For my experiment, I chose to test and evaluate a linear position sensor, and identify a possible use for this sensor in every day life. The experiment was done using a variety of apparatus, as seen in the list below, and was set up as shown in the diagram which follows.
Retort stand with two clamps
100g, 20g, and 10g weights with holder
Piece of string
Precautions and Safety
Safety was not a big issue when doing this experiment, as no harmful materials or apparatus were used. However, some precautions were taken. I made sure that I had enough space to carry out my experiment effectively, without worrying about knocking anything over; and when applying the weights to the sensor’s mechanical contact, I made sure to this gently, and not drop the weights on, which could damage the sensor.
After gathering and setting up my apparatus, the circuit was connected using a pair of connecting wires, a few crocodile clips, a power source, multimeter, and of course the position sensor. In this sensor, the exact circuit set up is not known, however there are two possibilities, both of which are shown in the two diagrams which follow.
Throughout the experiment, a constant supply voltage of 5 V was used, so as to maximize accuracy of the readings. The multimeter, set in voltmeter mode, was connected across or in parallel with the variable resistor, therefore measuring the output P.D or voltage across the variable resistor, and thus completing the potential divider circuit.
After this I began to carry out the experiment. The output P.D with no weights added was first recorded; and then 50g at a time, the other weights were added, and the corresponding output P.D readings recorded. After numerous repetitions of this procedure (to minimize error), the process was repeated, but this time in reverse order. The output P.D was first recorded for the maximum weight used in the previous procedure, and then 50g at a time, weights were removed and the corresponding output P.D readings were recorded. This process was repeated a few times to minimize error.
Each time weights were applied to the sensor’s mechanical contact, I made sure not to force them down, but allowed them to rest freely on the contact, with the string drawn tight enough to hold the weights more or less straight upright, without restricting them in any way. This ensured that the conditions for the readings to be taken remained as close to constant as possible throughout the experiment, and so helping to improve accuracy.
From the results you can clearly see that increasing the amount of mass applied to the mechanical contact, that is, the force applied to the mechanical contact, decreases the output P.D or voltage. As the mechanical contact experiences a force, it is pushed inwards, which consequently reduces the resistance of the sensor’s variable resistor. This therefore, in accordance with V= IR, produces a smaller output P.D or voltage. This type of circuit is known as a potential divider circuit, where the sensor’s resistance changes in response to the environment, (in this case an applied force) and so the proportion of the P.D across the sensor changes, which is used to give an output P.D or voltage. From the results we see that the maximum output P.D, is 4.95 V. This then means that at this point, the P.D across the fixed resistor (if first circuit) would be 0.05 V.
During the experiment, some anomalies were experienced. When carrying out the experiment by starting with a maximum weight and decreasing gradually, there was unusually low output P.D readings recorded, for the 950g- 850g masses. This however was identified to be because of the improper balance of the weights on the mechanical contact, which therefore gave inaccurate readings, as the conditions for the experiment changed. In light of this, we can make the assumption that the approximate operating force of the sensor is from 2.0 N to 7.8 N. This proves the relative accuracy of my experiment, as this assumption is very close to that of the manufacturers’, which was said to be 2.0 N to 7.5 N.
Also, after comparing the two sets of results and corresponding graphs, it was noticed that the output P.D readings of the sensor when gradually adding the weights, was higher for each amount of mass than those recorded when gradually decreasing the weights applied to the sensor’s mechanical contact. This could be because of random error; or possibly even systematic error, as an average difference of 0.05 V is seen throughout.
In the graph, we can see that there is a generally smooth curve throughout. It is hard to tell how accurate the results are. Firstly, much effort was put into making sure the weights were balanced on the mechanical contact in the same way throughout the experiment. However there is no guarantee, and is probably quite unlikely that each weight was balanced in the exact way as the one before. Also, due to the fact that the mechanical contact is not 100% efficient, that is, there would be some friction involved, this means that with lighter weights/force, detail or smaller changes are harder to detect, and therefore this explains the sudden jumps in output P.D readings for the lighter weights. An attempt was made to lesson the effect of the friction by the application of some WD 40, which helped to smoothen out the movements of the mechanical contact; however the problem of friction still was, and would be experienced. This then also relates to the sensitivity of the sensor, as gradual minor changes may not be detected as that itself, but rather as a big jump when it reaches a certain point.
From this experiment we can see that using a linear position sensor in a potential divider circuit gives an output P.D or voltage, relative to the changes in resistance, in response to the environment. An increase in the force applied to the sensor’s mechanical contact is associated with a decrease in the resistance of the sensor, and so the output P.D or voltage also decreases.
In light of this experiment, and other background research of this sensor, I believe that a contact linear position sensor would be ideal for automotive applications. This is because there needs to be a constant track of the positions of the engine compartments, especially the movement of the engine’s cylinders and pistons, and this sensor is protected against an engine compartment environment, (such as high temperatures), which can be harsh. It is also cost-effective, with a long endurance life, and can be ‘tailored’ and configured to fit the customer’s needs or specifications.