I wanted to test my understanding of the Load Cell / Wheatstone Amplifier Shield (2ch).
I provided 3.3V and GND to the input pins labelled accordingly and saw the LEDON light up. I then connected a power supply to the two middle pins of strain 1 and applied 5mV. I check the voltages with my fluke multimeter and get 1.6V output on Analog In 0.
I have tried varying the input voltage from .005-.1 volts. I had a technician swap out the gain resistor to a 499 Ohm for a gain of 100. I cannot get the output pin to vary its voltage with respect to the input voltage applied.
Please advise on ideas on what I could be doing wrong. Is there a circuit diagram for this shield anywhere?
Respectfully,
David
Hi David,
Can you attach some pictures of your testing setup ?
The two center pins should be the ones for the input according to the user manual. Make sure you are properly grounded on the same potential for all the equipment.
These are just my observations and what I “figured out”. Please do not take them as information from the manufacturer or that it is valid- it is just my observations. I wanted to use the system to avoid having to layout a pcb since this system would do exactly what I wanted.
I have the system working. Here are some take aways:
*the manual states that the 5V, 3.3V and GND pins are used to power the shield. This is misleading since the shield is only powered from the 3.3V pin, which took me a while to figure out. Note: you can apply 5v to the 3.3 pin and get the system to work except for the LED since all IC’s can handle that input.
*The manual states correctly the pinout as one can see by the square PCB pad; however, the “https://www.robotshop.com/blog/en/interfacing-a-load-cell-with-an-arduino-board-16247” is misleading. They show a black wire going into pin 1 and a red wire going to pin 4. In their setup it does not matter since the actual color of the wires is irrelevant to the system and polarity is irrelevant to a resistive network, but it is opposite of standard nomenclature of GND being black and power being red. As an aside, I am used to the pins going from 1 to increasing from left to right, but this is just common and not always. This problem was the biggest one for me as the picture stuck in my head. I was testing the amplifier system on the board and was directly inputting small voltages on the input and measuring the output, which means I had my S+ and S- incorrect because of my mistake on the pinout.
Note: I assume most people who use this board are hobbyist and would not see this problem if they directly connected the board into the arduino and just hook it up to a bridge.
Unsure if further development will occur with this board but one minor suggestion is to use a multi-turn potentiometer on the ref pot to make it easier to dial in the exact voltage that one needs since mV make a big difference in this setup. I would also consider taking down the picture on the blog and replacing it with one that shows the standard color configuration. If you ever make a alteration to the silkscreen I would be nice to people and put the number 1 next to the first pin, but not needed. If you wanted to further the use of the board to other applications you could put a small header with either screw in connection or molex specifically for power, and gnd, but I can see how that could be confusing to a hobbyist.
Respectfully,
David
Hi David,
Our main engineer behind the product on vacation. However I’ve sent him a message to follow up with you once he is back.
Thanks for your comments and recommendations.
Hi,
-
Correct, the shield can operate without issues with only the 3.3V and GND connected. The 5V pin on the shield only powers the green led.
Note : since the 3.3V line powers the AD8426 amplifier (can be powered with up to 36V) and the two MCP602 amplifiers (can be powered with up to 6V), you can power the 3.3V line with up to 6V without having any issues on the shield if the load cell you are using requires at least 5V excitation voltage.
-
Also, correct. We will update the pictures on the blog post (with the one attached) to show red wire (E+) on pin 1, green wire (S+) on pin 2, white wire (S-) on pin 3 and black wire (E-) on pin 4. As you have mentioned, the current picture shown on the blog post shouldn’t affect the working of the shield since the polarity is irrelevant to the resistive network of the wheatstone bridge on the load cell but it is not common, therefore we will correct it.
We thank you for your feedback regarding this load cell shield and we will keep your inputs on-file in case we decide to release a new version of this load cell shield. If such a product is designed, we will certainly revisit the silkscreen and add the pinout of the load cell on the strain connectors.
Best Regards
