nano Every, and UNO R3).
Tested with 2 different computers.
Arduino IDE 2.0.4 and appropriate Libraries installed.
Both Example programs compile and download to the Arduino.
Both AS7341 boards consistently report (basic_counts example) F1-F8, Clear, and NIR values, which change when a flashlight is shown on sensor and moved closer and farther away.
Both AS7262 boards consistently report (AS7262_test) "could not connect to sensor! Please check your wiring."
Power on Vin pin is 4.8V. Power coming out of 3V3 pin is 3.28, suggesting that the Micrel regulator is working. The IND LED always flashes at about a one second rate.
The Nano Every is a ATMEGA 4809 which requires "Register Emulation in the IDE called "None ATMEGA4809"
Attached pictures have a small piece of cardboard over the power LEDs so as to not saturate my iPhone camera
All 4 boards have been tested using minimal hookup thanks to level shifters on board: +5V, GND, SCL, and SDA.
Tested with 3 different Arduino boards (STEMTera, Nano Every, and UNO R3).
Tested with 2 different computers.
Arduino IDE 2.0.4 and appropriate Libraries installed.
Both Example programs compile and download to the Arduino.
Both AS7341 boards consistently report (basic_counts example) F1-F8, Clear, and NIR values, which change when a flashlight is shown on sensor and moved closer and farther away.
Both AS7262 boards consistently report (AS7262_test) "could not connect to sensor! Please check your wiring."
Power on Vin pin is 4.8V. Power coming out of 3V3 pin is 3.28, suggesting that the Micrel regulator is working. The IND LED always flashes at about a one second rate.
The Nano Every is a ATMEGA 4809 which requires "Register Emulation in the IDE called "None ATMEGA4809"
Attached pictures have a small piece of cardboard over the power LEDs so as to not saturate my iPhone camera
- Attachments
-
- AS7262 NanoEveryjpeg.jpg (261.86 KiB) Viewed 408 times
-
- AS7341 NanoEvery.jpeg (248.44 KiB) Viewed 408 times
- adafruit_support_mike
- Posts: 67454
- Joined: Thu Feb 11, 2010 2:51 pm
Re: AS7262 Problem
Your soldering looks fine, but the first few photos above show the SDA and SCL pins shorted to GND. There's no way you'll get them to work like that.
For the later photos showing connections to various development boards, the connections look reasonable. As a sanity check, do you see the same results when you use pins A4 and A5 on the Uno instead of the dedicated SDA/SCL pins near the USB jack?
For the later photos showing connections to various development boards, the connections look reasonable. As a sanity check, do you see the same results when you use pins A4 and A5 on the Uno instead of the dedicated SDA/SCL pins near the USB jack?
- robrlstn
- Posts: 24
- Joined: Thu Dec 02, 2010 12:16 am
Re: AS7262 Problem
Mike
My wiring is correct. The SDA and SCL lines ARE NOT shorted to GND.
I'm sorry I included pictures; I thought they would show how I do careful, systematic, color-coded wiring:
red +5 volts
black GND
yellow SCL
blue SDA
With all 3 different Arduino boards I tried 2 AS7341 and 2 AS7262 boards. The only required wiring change was the +5. My wiring worked every time for both AS7341 boards and never for either AS7262 boards.
And on 2 different computers. All 12 tests got color data back from the AS7341 boards and nothing from the AS7262 boards.
And an Arduino I2C scanner always saw the AS7341 but never the AS7262.
My conclusion:
Either both AS7262 boards are BAD OR there is something wrong with the AS7262 library.
My wiring is correct. The SDA and SCL lines ARE NOT shorted to GND.
I'm sorry I included pictures; I thought they would show how I do careful, systematic, color-coded wiring:
red +5 volts
black GND
yellow SCL
blue SDA
With all 3 different Arduino boards I tried 2 AS7341 and 2 AS7262 boards. The only required wiring change was the +5. My wiring worked every time for both AS7341 boards and never for either AS7262 boards.
And on 2 different computers. All 12 tests got color data back from the AS7341 boards and nothing from the AS7262 boards.
And an Arduino I2C scanner always saw the AS7341 but never the AS7262.
My conclusion:
Either both AS7262 boards are BAD OR there is something wrong with the AS7262 library.
- adafruit_support_mike
- Posts: 67454
- Joined: Thu Feb 11, 2010 2:51 pm
Re: AS7262 Problem
The I2C address scan is the most convincing evidence: that gives us a direct comparison between boards that show up and boards that don't.
One last question that might add some more information about the nature of the problem: have you tried doing an I2C scan while both boards are connected to the microcontroller? A short in the AS7262 boards could block the whole bus.
One last question that might add some more information about the nature of the problem: have you tried doing an I2C scan while both boards are connected to the microcontroller? A short in the AS7262 boards could block the whole bus.
- robrlstn
- Posts: 24
- Joined: Thu Dec 02, 2010 12:16 am
Re: AS7262 Problem
I2C results
A. AS7341 by itself:
Scanning...
I2C device found at address 0x39 !
done
B. AS7262 by itself:
Scanning...
No I2C devices found
C. Both boards connected at same time:
Scanning...
I2C device found at address 0x39 !
done
A. AS7341 by itself:
Scanning...
I2C device found at address 0x39 !
done
B. AS7262 by itself:
Scanning...
No I2C devices found
C. Both boards connected at same time:
Scanning...
I2C device found at address 0x39 !
done
- adafruit_support_mike
- Posts: 67454
- Joined: Thu Feb 11, 2010 2:51 pm
Re: AS7262 Problem
Thank you. That confirms the I2C bus is working in general.
Send a note containing a link to this page and your order number to [email protected]. The folks there will get you a replacement.
Send a note containing a link to this page and your order number to [email protected]. The folks there will get you a replacement.
- robrlstn
- Posts: 24
- Joined: Thu Dec 02, 2010 12:16 am
Re: AS7262 Problem
I have received the 2 replacement boards. Thank you.
Unfortunately, one of the replacement boards has the same problem as the original 2: an I2C scanner does not see it.
And fortunately, the second replacement board is happily cranking out color counts on my bench.
Unfortunately, one of the replacement boards has the same problem as the original 2: an I2C scanner does not see it.
And fortunately, the second replacement board is happily cranking out color counts on my bench.
- adafruit_support_mike
- Posts: 67454
- Joined: Thu Feb 11, 2010 2:51 pm
Re: AS7262 Problem
Hmm.. let's take a quick look at the one that doesn't show up to make sure all the connections are good.
Post a photo of that setup please, like the ones above.
Post a photo of that setup please, like the ones above.
- robrlstn
- Posts: 24
- Joined: Thu Dec 02, 2010 12:16 am
Re: AS7262 Problem
Done. Same problem as first two boards: not seen by I2C scanner, the IND LED continuously blinks.
The other replacement board plugged into this same breadboard has been sending counts for several days now.
The other replacement board plugged into this same breadboard has been sending counts for several days now.
- Attachments
-
- IMG_0252.JPG (382.46 KiB) Viewed 344 times
- adafruit_support_mike
- Posts: 67454
- Joined: Thu Feb 11, 2010 2:51 pm
Re: AS7262 Problem
Thanks. Again, there's no obvious visual sign of a problem, but the comparison with another working sensor is pretty solid.
Send another note containing a link to this page and your order number to [email protected]. The folks there will get you a replacement.
Send another note containing a link to this page and your order number to [email protected]. The folks there will get you a replacement.
- robrlstn
- Posts: 24
- Joined: Thu Dec 02, 2010 12:16 am
Re: AS7262 Problem
Thank you mike.
BUT, does this not suggest that there is a problem with more stock of this board at the company?
Out of 4 boards, 3 have exactly the same failure symptoms: no recognition with I2C scan and INT LED continuously blinks. Yes, a very small sample but 75% failure rate.
And looks like same problem with 2 more boards posted by chf78: same symptoms. That would make 5 out of 6 boards faulty.
BUT, does this not suggest that there is a problem with more stock of this board at the company?
Out of 4 boards, 3 have exactly the same failure symptoms: no recognition with I2C scan and INT LED continuously blinks. Yes, a very small sample but 75% failure rate.
And looks like same problem with 2 more boards posted by chf78: same symptoms. That would make 5 out of 6 boards faulty.
- adafruit_support_mike
- Posts: 67454
- Joined: Thu Feb 11, 2010 2:51 pm
Re: AS7262 Problem
I've already flagged this thread for review by our QA team.
When we have an issue like this, they pull all existing boards from inventory and don't list the item as available for sale again until they've done full testing.
That's extreme, so we don't do it for every report of an issue. The economics of testing pretty much guarantee that a few marginal devices will last long enough to survive the few minutes of testing done at the fab and in our own processes, but will die within the first few hours of service. The plot of failure rates over time is called the 'bathtub curve' and is a well known thing. That's why our replacement policy is generous out to 30 days, and our replacememt threshold is low for issues reported within the first few days.
This is a different failure pattern, and now I have evidence to justify pulling the AS7262 out of the shop until we figure out exactly what's happening.
When we have an issue like this, they pull all existing boards from inventory and don't list the item as available for sale again until they've done full testing.
That's extreme, so we don't do it for every report of an issue. The economics of testing pretty much guarantee that a few marginal devices will last long enough to survive the few minutes of testing done at the fab and in our own processes, but will die within the first few hours of service. The plot of failure rates over time is called the 'bathtub curve' and is a well known thing. That's why our replacement policy is generous out to 30 days, and our replacememt threshold is low for issues reported within the first few days.
This is a different failure pattern, and now I have evidence to justify pulling the AS7262 out of the shop until we figure out exactly what's happening.
Please be positive and constructive with your questions and comments.