Circular Saw Centrifuge




About: Build.Share.Destroy.Repeat. Follow me and try a few of my projects for yourself!

Start spinning more food in your kitchen than just salad. Bring the danger of the workshop into the kitchen with a homemade centrifuge! By using an electric circular saw as the motor, I was able to spin food up to 1800 G's and achieve separation in liquids, all for under $20. This small capacity centrifuge gives a taste of what a centrifuge can do for your kitchen.

*Do not replicate this project, it is incredibly dangerous!!

How centrifuges work:
A centrifuge is a rotational platform around a fixed axis. Using the sedimentation principle the centripetal force causes separation of substances with different densities in a liquid.

This rotational force is measured in G's, where G is the measurement of gravitational acceleration felt as weight. On earth we experience 1G, roller coasters experience anywhere from 2-5 G's, and many fighter pilots can achieve 9G's. Since the amount of G's is inversely proportional to the radius of the rotation, smaller radii can achieve some staggering G's. It's not uncommon for lab centrifuges to reach 10,000G's.

Why centrifuge food?
Great question. Those into molecular gastronomy and your average food nerd love when cuisine gets a modern, technical twist. By using centripetal force you can separate food into layers based on density. In a liquid, this means denser substances like pulp are forced to the bottom while lighter substances like oils float to the top.

This means you can separate pulp from juice that might take hours in a fine-mesh strainer, like carrots or peas. You can even clarify things like broths to make them crystal clear. There's no end to the amount of foods that can be spun and remixed.

This Instructable will walk you through my journey to centrifuge glory. I had success, and failure. I will show you both. Though I was successful with my experiment you would be better off buying an inexpensive one if you are interested in trying this out, both for safety reasons and larger volume.

Here's how I made my centrifuge (or centrifood).

Teacher Notes

Teachers! Did you use this instructable in your classroom?
Add a Teacher Note to share how you incorporated it into your lesson.

Step 1: Supplies + Science


What G's are achievable for this centrifuge?
Based on the variables for my built centrifuge [radius of rotational platform: 2.75" (7cm) and speed of motor: 4900 RPMs], this handy G-Force RPM Calculator works out my centrifuge to about 1879 G. I figure this number is optimistic, as this is not a precision machine. A conservative guess would put this circular saw centrifuge at around 1800G's.

Knowing that the 'slow' end of commercial centrifuges are around 1900G's I believe this is a acceptable entry to explore the lower thresholds of what a centrifuge can accomplish.


With this much power there is a real danger of failure, and the failure will be catastrophic. Considering that even professional centrifuges fail, making your own centrifuge is extremely dangerous. Here's some photos to show what lab centrifuge failures look like: Cornell, Perdue, and MIT.
I provided my failure pictures in Step 10 and Step 14

Don't try this.

Step 2: Remove Circular Saw Guards

The spinning action of this centrifuge is made possible by a circular saw I found at a salvage store, I bought this one for $7. The label reads as a Black & Decker 7 1/2" circular saw with a 4900 RPM motor (B&D model number 7301) Though this is a far cry from commercial kitchen centrifuges, it is adequate for a demonstration of the principle and yield acceptable results.

To allow for greater access to the rotor the guards were removed from the saw. I was able to remove portions by unscrewing, but the bulk had to be sawed off. Some portions of the guard were able to be unscrewed and removed, but the bulk of the guard was cast as part of the circular saw body. The saw was carefully disassembled and the guard was removed with a vertical bandsaw. Once the guard was completely removed the saw was reassembled. I saved all the guard pieces and screws for future mounts to this saw.

After the entire guard and removed the burrs and sharp edges were sanded smooth.

Step 3: Secure Base

Knowing this centrifuge would be spinning very fast and be potentially unstable the entire unit had to be secured to a sturdy base, I used some large scrap wood as a footing. I sketched the profile of the circular saw base onto a large flat 3/4" thick piece of plywood, then cut an opening using a jigsaw. I made the opening slightly smaller for a snug fit. I then built up the base with more 3/4" plywood. These plywood sections were clamped together tightly then screwed in place. After laminating a few layers the circular saw was completely secure.

For some safety redundancy I added steel strapping over the motor.

Step 4: Lower Bowl Enclosure

To prevent any failures from flying off at high speeds I decided to build an enclosure around the spinning assembly. After the saw guards were removed there was a few threaded openings I could use to mount my steel enclosure onto. I found 12" diameter (30cm) inexpensive mixing bowls at my local department store, I bought 2 to form a complete enclosure.

For the bottom enclosure I marked the center and bored a hole larger than the diameter of the circular saw spindle. I saved the guard collar and screws that fit around the saw spindle and used this as a mounting template for the bowl. Using a marker I transferred the opening locations to the mixing bowl and drilled corresponding holes. The mixing bowl could then be fit onto the saw and screwed into place using the guard collar.

Step 5: Engage Trigger for Remote Operation

I wanted to operate this centrifuge from a safe distance. With the circular saw unplugged I zip-tied the trigger down, this would allow me to operate the power by plugging it in from a safe distance and not by pressing the trigger up close.

Step 6: Upper Bowl Enclosure

The upper bowl enclosure has a small opening in the top to allow the spindle to pass and help center the spinning assembly when in operation.

The bowl center was found and a 1/2" opening was drilled. The bowl is thick enough to sustain an impact from my spinning assembly and the test tubes, but would wear from friction from the spinning assembly if the opening wasn't reinforced. I used a stainless steel bushing friction fit into the opening to protect the opening edge.

Later, after a few test runs, I decided to weld this bushing to a thick washer, then weld the entire thing to the bowl for a secure connection.

The top and bottom bowl enclosure was attached with binder clips for easy access.

Step 7: Rotatioanl Platform and Rotor

I started with a prototype for the platform out of plywood. Circular saw blades have a unique opening in the center that allows them to fit onto saws, this shape looks like a circle with the left and right side cut off. Matching the geometry of the circular saw where a blade would be, I made a plate of 5.5" (14cm) with 6 equally spaced indentations where the bottom of the test tubes would rest, then made a raised ring that would act as a block to prevent the test tubes from flying out. An upper assembly was made of 6 equally spaced ellipses to hold the top of the test tubes.

To match the threads inside the circular saw I used 5/16" - 24 (fine) threaded rod. The rod was measured to be long enough to protrude out the top of the top bowl enclose when both halves were closed. The top protrusion would stabilize the assembly when operational.

Step 8: Rotational Platform Assembly

To prevent the threaded rod from climbing its way out of the saw and allowing the rotational platform to fly loose I doubled up on the nuts to lock them in place.

The threaded rod was screwed into the saw and the bottom rotational platform was fit on to the saw like a blade. A large washer was applied and two nuts were secured on top and tightened.

Using a test tube as a guide the top rotational platform was positioned and secured using two nuts on either side along with washers, then tightened.

Step 9: Close Assembly and Start Spinning

After the spinning assembly is secured the loaded test tubes can be loaded in. Weight distribution is critical in a centrifuge, and with a home made version you can easily make a mistake and cause a catastrophic failure.

Step 10: Test Wood Assembly and Realize It Will Not Suffice


There's no mistaking the sound made when your centrifuge fails. Chances are there was one small failure, but with high speeds and force in play this small failure will likely cause everything inside to explode.

Here's a failure from my wood prototype spinning assembly. I had the machine running successfully for about an hour, then switched the test tubes for different liquids and started it back up again when it failed immediately upon restarting.

Step 11: Redesign With Metal Spinning Plates

I remade the spinning plates from 1/8" aluminum, which offered much more structure for the centrifuge and didn't shatter.

With a more sturdy assembly I was confident I could run the centrifuge at speed for long durations without failure...assuming I had proper weight distribution.

Here's an animation of the centrifuge working, unloaded, with the top bowl enclosure removed.


The centrifuge could now be loaded with materials.

Step 12: Running the Centrifuge (aka "the Death of Hearing")

Running this centrifuge is loud. Very loud.

I've never used a large commercial centrifuge, so I'm unsure how it compares. Circular saws are loud to begin with, when it strained under the modifications it was never designed for the sound was deafening.

Imagine a high pitched grinding drone mixed with the sound of metal being churned in a blender. The piercing din seemed to bypass my hearing protection and pierce my brain.

The sound was so loud that there were concerned co-workers coming in from other areas of the building to ensure that the sound of a robot doing breakdancing moves was intentional and things were okay.

I ended up having to tuck the centrifuge away in small room and had to promise only to operate it outside of business hours.

Step 13: Results!

Centrifuge success!

I spun my tests for 15 minutes and was satisfied that density separation had occurred. I tried a variety of liquids, and other soupy ingredients with suspended solids. Shown here are carrot juice, fruit+veg juice, hot sauce, and a green health drink.

The drinks separated beautifully with pulpy mass being forced to the bottom. The hot sauce separated into interesting layers, with the spice mix being the densest layer, followed by a think layer of pulp, then the vinegar solution.

I separated salad dressing, though this was visually appealing it is not that special as salad dressing separates naturally on the shelf. I also blended peas and corn, then spun those, which separated into the coveted 'pea butter' (soft concentrate of pea starch).

Through experimentation I have destroyed almost all my test tubes. I'll be making an update with more centrifuge food results.

Step 14: Final Thoughts

Even with a more robust rotational platform and safety precautions making a homemade centrifuge is dangerous. I had great success with my centrifuge, but there are limitations with what I was able to accomplish.

Aside from being much safer than my creation, commercial centrifuges offer a larger volume for vessels which can be fit into the device. If I were to attempt to scale this design to a larger centrifuge there's be all kinds of balance problems. Commercial Centrifuges have this covered by having a large heavy rotational platform that can accommodate larger volumes of differing density. Commercial centrifuges are able to spin at much higher G's. For comparison the lower spectrum for a decent lab centrifuge is over 10,000 G (mine was a measly 1800G).

Finally, I imagine commercial centrifuges are also a great deal quieter than mine.

Plenty was learned through making a centrifuge. I'm hoping to revisit this concept again but am also happy to put it away for a little while to let some new ideas percolate.

If you have any suggestions I'd love to hear them!

Power Tool Kitchen Contest

Participated in the
Power Tool Kitchen Contest

Full Spectrum Laser Contest

Participated in the
Full Spectrum Laser Contest

Green Electronics Challenge

Participated in the
Green Electronics Challenge

Be the First to Share


    • Assistive Tech Contest

      Assistive Tech Contest
    • Reuse Contest

      Reuse Contest
    • Made with Math Contest

      Made with Math Contest

    31 Discussions


    1 year ago

    Here some attempts to get tonic water (boiled down) with carbon to settle out. It failed since the sugar (was probably a carbon-sugar complex) with the same density as water. Thus the water and acid + methanol even at 1200-1500 RPM for 15 minutes had NO EFFECT. Still learning about it centrifuging Properties. A commercial centrifuge for 120 ml saves me tons of $$ on expensive solvents like Hexane, Dichloromethane, Tetrachloromethane, Ethyl Acetate, and even perhaps Chloroform as well. It saves time and the expensive of recycling solvents through distillation.

    Also, I am doing a project on Limit of Detection of Alkaloids including Nicotine, Yombine, Ephedrine, Dextromorphan Bromide, Pepperdine, which react with Potassium Bismuth Iodide in HCl (dragondroff's Reagent). This is through a serial dilution and Potassium Bismuth Iodide in HCl for color change detection is normally limited to 4-6 mg an L. Centrifuging, Freeze-fracture, boiling and extracting removes lots of these impurities and gives you an effective extract. For HPLC, Mass spec it Limit of Detection (Potassium Bismuth Iodide) is lower near a microgram an L!


    1 year ago

    Yeah a Rotary, if it spins or breaks at 2000 to 4000 RPM, is no LOL matter! It can seriously injure or kill a person. I think if you can get to work for 20 dollars is very cool. I don't have the time or the skills to do that but I know you can get on E-bay for 80-100 dollars the "800 in 1 centrifuge." It goes from 500 RPM to 4000 RPM for 0 to 60 minutes. Here some results for coffee and nutmeg.

    Especially if you make the centrifuge yourself you are bound to have failures so don't take this too personal. Nice Instructable tho!! Note that Chlorphyll unless it in really high concentrations (in methanol and HCl (1 ml conc Hydrochloric acid) doesn't form a pellet that easily even for high speeds.

    1 reply

    Reply 1 year ago

    My project was incredibly dangerous, more a proof of concept (and, I was interested in the results). A commercial unit is a much better idea :)

    Your results look great! Food science is so neat


    5 years ago on Introduction

    AWESOME project! :D Can I ask your opinion about using a brushless motor?

    Dairk Laird

    5 years ago on Introduction

    Cool. When's the gas powered lawn mower engine version coming -lol.


    5 years ago on Step 14

    Get yourself a wall fader switch (the kind used for lights). That way you can turn it on gradually and avoid that initial shock to the system. If you have an RPM meter, you can put some black tape on your assembly, and measure the RPMs then mark it off on your fader switch so you know your RPMs for any given setting on the dial. Cheap, easy improvment.


    5 years ago on Introduction

    The simplest food centrifuge is a lettuce spinner--a commonly used replacement in microbiology labs with low budgets. You don't need incredibly high speeds for a useful centrifuge.

    But I'd still like one capable of dealing with larger volumes for extracting herbals...


    5 years ago on Introduction

    Very nice, etc… etc… But why on earth would someone use a centrifugal force macine when cooking ???

    What we need is how to combine ingredients, ie. blend them, not separate them, ie. "un-blend" them.

    2 replies

    Reply 5 years ago on Introduction

    You can take fat out of non-low-fat salad dressings, or remove water from some food for whatever reason, and there's likely a bunch more you could you. Yes, a way to use a power tool to blend stuff would be cool, but I think this is cool too. It doesn't have to be just for food, either.


    Reply 5 years ago on Introduction

    That it may be useful for other purposes than food I readily admit.

    As for salade dressing, one can make his (her) own so he (she) won't worry about how much fat they put in manufactured dressing. Also home made dressing takes only 2 to 4 minutes to make at the most, and always tastes better than off the shelf stuff. Again taking water out of food is a good reason why to use such a contraption … but I'm afraid it would be difficult to pt salad leaves, or tomatoes or a steak in one of those glass tubes !!!…

    Nicely made all the same.

    Have a good week.


    5 years ago on Introduction

    Glass Cigar Tubes would work too... Just a thought


    5 years ago on Step 12

    Commercial centrifuges are actually rather quiet. My guess is the saw housing provided more noise dampening than it would appear. Or maybe when we use the saw holding it absorbs the vibrations so the motor seems quieter.


    5 years ago on Introduction

    I love your build!! It shows great ingenuity. Having worked on many styles and sizes of "fuges" during my career in the medical repair field, I have learned that it isn't always the imparted G's from rotation speed that does the separation. It is the separation angle and time at speed. That is why most medical fuges use pivoting tube carriers. That way they impart a 90% rotational separation angle. A blood separation fuge only turns at about 5k rpm yet separates 1 unit of blood (495gm) at a crack into plasma and platelets. two very important aspects are load balance and the addition of an upper shaft bearing to diminish wobble. Your cracked tube is the reason medical fuges use tube shields and cushions in the carriers. For non-medical use, a more reliable tube would be a food-grade plastic one. They are much tougher and pose less of a hazard if they do crack or shatter.


    5 years ago on Introduction

    If you used a drill press motor, you would not only be able to reduce the noise level, you would have better control over the specific RPMs.

    The page you linked to shows that the centrifuge has an RPM of 3300. Your

    4900 RPM motor may be just to fast for what you are trying to do.


    5 years ago on Introduction

    FAIL--Sometimes wisdom lets you know just because you CAN doesn't always mean that you SHOULD. The cracked tube. wouldn't have happened if you had started level. else gimbaled the carrier .


    5 years ago on Introduction

    Hey mate, just a little idea to offer to you.

    To increase safety & stability, I'd build an arm with a bearing to goes over the top bowl. That'll soak up imbalanced loads, lengthen the life of the circular saw bearing, and remove a lot of risks.

    Love the idea anyway!


    5 years ago on Introduction

    Now hook it up to a Variac and limit the wattage to the circular saw....and spin it at a lower RPM. Good Job.

    So, after reading this, I got to thinking...

    Would it make this a safer build if there was some sort of kill-switch to shut down the machine if something went out of balance, like a pinball tilt sensor, thereby minimizing collateral/personal damage?

    Or would it be too late at that point with these machines? I'm not very familiar with centrifuges.