I have been building some Rocket-candy engines for fun, and informally experimenting with different designs or techniques.

Now I'm interested in figuring out a way to compare the performance of the different engines I make. What kind of hobby-level experiments might I do, and what simple metrics might I use to try to decide if one engine is "better" than another, or what "better" would even mean?

I've looked at some videos of engine tests and have been experimenting with trying to use a weight scale (or my own spring+ruler) to measure the thrust, but I'm not sure the best way to record the data, or use it to try to calculate standard things like impulse or rocket acceleration. What about going a step further and trying to predict which rocket might go higher or lift a heavier payload to the same height?

And are there other instruments I should look into buying to do other measurements on this kind of hobby rocket engine?

Where to begin? Thanks!

  • $\begingroup$ There is a little bit of helpful in this link found in this answer but I think a good answer or two will be posed here if you can at least give a little bit of detail on your situation. What resources do you have? Can you build test equipment? Have you looked at anything so far before asking here? Even a few other YouTube videos about sugar rocket testing? Right now it looks a little bit like "please write a book chapter on rocket testing for me". $\endgroup$
    – uhoh
    Commented Jul 21, 2017 at 14:13
  • $\begingroup$ It would be better if you explained some tests you were thinking of doing. Can you think of any connections between velocity and acceleration? $F=ma$? The Tsiolkovsky rocket equation perhaps? People will be more likely to help if you describe some efforts you've made yourself first! $\endgroup$
    – uhoh
    Commented Jul 21, 2017 at 14:17
  • $\begingroup$ @uhoh I have look at some youtube videos and I've seen 2 ways to calculate thrust, one is the weight scale and a spring and ruler contraption. And yes I can build test equipment. I just want to make a multitude of R-Rockets and find the one with the highest average in all the stats. Aka the best possible rocket i can build $\endgroup$
    – jack
    Commented Jul 21, 2017 at 14:21
  • 1
    $\begingroup$ That's great! This is exactly the kind of thing that you should include directly in your question. In stackexchange comments are good for clearing up minor points, but anything central to your question should be added up there. I've made some edits to your question based on your comment. You can edit further, or if you don't like it at all, click on the word "edited" to the left of your name and icon, and when a list of edits shows up, look for roll-backnext to an earlier version you prefer and click that. $\endgroup$
    – uhoh
    Commented Jul 21, 2017 at 17:21

1 Answer 1


First, I'd like to say props on pursuing your curiosity. Please make sure you're trying to adhere to some safety guidelines while pursuing this - use your best judgment.

Now, it sounds like you're willing to do some learning/already have some engineering hobby experience. I will be assuming that you're willing to make some purchases for this.

Experiment: This experiment builds off your spring-ruler idea. It can be tedious to measure the spring constant of the ruler and the subsequent deflections due to the rocket thrust. My suggestion is that you pursue the cantilever beam strain gauge experiment with slight modifications. enter image description here Image Source: Chegg

Above you see the general setup of the experiment. The strain gauge provides you an instrument for measuring the deflections by correlating beam deflection, induced strain and the force experienced by the beam. In the image above, you see a force vector applied to the end. This is where you will modify the beam setup so you can install your rocket motors to this end, allowing you to measure thrust (force) by measuring the strain and subsequent deflection of the cantilever beam.

Attached to the strain gauge will be some type of data acquisition device (i.e. DAQ, microcontroller) that will read and transfer your data to your computer. This part gets a little difficult because you may need to tune your controller to be customized to the beams material properties, dimensions etc. I suggest looking at some basic mechanics of materials topics to help with this portion.

During the data phase, you will only be able to measure voltages so you will need to dive in depth into the workings of a strain gauge. From this experiment, you'll be able to get thrust, burn rate and other performance parameters - most importantly specific impulse. You may use the linked performance parameters to determine which rocket motor is better. Of course, in a 'real-life' scenario the most powerful engine may not be universally desired, however, generally speaking, the most efficient is. This is a very on the surface explanation of the entire experiment but I think it provides the best path for what you would like to learn

Required Materials:

  • Strain gauge (Wheatstone Bridge)
  • Metal Beam (metal ruler)
  • Data Acquisition Device (DAQ)
  • Amplifier (for mV signals from strain gauge)
  • List item

Required Experience:

  • Programming
  • Basic understanding of electronics (i.e. strain gauges, microcontrollers)

As far as predicting which motor will go further, you can make that deduction from the experiment listed above, particularly from the performance criteria. Hope this helps!

  • 1
    $\begingroup$ You can save yourself a lot of headache in the hardware setup if you just go straight to a load cell, which replaces the need for a strain gauge and beam with a compact, calibrated package. Load cells exist with a variety of precisions, ranges, and price points, so I would advise going down that route instead of introducing the additional pain of "rolling your own," so to speak, with the beam and strain gauge. $\endgroup$
    – Tristan
    Commented Jul 26, 2017 at 15:24
  • $\begingroup$ @Tristan and Inti So in other words weight the force being produced? $\endgroup$
    – jack
    Commented Jul 27, 2017 at 3:33
  • $\begingroup$ @jack Yes. Essentially, the experiment is a weight(force) scale. $\endgroup$
    – Inti
    Commented Jul 27, 2017 at 3:35
  • $\begingroup$ so instead of going through this more "expensive" and harder route I can buy a scale and record the weight change? a $\endgroup$
    – jack
    Commented Jul 27, 2017 at 13:08
  • $\begingroup$ @jack well, the difficulty with that is that it will affect the quality of data adversely if you're collecting data by hand. With a load cell or strain gauge you can collect the data electronically at a specific frequency - 1000 Hz (1000 time points per second).Unless you're referring to an electronic scale that lets you automatically feed and export your weight data directly to your computer. $\endgroup$
    – Inti
    Commented Jul 27, 2017 at 13:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.