Recording what you did

Every experimental record should report what actually happened while you were in the lab. It is really important to record exactly how the data you measure and was obtained because, if after the experiment has been conducted, the results are interpreted and something unexpected, interesting or valuable is discovered, it is critical to know if this has occurred due to how the experiment was conducted.

For example, say you are considering purchasing a battery from a supplier that you want to use in your new range of products. You test the battery and it shows that the performance isn’t what the manufacturer claims. The manufactures respond saying the battery must have not been fully charged when the test was performed. Unless you have recorded the depth of discharge before, during and after the experience, you have no way of knowing if the supplier is lying to you or your test was faulty. Other factors that could influence the results you obtain during an experiment are the pertinent surrounding environmental conditions, such as room temperature or pressure, the pieces of equipment you are using or the procedure you choose to adopt. You should consider adding all of these into your experimental record. The order in which you conduct sequential tasks, tests or measurements should be clear from your record. For example, experiments that involve hysteresis may give you a different answer is you different answer depending on if you are increasing or decreasing your independent variable.

The information about what you did and how you did it usually goes into a “procedure” section of an experimental record. As with all information in an experimental record, the level of detail in the procedure should be sufficient to repeat or recreate the experiment, without anything unnecessary.

Referencing existing procedures

When you execute an experiment, you should have planned in advance, as far as possible, what you intend to do, and written this down in an experimental plan. There is little value in copying out something that has previously been written down in another location. So the procedure section of the record only need make reference to any previously written experimental plans and, critically, any deviations from them that may have occurred during the experiment. As well, experimentalists will often use testing equipment that has been specifically designed to perform a certain task, for example an oscilloscope can be used to measure the frequency of a signal. This equipment may come with existing documented manuals, protocol or procedure for performing certain tasks, often referred to as a “standard operating procedure” or "SOP". For an experiment that is part of a University course, these may be provided in a lab sheet. In the experimental record, reference should be made to SOPs or other standard documents, rather than copying them out.

There will also be aspects of a procedure that cannot be accounted for in the plan a priori [1] and decided while the experiment is ongoing. In this case, those decisions should be recorded in the procedure section of the record, preferably with a justification as to why the decisions were made. For example, say your experiment is designed to determine the optimum input conditions, to give maximum efficiency of a particular device. You don’t know in advance what that optimum input value will be. Your plan says to take a range of initial readings to work out roughly where the peak efficiency is, and the subsequent readings based on that information. The value of the subsequent readings will need to be recorded, as they won’t be part of the plan.

Equipment

The equipment used in an experiment can have a significant impact on the results obtained, so therefore it is critically important that this information is recorded, either in an experimental plan or in the procedure section. The equipment used can be recorded in text or as a diagram, depending on what is most appropriate.

When recording items of equipment as a written list, it is important to include specific details particular pieces of equipment. For example your experimental plan might say you intend to use an Armfield C15 Wind tunnel in the Diamond’s Fluids Engineering lab. This is good, because it includes the module number and you can make reference to the manuals to provide details on this equipment. However, you may not know until you arrive which one of the Armfield C15 you will be assigned when you arrive. If you obtain an interesting results you need to be able to identify if it was truly interesting or due to a malfunction with a particular piece of equipment. In your record should have a mechanism to track back to the unique item. This can normally be done with a serial number provided by the manufacturer or a time number from the operator, for example “Wind Tunnel 4”.

You should also record the precision of any instrumentation you are using. This will be critical for later analysis to understand and process what errors may have occurred in your results. For example, if you are using a ruler to measure a length, specify the smallest number of division or if you are using a piece of digital instruments, record the number of decimal places it is capable of recording to, the published error (which should be in the manual) and record you have checked it has been (or for that matter, has not been) appropriately serviced and calibrated.

A diagram of the equipment can be very useful to record how your experiment operates or fits together. This is because sometimes it is easier to convey information visually rather than in text. For example, if you want to describe the location of specific items of instrumentation, show how water flows around the rig or how the components were wired together, a diagram would be the ideal medium. It is almost always preferable to have a 2D schematic sketch rather than a photograph or rendered 3D image to record the important aspects of the experimental set up.

When using a diagram, it is critical that it is suitably labeled so that parts of the diagram can be identified. Once you have used a label in the diagram, this label should be used consistently throughout the experimental record. For example, if there are a number of pressure gauges in your experimental set up, you may choose to label these up as P1, P2, P3...etc. In this case, a diagram would be ideal for illustrating where those gauges are, including dimension if appropriate. The same naming convention of P1, P2, P3...etc, should then be used when you record the data, so the reading from each gauge can be tracked back to the location where it was taken.