Faulty SharpIR GP2Y0A51SKOF

I purchased 4 pololu IR sensors of model mentioned above from the local distributor that you have used in Sri Lanka today. 3 of the sensors are working and giving an accurate reading when implemented with an arduino using the library that you have referenced. However, the fourth module does not return accurate readings. I contacted robo.lk ([http://robo.lk]) who is the authorized distributor for Pololu in sri lanka and they mentioned that Pololu does not replace parts which have been opened or used. I soldered a capacitor between the power and the ground pins as instructed by you in the details for the usage of the module. Am I eligible for a replacement ? If so would I be able to get the replacement from robo.lk or will I have to wait for it to be shipped by you? I need the sensors urgently for a robot competition and would appreciate it if you would look into this matter with haste.

Thank You


I am sorry to hear that you are having problems with one of your Sharp distance sensors. Returns of any products you purchased from a distributor should be handled directly through them. Our distributors are independent businesses, and how they handle customer service is their decision. However, we can generally assist with troubleshooting to try to determine what might have gone wrong, especially for products that we make.

In this case, this is not a product we make, but there are still a few things we can look at. In particular, I am wondering if you have tried swapping the sensors around in your setup (both in terms of physical location and connections) to confirm the problem really is with a specific sensor module. Also, are the readings still inaccurate if you reduce your system to just the problem sensor by removing the other ones completely?

- Ben

Well we tested each module with the same setup and didn’t set it up on our robot yet. So I’m pretty sure it’s something wrong with the sensor.

My expectation is that Sharp tests every unit, and we almost never hear of problems with them, so my immediate suspicion would be the capacitor you added is not doing what you expect. Have you carefully inspected the solder joints or tried reworking them? Can you quantify how the readings are inaccurate?

- Ben

The readings are always off by a centimeter. For example, if I position a white board at a distance of 30mm from the sensor, I get an output of 40mm and this is true for all readings; the offset is always +10mm . Due to this constant offset, we can use this sensor by adjusting the code we write. However, I fear that the sensor might start malfunctioning during the task. Also, I measured the resistance between all the pins of the faulty sharp IR and all the units which are functioning properly. The resistor values are quite similar for the ones which are properly working, but different for the faulty one. I guess it’s something to do with a voltage divider ?

What you are describing sounds like it is well within the normal unit-to-unit variation of these sensors. The sensors just output an analog voltage, and you typically need to calibrate each sensor to figure out how to convert that voltage into an absolute distance if that is important for your application. For example, you can see from the datasheet that the voltage corresponding to a distance of 15 cm can be anywhere from 0.25 mV to 0.55 mV, and the voltage difference between 2 cm and 15 cm can be anywhere from 1.35 V to 1.95 V.

- Ben

This was the issue. I calibrated all four sensors using a custom function. Thank you for your help.

1 Like