View Full Version : Grinding wheel balancing fixture

08-25-2011, 12:34 AM
I'm looking at making a grinding wheel balancing fixture. The types I've seen use a pair of parallel surfaces on which the arbour is supported and the balance point determined. However I've seen another type that uses a pair of rotating disks on either side on which the arbour can sit. I believe the latter type has the advantage of not requiring precision levelling. Has anyone made the latter type or can provide information from which I could fabricate one?

08-25-2011, 12:37 AM
McGyver did a balancer project some time back.


08-25-2011, 12:24 PM
I'm not sure if you're talking about a normal bench grinder or not, but Oneway makes a very simple balancer that made a world of difference on the grinder I use for sharpening lathe tools.


08-25-2011, 10:38 PM
I'm thinking about both bench and surface grinding wheel. I wonder if a small surface grinding wheel is able to overcome the stiction involved in the bearings of a simple stand such as that above?

The type of stand I would like to find out more about is the bottom one in the page below.


08-25-2011, 10:51 PM
I bought some open (unsealed) skateboard style ball bearings for building a balancer. Oiled not greased. Amazingly low stiction/friction. Definately let you balance a wheel easily.

Dang, I just noticed those stay with the wheel, iv been wanting a system like that... oh well. Allready gave away my crappy grinder.

08-25-2011, 11:08 PM
One of the problems of course is being able to perfectly center the wheel on the balancing shaft, then subsequently mounting the wheel to a grinder while keeping that same exact centering. And before you mount the wheel to the grinder, does the grinder shaft run perfectly true?

There are always going to be errors here, and those errors can easily be shown with the simple shaft-on-rails as the first picture shows.

All I'm saying is that even if you took every precaution to mount the test shaft well centered to the wheel, then carefully balanced the wheel, you could still find that it's off balance once you mount it. This kind of suggests that there's a practical limit to how close you should try to get it. You don't need anything more elaborate than what the first picture shows.

On the other hand, the model with twin discs each side gives you better control over the thing you're balancing- it stays centered within the machine. This would be important when the weight of whatever you're balancing starts to get up there. Motorcycle wheels, while not particularly unweildly are heavy enough that the disc type balancer is normally used. The additional weight and the degree of unbalance tolerable are larger factors than bearing friction, so the disc type has its place- but if you try to balance anything light you'd be better off with the first type shown. Very important is to have the contacting surfaces free of any particles of dust or anything.

08-25-2011, 11:18 PM
The disk type doesn't require perfect levelling for use. That is its primary advantage. The beam type is simpler however more time consuming to set up.

Dale Lusby
08-26-2011, 12:10 AM
As to the disk type is there anything special about making one? Do the disks spin while it seeks the heavy spot? If so would you use plain bearings or ball bearings or what? I would like to make one or the other and have it on my list of things to do.

08-26-2011, 12:47 AM
As to the disk type is there anything special about making one? Do the disks spin while it seeks the heavy spot? If so would you use plain bearings or ball bearings or what? I would like to make one or the other and have it on my list of things to do.

the disks do spin, And you'd use ball bearings, Idealy unsealed if you can stand it, or metal sealed but oiled filled. (less friction then rubber contact seals)

definately don't use grease filled bearings but oil filled ones, much lower friction. (VXB bearing has unsealed skateboard bearings for like.. free if they where any cheaper.) Mine only uses 1" OD bearings and no 'disks' but seems to work well enough. the idea of the disks is it incresses the torque on the bearings, reducing thier effective friction. (Of course, your disks must be balanced, and mounted on center, and have a smooth rolling surface free of defects.. Bearing races come prehardened/polished and reasonabley balanced. (also small enough to not matter much)

08-26-2011, 01:03 AM
BM do you have one of these bearing type balancers? Did you make it yourself and would you mind posting a picture. In principle they seem quite simple but I have no practical experience with them and how much the bearing stiction and friction will affect their operation.

08-26-2011, 04:28 PM
I used to work for a company that makes electronic balancing equipment, and have often thought about making a simple DIY electronic balancer. The idea is to make a gadget spins the offending object, said gadget being euipped with a vibration sensor, a strobe light (or LED), and a degree wheel that rotates with the work. The signal from the vibration sensor goes into a zero-crossing detector, which emits pulses, which trigger the strobe light, which indicates the angle of the imbalance on the degree wheel. The same vibration signal can be rectified and fed to a meter to indicate the amplitude of the imbalance. When the amplitude is zero, you're done. The electronics should be PDS (pretty darn simple).

Has anyone already done this? Is anyone interested in doing it?

08-26-2011, 05:30 PM
I'm nearly finished with a small soft bearing dynamic balancer and am currently working on the electronics. I have a linear velocity transducer and a MEMS accelerometer for each shaft cradle. I've looked at both maximum voltage and zero crossing detectors - I should say I have electronic engineer friends who have helped me look at the options. The max voltage appears to be easier to implement - the zero crossing detector that I have triggers the LED at each crossing - that is at 180 deg. I haven't found a circuit that will trigger at only one crossing.

Another necessity is a filter to eliminate mechanical and electrical noise. One of my bigger problems is vibration from the drive motor - I can't yet balance it. That will be the first use of the balancer.

All this to say that I am interested in working on this project. I can build and test circuits but am not a designer.



08-26-2011, 07:21 PM
Millhand: I think you want an edge detector to turn the output of a compairitor into a single pulse once per revolution.

Zero cross is easyer to detect accurately then peak, because there is much higher slew rate (assuming a sinewave signal) at the zero cross, while its nearly 0 at the peak (Any noise/varation in supply voltage will give false peak triggerings)

Tanto: its just two bearings side by side with basicly no gap inbeween, bolted to an L bracket I made (Two of these made)

I clamp em to a table when I need to use em.

08-26-2011, 11:29 PM
Black Moons:

Thanks, I'll pass the edge detection suggestion to the guys who are helping me. I did a Google and it looks like that is what I need.

Do you have any familiarity with the Retrotek balancing instrumentation package? The local balance guy said I should look for one but he suggested that the current price was in the range of 5K USD. Well that's a non-starter but I'd sure like to know the basic outline of its circuitry.


08-27-2011, 12:24 AM
It is possible to use something as simple as a dual-trace o'scope and a pair of tape recorder heads pulled from a couple of junkers. The heads are placed at 90 from each other on a shaft that holds the object you're balancing. A shaft greater that one foot length will do.

The out of balance object on the shaft will produce a circular Lissajous figure on the o'scope and the maximum deflection will happen at the natural frequency of the shaft and wheel. Attaching a strobe to the Trigger Out will provide orientation of the heavy spot relative to a painted spot on the wheel.

This is all added complexity but not much added expense.

08-27-2011, 04:45 PM
dp: Thanks for the suggestion re recording heads. I regularly see old tape decks at ham swaps and now have a reason to buy one.

I got lost on the trigger signal. Is that an output function from some oscilloscopes? Neither of my dual-trace scopes have that. And how is the trigger level determined?


08-27-2011, 07:10 PM
The type of zero-crossing detector you want is just a comparator referred to one-half the peak-to-peak voltage swing. That way, you get one positive transition when the signal goes positive, and one negative transition when the signal goes negative. You trigger the strobe on one transition or the other. That gives one strobe per revolution.

08-27-2011, 07:13 PM
Obviously you have to avoid picking up vibration from whatever is making the work rotate. One method is to put a fan blade on the arbor and turn it with a jet of compressed air. The first thing you balance would be the fan blade/arbor combination.

08-27-2011, 10:27 PM
Balancing grinding wheels:

Most surface and tool and cutter grinding wheels - particularly "Norton" - of say 7" and less, are very well balanced right out of the box when new and can be used "as is".

Tool and cutter wheels and wheels for pedestal grinders normally found in a HSM shop are not normally fitted with "balancing hubs" (which allow balancing).

They can be dressed by hand or machine as that dressing will normally leave the wheel sufficiently balanced.

Surface grinders normally have a "balancing hub" on which the wheels are mounted, dressed and balanced. The wheel is mounted on the hub which in turn is mounted on the surface grinder spindle.

To achieve an accurate balance of the grinder motor rotating assembly (rotor, hub and wheel) the the rotor assembly is very accurately dynamically balanced. The wheel and balancing hub are separately statically balanced while not mounted on the grinder motor rotor. The complete assembly (motor rotor, wheel and hub) should also be very accurately balanced when run as an assembly.

The balancing process is an iterate one as the wheel is balanced on the hub, the wheel is dressed and hub and wheel assembly is removed and re-checked and re-adjusted for balance. This is repeated until the wheel is dressed ready for use.

These pages should clarify the process:










Here is my wheel balancing set up (shop-made) as well as wheels and balancing hubs for my surface grinder:





I will go through the process later as its very simple and very accurate.

My surface grinder wheels are 7" x 1/2" x 1 1/4". I use them until they are worn down to 6" and then I either use them on the tool and cutter grinder or break them up and put them in the garbage bin.

I have found consistently over time that once I have a balanced and dressed a wheel on a balanced hub that the wheel rarely needed re-balancing during its life on its hub.

08-27-2011, 11:38 PM
Welcome back Tiffie. :D :D :D

08-27-2011, 11:43 PM
Thanks Evan - appreciated.

08-28-2011, 12:10 AM
Try this Google link for more info on surface (or other??) grinding wheels:


08-28-2011, 09:08 AM
yestrday ..i had it in my watched items

should i have bought it though ..
i gave it a miss


all the best.markj

08-28-2011, 10:32 AM
It is possible to use something as simple as a dual-trace o'scope and a pair of tape recorder heads pulled from a couple of junkers. The heads are placed at 90 from each other on a shaft that holds the object you're balancing. A shaft greater that one foot length will do.

Another possibility:

I once made an accelerometer for vibration testing by taking apart a piezo buzzer, the type without internal drive circuitry, and soldering a small brass cylinder to the face of it. I mounted this in a little aluminum housing I made such that the piezo disk was supported by it's edge and the brass weight was hanging free. The piezo was connected to a wire that plugged into the microphone input to my laptop and I read it using a software package called "daqarta" that can show it as a scope, spectrum or spectrogram type display. Daqarta has a freeware version and the full version license is available to non-commercial users for something like 30 bucks. Lots of capability for short money if you have need for something like that. Google it for more info.

Worked surprisingly well for a random first try, though I did not end up experimenting with it too much. If I can put my hands on it, I'll post a few photos. might be a fair starting point for someone.

There is a pretty good thread over on PM about balancing using cheap-tech hardware, with some very informative links, spreadsheets and freebee software:


J Tiers
08-28-2011, 12:14 PM
Tiffie's info shows the process nicely.

As for darryl's questions.......

The wheel is mounted to a hub, which travels with the wheel...... you don't take it off the wheel unless you are OK with re-balancing.....

And the hub is a taper-mount, so once you balance the assembly, it has no reason to change until the hub is removed. You can take the hub and wheel off as an assembly, and when you put it back, if the spindle is ground concentric, there won't be an issue.

08-28-2011, 08:22 PM
Thanks JT.

I will post my shop-made and very satisfactory wheel-balancing set-up and its use shortly.

My wheels and hubs for my surface grinder are, as you say, a complete assembly balanced on the balancing rig and are mounted and dismounted a s whole to and from the grinder rotor/spindle. I can change wheels quite quickly, but more importantly they are re-mounted in a balanced state quite accurately without a need for re-dressing if the wheel is "sharp". But if it is "blunt" a re-dressing does not seem to affect the balance at all from a mainly practical perspective.

A HSM shop sized 7" wheel really does not need balancing - if it is a "Norton" wheel. All that is needed is a dressing to remove any eccentricity at the cutting face (periphery) and you are ready to go.

I am not trying to ignore the OP's proposed electronic balancer at all as it is quite an innovative concept.

My previous remarks about not needind to "balance" wheels on a tool and cutter grinder or a pedestal grindeer on HSM shop type grinders still apply, but having said that an accurate balancing will certainly improve the grinding process on them.

08-28-2011, 10:36 PM
The (re-)alignment of the balancing hub (to which the balanced wheel is attached) to the surface grinder spindle tapered nose is similar in principle to the D1 series lathe chucks (female) D1 to the (male) taper at the spindle nose. The D1 cam-lock location system is very accurate and reliable if used correctly.



As a matter of interest, the only reason I have a balancing hub set up on my grinder is that I'd have been quite satisfied with a "plain" (1 1/4") wheel-mounting spigot on my grinder but my supplier only had balanced-hub mounting on grinder I wanted at the time.

But of course, it only had one hub and of course I needed 3 more - at $200+ a time which made the grinder $600+ dearer at the time. But I think it was worth it.

But as I've said before, the grinder would have done pretty well just as well with no hubs at all (ie with the 1 1/4" wheel bore mounted directly onto the grinder 1 1/4" parallel spigot).