Balancing Harbor Freight Bench Grinder Wheels: An Electronics Engineer Way

The video explains how to use MEMS accelerometer to quickly balance desktop grinding wheels (stones) in place, when the wheel is mounted on the grinder. The setup uses oscilloscope to find the angular position and amount of counterweight to compensate vibrations. Note that 𝑭=𝑹∗𝑴∗𝝎^𝟐
The phase of the signal (location of the top of waveform) relative to reference pulse period gives the angle where the compensating mass must be placed, and the amplitude is proportional to the required compensating mass. The feedback and likes are appreciated.
Source of Accelerometer module: www.amazon.com...
Datasheet: www.analog.com...
Using Soundcard as oscilloscope: homediyelectro...
www.zeitnitz.e...
You can find many other software oscilloscope software for PC and Android online. Note that the soundcard should have high input impedance ( greater than 100kOhm). Usually, line input of typical sound card is suitable.

[@billl7551]

Interesting that I just watched another video on this unit that took 5.4 grams to balance. I dont have an oscilliscope but do have a lathe. I made a ring with 12 locations for a 1/4 screw. All roundness and wobble was addressed first. Then, added the balance ring. Using a brass screw - progressively moved the screw to locate max and min locations. Then proceeded to tune to minimal vibration with weight. It does run smooth now, all the imbalance was in the wheel. This is a good reference video for a low cost balance sensor. Bravo!!! Well done.

[@stringlarson1247]

As a fmr. EE turned software engineer; I appreciate this. Esp. the real oscilloscope. What's really needed here is a smart wheel with a WiFi connection and an 'app' for iOS/Android/Web with a cloud-based solution to rebalance the wheel internally via hydro dynamic changes in micro-cylinders. The system will also monitor the number and type of things you grind and order new parts as needed. Or one could purchase a better quality product? Nice touch with the wire cutters as pliers. :)

[@wavedrom]

Time to try it on the larger grinder!

[@celsiusswatt6021]

The phase of sinewave on the screen relative to the reference pulse tells the angular position of the counter-weight. The period is 360 degrees, so if the top of the sinewave is at, say, 3/4 of space between pulses, the angle is 270 degrees from the place where the reflecting sticker is attached. If you can measure the delay with cursors, the angle in degrees is equal to the time (in microseconds) divided by 43 (for the 60Hz motor free running). In other situations, the angle (in degrees) is the time offset from the pulse (in microseconds) divided by distance between pulses (in microseconds) multiplied by 360. I.e., Angle=Toffset/Period*360. The weight may be also calculated from the effect of placing calibrated weight and measuring corresponding change of the amplitude. That allows to calculate weight and position and perform balancing in one step. In fact, similar technique is used in automatic wheel balancers in auto service.

[@ianboard544]

I just bought one of those and have the same issues ( I'm also an EE). I was thinking another way to do this would be to make a stroboscope and run it at a frequency slightly different from the rotational speed of the wheel. The space between the wheel and the rest should 'breathe'. Some simple markings on the side of the wheel ought to tell you where the heavy spot on the wheel is. Your way looks pretty complicated.

[@celsiusswatt6021]

Thanks for watching! Please let me know if your experiment with a stroboscope was successful. I guess that if you want to track the (deflecting due to misbalance) center of rotation with stroboscope, it will be very difficult because the amplitude of displacement is very low, especially near perfect balance. Besides, you need to properly dress the wheel before balancing (but you need to do it anyway). One other alternative is to place the grinder on the scales and observe the AC component of the weight, which is nearly the same as I did but will be less sensitive and needs an amplifier for load cell signal. On the side note, someone suggested that the same method will be useful for balancing the ceiling fans. This case may also work with stroboscopic observation.

[@ianboard544]

​@@celsiusswatt6021 You got me thinking about this. I worked for a company that make ultrasound probes for surgery. We tested them looking through a microscope where the illumination was coming from an LED being pulsed close to the vibrating frequency. I looked like it was waving back and forth really slowly - very cool. I was going to write a program for one of the ARM cortex eval boards I have to pulse an LED using a timer. I was going to just put marks on the side of the wheel and watch the gap between the wheel and the tool rest to get an idea where the imbalance was happening. If I get around to this soon I'll post what I find. I was a bit surprised at how out of balance the grinder was.

[@celsiusswatt6021]

@@ianboard544 Note also that the spacing between the stone and toolrest will not depend on imbalance, because they are fastened rigidly together, and they will vibrate together. The best you will see this way is the non-concentricity of the stone. You need a kind of marker on your steady table and let the grinder wiggle on its flexible rubber feet to see where the imbalance is. The Young's modulus of lateral movement is lower than that for vertical, but you may want to observe vertical vibrations if it is easier... You mentioned a keyword: a microscope. When rotating at normal speed and close to perfect balance, the displacements you need to detect will be

[@ianboard544]

@@celsiusswatt6021 good points. The one ive got moves quite a bit. Im curious what Ill see. I have a project i can base the strobe on so it should be pretty simple.

[@andreyl2836]

Like this video! The only thing I didn't get: Did the circuit and oscope are helping somehow to find the angle position for the weight to put on? If not, it could be quite a challenge...

[@celsiusswatt6021]

Yes it does, see below.

[@andreidioumaev4928]

Unfortunately, the links "below the video" are not clickable

[@LaLaLand.Germany]

Well- could I just send You my grinder with the wheels I´d like to have on? Since my grinder is equally shitty but I am no engineer nor do I have access to a lathe (or know anyone that does) to make those weight thingys. But I shure would like my grinder to run nice, I know they can but I´m not gonna glue on weights to the wheels (as suggested by other videos) You could make a pretty business out of, I guess if it is quick and cheap to do.

[@celsiusswatt6021]

The main purpose and advantage of the method is balancing in place together with the rotor, and I am more than happy with the result. One advice, though: gluing some weights to the side is a chance to be hurt as the glue is not reliable. So putting the weights in the way that they are held radially (as in my video) is a much safer option. However, someone mentioned that I should simply buy more reputable stones and grinders 🙂. So it could not be a business... I bought CBN (Cubic Boron Nitride) Grinding Wheels (they are well factory-balanced) for grinding carbide bits (Note: you cannot use diamond wheels for carbide)

[@roulbook1921]

@@celsiusswatt6021 Did´nt you mix it up? diamond for carbide is good.Cbn for carbide don´t work.

[@rtz549]

The wheels should be better balanced from the factory.