GUVA-S12SD & ADS1115 accuracy

Post here about your Arduino projects, get help - for Adafruit customers!

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
nirbec
 
Posts: 5
Joined: Mon Jan 23, 2023 6:58 am

GUVA-S12SD & ADS1115 accuracy

Post by nirbec »

Hi,
I have a project that require power measurement of 8x UV LEDs in parallel,
For that i connected 8x GUVA-S12SD (https://www.adafruit.com/product/1918) & 2x ADS1115 (https://www.adafruit.com/product/1085),
While testing the results i noticed the values are very different between sensor to sensor,
ADS1115 configuration is 0.256V range & slowest speed (tested other configuration as well).

It is important to note that values are stable, but different between sensors.

Attached are the digital values i receive.

Could you please advise if this difference comes from the photodiodes? or the ADC?

Something can be done to improve it?

Thank you,
Nir.
results
results
Screenshot 2023-01-23 135233.png (8.3 KiB) Viewed 126 times
Last edited by nirbec on Mon Jan 23, 2023 8:12 am, edited 1 time in total.

User avatar
adafruit_support_bill
 
Posts: 88090
Joined: Sat Feb 07, 2009 10:11 am

Re: GUVA-S12SD & ADS1115 accuracy

Post by adafruit_support_bill »

Not enough information to draw any conclusions. The labels on your data refer to different LEDs. It is possible that the different LEDs have different outputs. To make meaningful comparisons between sensors you need to test them against the same light source under identical conditions with precisely the same alignment.

We also have no details regarding the LEDs or how they are driven. And no details of the physical aspects of your test setup - including alignment and ambient lighting conditions.

User avatar
nirbec
 
Posts: 5
Joined: Mon Jan 23, 2023 6:58 am

Re: GUVA-S12SD & ADS1115 accuracy

Post by nirbec »

adafruit_support_bill wrote: Mon Jan 23, 2023 8:11 am Not enough information to draw any conclusions. The labels on your data refer to different LEDs. It is possible that the different LEDs have different outputs. To make meaningful comparisons between sensors you need to test them against the same light source under identical conditions with precisely the same alignment.

We also have no details regarding the LEDs or how they are driven. And no details of the physical aspects of your test setup - including alignment and ambient lighting conditions.
Attached are images of the mechanical connection & setup,
I tried also going over with the same LED on all PDs, the lower values PD just can't reach the values of the higher ones.

Edit: when i switch the PD 'signal' pin between the ADC inputs, value stays the same (on different channel of course), meaning that the difference probably comes from the PD.

The LED driver is very high end used in industrial machines for years.
1.jpg
1.jpg (223.5 KiB) Viewed 115 times
2.jpg
2.jpg (120.16 KiB) Viewed 115 times

User avatar
nirbec
 
Posts: 5
Joined: Mon Jan 23, 2023 6:58 am

Re: GUVA-S12SD & ADS1115 accuracy

Post by nirbec »

this is image of the readings when the PDs just sits inside the jig like the picture i attached before,
even without LEDs the difference is high.
ambient.png
ambient.png (8.96 KiB) Viewed 99 times

User avatar
adafruit_support_bill
 
Posts: 88090
Joined: Sat Feb 07, 2009 10:11 am

Re: GUVA-S12SD & ADS1115 accuracy

Post by adafruit_support_bill »

Attached are images of the mechanical connection & setup,
The position of the sensor relative to the LEDs is not clear. It looks like there is some shielding to block spill from adjacent LEDs. But distance and alignment are still important. Both emitter and detector are planar devices so the emission and sensitivity patterns will be approximately Lambertian. As a result, alignment becomes more critical at closer distances.
I tried also going over with the same LED on all PDs, the lower values PD just can't reach the values of the higher ones.
How much variation are you seeing in that test? Please post the readings.
The LED driver is very high end used in industrial machines for years.
Is there a separate driver per LED? Constant voltage or constant current? Does it monitor the temperature of the LED? Does it use PWM?

User avatar
adafruit_support_bill
 
Posts: 88090
Joined: Sat Feb 07, 2009 10:11 am

Re: GUVA-S12SD & ADS1115 accuracy

Post by adafruit_support_bill »

this is image of the readings when the PDs just sits inside the jig like the picture i attached before,
even without LEDs the difference is high.
You picture before did not clearly show how the sensors were positioned in the jig. And I'm not sure what you mean by "even without LEDs". Without a known constant reference, any comparison is meaningless.

User avatar
nirbec
 
Posts: 5
Joined: Mon Jan 23, 2023 6:58 am

Re: GUVA-S12SD & ADS1115 accuracy

Post by nirbec »

adafruit_support_bill wrote: Mon Jan 23, 2023 9:21 am
this is image of the readings when the PDs just sits inside the jig like the picture i attached before,
even without LEDs the difference is high.
You picture before did not clearly show how the sensors were positioned in the jig. And I'm not sure what you mean by "even without LEDs". Without a known constant reference, any comparison is meaningless.
Hi bill,
Thank you for your replies, i will prepare a dedicated fixture to test individual PDs with the same light source, distance, angle etc..
will update soon

Thank you

User avatar
adafruit_support_bill
 
Posts: 88090
Joined: Sat Feb 07, 2009 10:11 am

Re: GUVA-S12SD & ADS1115 accuracy

Post by adafruit_support_bill »

In your second photo, are those the sensors or your LEDs shown in the jig? Looks like it could either be the GVA or typical SMT LED in a 5050 package.

Most through-hole LEDs are more directional than the planar style, so alignment is even more critical.

Locked
Please be positive and constructive with your questions and comments.

Return to “Arduino”