Dynamic balance on props

[Erfolg]

SDF wants the maths, look at this, with particular emphasis on the parallel axis theorem, considering the propeller in two parts, where it will quickly be seen that the forces are unbalanced with differing moments of inertia, deriving from the mass of the material not being identically distributed.

Or this, which is about rotors, far more descriptive, with a few relationships, where you would have to supply your own values.

I have just remembered something from my youth, which was used to convey the effects of inertial forces, I assume at college. Two apparently identical balls were rolled down an incline, one reached the bottom first, after a few demonstrations is was apparent that one always was faster. I seem to remember that the difference was not the mass or diameter, but how the materials were disposed. Indicating that the moments of inertia were in that case the determining factor. Being honest, it is not a vivid memory. But then again neither is Fletchers Trolley, and a multitude of other experiments.

Edited By Erfolg on 20/09/2012 11:04:01

[Myron Beaumont]

Erfolg

Now you've got me ploughing through all the stuff I'd had to do years ago but the difference being that it has a relevance (or doesn't- regarding the dynamic balancing of propellors) I have noted a few relevant points ,having read your links that I think are relevant .It's basic engineering stuff I know but many folks are put of by anything to do with mathematics .

quote-If a rotor has a dia. of more than 7 to 10 times its width,it is usually treated as a single plane rotor.That is for static balancing mentioned in link one . To me I would add "And certainly doesn't need dynamic balancing even more so for part-rotors like propellors where other factors ie airflow ,different angles of attack on each blade ,Side winds, and many more are involved."

This is reinforced at the top of page 8 of your second link .

All good basic stuff Eh!

Edit --Just remembered that when I was servicing BN Islanders at St Just aerodrome ,we regularly sanded down LEs on the props fitted to the Lycomings with no regard to balance .Just to smooth out damage caused by grass strip runway debris & the odd seagull.

Edited By Myron Beaumont on 20/09/2012 12:04:31

[Ultymate]

Some guys flying large scale big props 20-30"+ range balance the hub as well as the blades, but this as far as I see it is still a static balance. Back in the day when I was in the motor trade we had a dynamic wheel balancer that involved spinning the car wheel and having a sensor connected either vertically or horizontally to the supension thus sensing out of balance dynamically in two directions. This meant that not only the tyre and wheel were balanced but also the hub and brake assemlbly which was ok but meant that the wheel had to remain fitted to that hub and in exactly the same position.Any ways here's a link to a video relating to balancing the hub on a prop, it's a few years old and sadly is very americanised.

 

Edited By Ultymate on 20/09/2012 12:30:09

Edited By Ultymate on 20/09/2012 12:31:43

[Erfolg]

I do agree practise does indicate that dynamic balancing is not necessary. That is always with the proviso that the mass is not massively unevenly distributed, as was the case with the AEI demonstration model.

From personal experience, dynamic balancing can matter. In the dim past I had a Triumph Spitfire. In those days it was possible to dynamically balance a wheel on a car, I have not seen the equipment in recent years. Due to the design and control of quality, there was an issue with vibration if the wheels were balanced of the car. The whole drive train to the diff needed balancing, dynamically to overcome the out of balance forces. An example where a thin disc needed dynamically balanced.

When you think of grinding wheels etc. you quickly realise that the density is pretty homogeneous across the whole of the wheel, the issues are more likely due to out of centre bores, which have not been trimmed out.

[SDF]

Erfolg, Interesting paper on balancing rotors. Not really sure what your getting at with the parallel axis theorem though. It just says if you rotate a body about its centre of mass the moment of inertia is a minimum about a given axis. Our staically balanced prop is being rotated about its centre of mass.

Say our staically balanced prop has the mass distributed so there is a lump of mass M at radius -2R and a lump of mass 2M at R with respect to the hub centre, i.e. one lump in each blade. Clearly not an even distribution. The centripetal acceleration at any radius (r) is given by -r*w*w (w should be omega, can't do greek i.e. the angular velocity). The acceleration is linear along the blade in the rotating axis set so the net force (F=ma) at the shaft is 2R*w*w*M - R*w*w*2M = 0. No out of balance. So the radial distribution of mass doesn't matter a jot at fixed RPM and single bladed props and bob weights work fine.

Under angular acceleration (changing RPM) there is an out of balance as the moment of inertia of each of our two lumps of mass is proportional to the square of their respective radii from the hub. There is a net force perpendicular to the axis along the blades as a result of the uneven torques required to achieve the same angular acceleration. Don't know how significant this effect might be but I suspect small on a normal prop.

There could also be a coupled moment perpendicular to the axis of rotation if the masses are not evenly distributed throught the thickness of the blade (principal axis misalignment). The same effect as the car wheel dynamic balance already mentioned. As the blade thickness is small compared to the diameter the misalingment is likely to be very small and I wouldn't expect this to be a significant effect.

I shall look forward to reading the article to seeing what it actually says.

[Biggles' Elder Brother - Moderator]

I wondered when this would come up! Yes SDF you are correct - if we view the prop as simply a thin rotor we can't have dynamic inbalance! But...the prop is not simply a rotor! Its a prop!

In a prop the blades twist out of the plane of rotation, and that twist varies with position along the blade, its much greater near the hub than it is at the tip. So non-uniform mass distribution along the blade could led to a non uniform axial (ie in the direction of the drive shaft) mass distrubution. Think of it this way, suppose the "heavier bit" was near the hub on one blade, but near the tip on the other. Now there would be an out of the plane, axial, mass assymetry between the blades.

What does this do? Well it results in the principal axis of the moment of inertia of the whole prop failing now to align with the axis of rotation - that is dynmic inbalance! The reason why your example of the 2xmass at unit distance and 1xmass at twice the distance is not dynamically unbalanced is because the axis of the moment of inertia still lies along the shaft - no probelm, you are quite right it will be balanced both staically and dynamically.

But once we allow for the fact that mass distribution could be non-symetrical along the shaft axis then we get the possibility of dynamic inbalance.

Now theoretically fasinating as all this is - its a bit like arguing about how many angels can dance on the head of a pin! Because practically it simply "ain't gonna 'appen"

BEB

[Erfolg]

The parallel axis theorem comes from considering each blade as a separate entity with each having an effective differing disposition of mass.

I think the reasons given for a lack of vibration and also for vibration are both correct. The problems arrive where the masses are different. During the acceleration phase you can see a problem, thereafter the vibration is most probably going to continue, as no consistent revs are achieved, due to the vibration. This vibration can easily affect the airframe, making the prediction of what will occur next, as a problematical. You are now into 3 axis vibration, where predictions are great when things have reached an equilibrium, other than the maths, do they actually get there?

Given E=0.5* I *omega^2 then I will have a different value on each blade, so the rotational kinetic energy is different for each blade.

There appears to be a lot of reasons why a something can be dynamically unbalanced and it matters, conversely there appears to be a lot of reasons of why you can get away with it.

It does not appear to matter on bicycle wheels. It has mattered on a machine I once commissioned, requiring an army of experts and many changes to obtain a satisfactory result. Two examples, two different results

Early on I did say that in practise I do not see dynamic balancing much of a issue in the real world of models. Yet surprisingly (for me) I have had a model where after a crash/arrival I suddenly had an issue. I found that the prop was statically out of balance. For the life of me I could not visually see any damage to the prop. I decided that I would balance the prop (a carbon or glass stranded type) by painting the lighter blade. Even after this treatment, the motor still has a slight vibration. I have put this down to a mass distribution imbalance somewhere. I have no proof, although I know the prop balances correctly. I have changed the motor shaft, just in case. All to no avail. Something is not spot on.

So restating my position, in general terms it does not seem to matter if statically balanced. Yet I can envisage there will be times that for some extreme reason it does.

About a year ago, I would have dismissed critical frequency as something that did not matter with model planes. Yet had a motor tear itself out of the front of a model (lightly built), twice while trying to set it up.Yet on a piece of plywood, no vibration at all.

These days I am loath to dismiss some theoretical issues as not being applicable to model aircraft, as I feel I am about to have an irrelevant bit of theory bite my nether regions.

[Martin McIntosh]

Not read all of that lot yet but from practical experience I have found that up to 16 inch APC are fine. Over that I use wood on petrol motors and static balancing is an absolute must. I usually re-sand to a better shape then balance with coats of varnish, no matter what Brian Winch says. Works for me.

Martin

[SDF]

BEB, I see what you mean but the dynamic imbalance is likely to be very small as blade thickness is small compared to the diameter so the scope for misalignment of the pricipal axis is, I would have thought, quite limited. This would probably also show up as a static imbalance with the prop vertical, but I don't suppose may people check the balance of their prop in that orientation.

Erflog, I suspect the vibration would damp out pretty fast without any stimulus but can't be sure. Might be an interesting exercise to try and computer model it, not done any of that for a while.

The kineic energy of the blades may be different but I think that only becomes relevant when you try to change it. Same effect as the angular acceleration I described.

Could your post arrival vibration problem be due to a partial failure in the internal model structure making it more flexible than intended.

I have a control-line twin (same configuration as a F82) with two diesels and on the ground there is a beat frequency oscillation when both motors are running that makes the two fuselages forward of the wing visibly flex from side to side quite alarmingly but there is no problem in the air. Funny old stuff vibration.

[Biggles' Elder Brother - Moderator]

Posted by SDF on 20/09/2012 23:06:26:

BEB, I see what you mean but the dynamic imbalance is likely to be very small as blade thickness is small compared to the diameter so the scope for misalignment of the pricipal axis is, I would have thought, quite limited. This would probably also show up as a static imbalance with the prop vertical, but I don't suppose may people check the balance of their prop in that orientation.

Couldn't agree more - but that's the only way I can see, even in principle, a model prop being in a state of dynamic inbalance - as I said "how many angels" territory! Its theoretically possible - but I just can't see ever being a serious issue unless you bought an absolutely disastrously poorly made prop - and then you'd just bin it!

BEB

[Erfolg]

I know that as engineers we have often/mostly studied a condition as being in equilibrium static, that the forces have remained constant. Yet in the real world, very little is static, particularly in a dynamic system (erm, well, I think you know what I mean). The irony (for me) is that all our lecturers have actually told us this, they have often emphasised the limitations of relationships, the provisos, the assumptions. Above all we were taught how certain relationships have been derived and learnt the proofs. Yet most students (including me) have been more interested in passing the exams, less concerned with what really happens in a non academic system, than passing the next exam.

There is also perhaps issues with naive students such as myself. We frequently have not recognised that the lecturer and lab assistant have constructed an experiment that would clearly show the phenomena. Again not recognising the set up and values were not as typical as you would find in the physical world. Real engineers from collective experience know the norms of what works.

With respect the propeller, I see it as completely non static type situation. Every change in direction causing accelerations on the motor and propeller system, Taking an "A" level type question, what happens to the centripetal forces as a bob weight rotates towards the top of the circle and then moves to the bottom, as we know that in one condition it resists g (little) and the other gains. A propeller also sees these forces, yet the axis is probably seldom horizontal for more than a split second. Are these forces significant in our model system? Particularly when the differences in mass and any non symmetrical disposition are small Our observations are that they are probably not.

We have only considered one component, and only then, one small aspect. The more we look, the more complicated the system can be made to seem. The skill is knowing which things do matter. My experiences suggest most engineers do steer away from the trivial issues, and are taken aback when things go wrong, then they start identifying which aspects actually did matter, thinking they were trivial, of no consequence and then explain, why the designer was so lacking in good judgement.

In my case (my model), I have undertaken a number of small repairs, just to maintain airframe integrity. As to knowing the cause of the low level Buzz, I am no nearer. When first operated, the drive train was like a turbine (that is one operating some way from the resonant speed,a joke, well almost). It was silky smooth, to an IC man, they would think the motor must be switched off.

As I think we are all agreed, there does no appear to be a fundamental issue with just statically balancing a prop.

At the moment, there are events in my life which make all this seem so trivial and so unimportant, particularly when there is little evidence to suggest the issue is a problem which haunts us. For IC modellers where the angular velocity changes over a single rev ( 2 pi radians another little bit of trivia to modellers) and the effects of the piston, most other effects are staggeringly small. But then the propellers are also generally beefier. And the debate goes on, and on, getting nowhere.

That is other there is no problem in reality.

[Tony K]

Posted by Erfolg on 20/09/2012 23:50:33:

As I think we are all agreed, there does no(t) appear to be a fundamental issue with just statically balancing a prop.

This sentence, along with the rest of Erfolg's post, just about sums it up.

[Simon Chaddock]

Just a bit of a teaser.

Assuming its hinge point runs true can a folding prop ever be out of dynamic balance?

[Myron Beaumont]

Simon

If it is out of balance ,then aerodynamic forces will mean that the thrust from each blade will be different resulting in maybe more "out of balance" criteria .

By the way ,I havn't a clue what I'm talking about ! We're well and truely in the realms of hypothesis and theory .Nothing to do with what happens that has no significance in this field .Never bothered me anyway !

And you can't assume the hinge point runs true .Can you ?

Myron - hinge whinge dept

The whole topic is getting silly isn't it ? I shall now read the mag article !!!

Edited By Myron Beaumont on 21/09/2012 13:07:53

[Myron Beaumont]

OK,I've now read the article. I think it was maybe meant for next Aprils edition (1st -that is ) .I balance my props NOT using the cone shaped thingys on the shaft provided for obvious reasons (to me) .but clamping the prop as per engine .In other words ,how it is properly held .

[Martyn Johnston]

Simon,

If you mistakenly assembled your folding prop with, say, a 12"x6" prop on one side and a 12"x5" on the other, would this give you what you're thinking?

[Peter Beeney]

With regard to the earlier post concerning the hollow backs of APC props, I don’t think it makes much difference regarding the surface area of the back of the prop. The amount of friction will be the same whether large or small. If we have a one square inch of surface with a weight of one pound standing on it the rate of pressure is one pound per square inch. If the area is reduced to a half a square inch with the same one pound weight on it the rate will be two pounds per square inch. The APC props have a narrow rim at the back, which I think is a good idea, this ensures that a relatively small area is bearing with a lot of pressure on the metal prop driver. It also ensures that the serration or knurling on the driver bites into the plastic and helps to increase the friction. You can see this clearly on inspection. I’ve used APC props for a long time without any problems whatsoever. If the prop nut is tightened sufficiently, and quite often it’s not, the electric starter will never spin the prop off. I very much suspect that many others have also used APC props with impunity.
Once I’ve read Brian’s chronicles, hopefully I think tomorrow, it might be possible to comment on what he says. I’m now slightly confused as to what he does exactly say. On the face of what’s been said so far it seems to be just a rather futile exercise in just trying to BBWB, (Baffle Brains With Bullshine), where indeed has there ever been any evidence in the past that this has been a significant factor in any model propeller operation? But I really don’t think that I could ever justify drilling holes in the hub, in any circumstances. Looking at one closely it’s quite possible that you could get away with it to some extent, it might well function for a time but I don’t think that I would ever consider using a prop in this condition, there is always a chance that it might break up and then would fly off in any direction.

Having now read Brian’s column, (where did the Dr title spring from??), I’ve not changed my mind one jot or tittle. I only use APC types, so I can only really speak for these. With the greats respect to Brian, I certainly have no problems if he wants to statically balance the prop, but in practise how many actually are balanced by the pilot, 1 in 10? Or in100? Also what I can say, having just been tinkering, is that as the back of the boss is hollow the centre hole is relatively short; and I’m convinced that the squareness of the prop has not too much to do with the squareness of this hole, rather all to do with the squareness of the back of the boss that bears against the prop driver. When the prop is on the shaft with the nut tightened correctly, that’s Tight, the two surfaces are close together, and square. Or, at least, as square as the prop driver is. I think the hole would have to be quite a long way out of centre to make a noticeable difference here. I’ve just checked 4 props for tracking and they are spot on, and looking at the prop, I suspect they are all made in the same way, and the back of the boss is always going to be square with the blade. Also I’m not convinced, not even a tiny bit, that the machines that drill the holes are going to be all over the shop at odd angles, either.

Now that has led on to one other little query I now have. We often use the orange Flexi props, usually on model SPAD types, (Simple Plastic Aeroplane Design), and suchlike, these tend to frequently hit the ground usually because they are flown too low; whether by accident or design. The Flexi props do not break. But when they run they do flex to and fro considerably, as the motor speeds up and down. If this tracking is so important why then are these props actually so good? We have always been very impressed by the efficiency of these things, the models fly perfectly, to the point that sometimes it seems almost worth while sticking them on everything, but the size range is very limited.

And as before, I would personally not be involved with any drilling of the hub. At all. It seems to me that any benefit is questionable at best, if this were a common occurrence it would have been highlighted long before now, and if the drilling resulted in the prop failing I think that would be, for me anyway, unforgivable. And as in a number of other such circumstances, it’s most likely to fail when the device is operating at full power.

If I obtain a balancer I can then do a few checks for myself. I cannot speak that convincingly for electric either, but we do have quite a few electric pilots and I’ve never know anyone to have any particular problems with vibration. That said, it could of course still be perfectly possible. I can also appreciate that balancing electric props might well be beneficial.

Now please forgive the little diversion from track, but it’s still relevant to the mag, I just happen to open it at page 46 and there right in the middle there is a photo of two meter prods in the back of a balance plug. I would like to suggest this is definitely not the way to do it. If they touch it will short out the cell. A thread with some much better thoughts of how to do this might be a good idea……

PB

Edited By Peter Beeney on 22/09/2012 15:17:43

[Hugh Joseph]

Hi All, new to the forum but interested in the comments.

I don't know too much about prop blades but assume that there is some deflection of the blades at speed so this alone will affect the imbalance.

Whether or not the prop need balancing depends on the mass, speed and permissible residual imbalance - dynamically balanced rotors run smoother with less vibration that can cause premature wear to bearings of fatigue in other system components. In some cases smoother running rotors will run faster.

I have many years experience in high speed air bearing rotors and now also balance turbochargers which are a bit basic and aren't truly dynamically balanced.

One point I would like to make is with regard to static balance - this might be fine for a very slow rotation but in my experience if balancing is required then just correcting the static is not enough. The best way to view static balance is to picture a long shaft with a disc on each end. If the assembly is perfectly statically balanced the can slowly be rotated and it will never stop in the same position. If you then add a 100g weight to 0 degrees on the front disc it will then always settle at the bottom. The static can be corrected by adding a weight to the rear disc 180 degrees to the first weight. So the static is now perfect but what would happen if you rotate the shaft at 15,000 rpm! It would not run very smoothly at all.

I am not sure if dynamic balancing would suit this application but if anyone wants to do some trials I would be interested in doing some development work with them