Strobe Balancing

[hugh hargett]

Read an article sometime back By Andy Kunz on this. Sounds like a great idea. Seems like something that could be put together with a small brushless motor/esc, a propeller shaft a strobe and microcontroller and like the audruino you fellas are so fond of.Built and executed correctly this has alot of potential. Ideas????

"Spin Balancing

 

After you have finished static balancing, you need to make sure your prop will stay balanced when running at top speed. This is a lot more difficult to do, and it's where I sometimes cheat myself. If you've taken the time to balance your prop as described above, you're already doing better than 95% of the rest of the guys out there. Here's how to get that extra 5%.

 

Spin balancing is what they are doing to your car tires when you have them balanced at a garage. The tire is spun, and a computer inside the tire balancer tells the operator where to place a weight and how big it should be. He will spin the tire multiple times, until the computer tells him it's done. Since there aren't any such tools available for us, we need to make do with what we can. This is another one of the other reasons I like my Top Flight balancer.

 

Basically, we will be spinning the propeller while using the stroboscope to make it appear as if it isn't moving. If it isn't running true, we remove weight from the heavy blade until it is balanced. Here's how to do it:

 

First, make a small mark with a permanent marker on the tip of one blade. Next, you spin the prop on the balancer shaft. While it spins, shine the strobe on the prop, watching carefully to see if the shaft is rotating perfectly. By watching carefully, you will be able to see if the shaft is running true or moving in a circle. The ink mark will allow you to identify which blade is heavier (the heavy blade will appear to be on the outside of the wobble). Lightly oil sand the heavy blade to adjust it. Repeat this until the prop is moving in a perfect circle, with no movement of the shaft. If you have done a perfect job of this, your prop will still be perfectly static balanced. If it isn't, you will have some vibration at certain speeds but not at others. The goal is to make it perfect at all speeds, but if you can't do that, just make sure it's perfect at operational speed. The ideal spin balancer would spin the prop at the speed you would be turning it on the water. I don't have a tool, which would allow me to do that, so I just spin the shaft with my fingertips. If I ever figure out how to get 30K RPM on the balancer, it will be a good day at the races!

 

Mount the prop on the drive shaft in your boat, and run the motor up to the expected operational speed, using the optical tach to find that speed. It will probably be between 50% and 75% of the maximum speed your motor turns, so you can work it from that angle as well. (This is where stick radios with the spring return taken out can be very handy.) Check to make sure the prop runs true at operational speed. If it doesn't, you can find the heavy blade as you did on the balancer, but it might not be the prop that's out of balance! Try making sure your drive shaft is balanced as well - it should run true at even full RPM. This is a very fine detail, but it will help keep you from wasting power. One little hint - always mount the prop the same way on the drive shaft. I like to make a very tiny notch, which identifies how the prop mounts to the drive dog. This will help you keep your drive system balanced, not just the prop.

 

Author, Andy Kunz."

excerpt taken from http://www.modelpowe...eller-balancing

COOL B)

 

[hugh hargett]

cool idea

 

[hugh hargett]

A cool project for all of us .

David, how you going with that aurduino language?? If youre out there...... David Wilfong has been working with the atmel microprocessor syntax on a lambda nitro engine management system - it is something that would be good for this. You could put an optical shutter on the blades or on the drive motor for your rpm input. A high intensity white led on and output for strobe. Incorporate a simple alphanemeric 2 line display. The audruino will interface into a hitachi hd44780 lcd driver chip easily.That will give you a speed readout. Ud use a pot for the speed on an input and drive an esc off an output. Its doable.As far as drawing the dot thats cool but maybe photo recogknition through hd camera or laser.

Dont know if you know it David but atmel mcu's reside in alot of the FE esc's you see on the market today. Knowing atmel chips can make you some ca$$$$H!.

Hugh

 

[Jim Allen]

I have read & studied carefully the article written by Andy Kunz some time ago. His assumption that the blade is statically balanced when it sits level is incorrect. If the propeller blade sits with the same leading edge down & the same trailing edge down, everything below the center line is heavy. This would also be true for 3, 4...6 bladed propellers. If the blade in question is balanced it will sit anywhere it is placed! If the propeller blade is statically balanced in a magnetic balancer, you will be able to blow compressed air on the blade while it is in the balancer. It will spin about it's center axis without any wobble to the axis. This becomes very difficult to do if the blades are not mounted to the propeller's hub in exactly the same plane. With the pitch gage I use, rotating 2 bladed propellers 180 deg without moving the drive dog instantly tells me if the blades are in the same plane. The same thing can be done with 3 bladed propellers when the indexing of the drive dog is changed to 120 deg.

Jim Allen

 

[hugh hargett]

Thank you for the info Jim . The first thing I want to try is to chuck up a propshaft on a brushless motor with an octura collet and see how tru it runs anyway without a prop.

 

[Jim Allen]

Do this with more than one collet. You will be surprised at what you see.

 

[Mark Bullard]

What your talking about is way to much money. They make machines that will balance something that small but very costly. I have two hard bearing machines in my shop and when the guy that sold them stop by and showed us a machine that would do props and cranks. Price tag $20,000 dollars. It was his show machine that fit in a suitcase for selling customers. I tryed to get him just to leave it and he said he could for the price of the machine. It was cool and would be nice to have one but I do not balance anything that small.

There is one thing I do know that our props are couple unbalance a lot of the time. What this means is the the prop is out of balanced 180 degrees apart. The left side will be one position of out of balance and the right side will be about 180 degrees the other way. This will make a prop have a gyro effect and can not be balanced out by static balance.

The machines I have are a 15,000lbs and a 200lbs total weight machines. The big one is the same that they use to balance jet engines. Very good machine.

 

[dwilfong]

This balance thing is some thing that I need to figure out.

yes Hugh I am still here.

I am more concerned with the balance in the eng than the prop.

The mass of the prop is much less the mass of the crank and rod piston assembly.

If this is out of balance the prop is a after thought.

Lets just say I have encountered problems that the manufactures have not taken in to account when making the Larger engines. Yet alone adding a heavy billet piston to the mix.

Now with building a hybrid eng it will be more of a issue.

So what is the answer?

How to balance a rotating assembly with out guessing at what the % of out of balance will be?

Any input to this question?

David

 

[Mark Bullard]

This balance thing is some thing that I need to figure out.

yes Hugh I am still here.

I am more concerned with the balance in the eng than the prop.

The mass of the prop is much less the mass of the crank and rod piston assembly.

If this is out of balance the prop is a after thought.

Lets just say I have encountered problems that the manufactures have not taken in to account when making the Larger engines. Yet alone adding a heavy billet piston to the mix.

Now with building a hybrid eng it will be more of a issue.

So what is the answer?

How to balance a rotating assembly with out guessing at what the % of out of balance will be?

Any input to this question?

David

You could do a static balance on the crank or look around and fine someone with a small machine. The balance job will cost about $100 to $200. I could call my rep for my machines and find who may have a small machine that does balance jobs. But most of the small machines like this are used in companies that only balance there own part that they make and do not do outside work.

The setup on a crank is the same for static and dynamic balance. You must make a weight that weighs 60%of the total rod, piston, and pin together that can be attached to the crank pin. You can then static balance this crank pretty good this way.

 

[jefffro007]

I have had the pleasure of balancing many rotating assemblys on 3 different Hines hard bearing balancing machines very similar to the one Mark has pictured above . I have always wanted to try to do one of these small single cylinder engines but am un sure if the smaller engines would have enough mass to read it .... I think the gas engines may benifit from it if theres enough mass. David i do have some single cylinder percentages in a AERA Automotive Engine Rebuilders Association book i have used for Briggs and Stratton engines. If i remember they hat two different ones race "high rpm " and stock .

Jeff

 

[jefffro007]

Here's what i have for a few single cylinder engines .. Note the two cycle and race percentages are higher due to higher rpm .

Briggs race %67 Recip %100 Rotating

Briggs stock %50 Recip % 100 Rotating

Kohler %50 Recip % 100 Rotating

Yamaha %61 Recip %100 Rotating

Jeff

 

[dwilfong]

thanks guys for the info.

On the one sided cranks it will be simple But on the two piece pressed cranks if the rod is more than 60% of the total it will not be.

When I get home I will give it a try.

David

 

[hugh hargett]

About spin strobe balancing props is it possible or impossible? Is it possible to put something on the table top for less than a 1000 dollars that would accomplish this. Theories anyone? I can appreciate whats been said so far as $$$$$ is concerned, but we would likely be looking for something that spin balances up to a max of say 50k .Something you or I could put together as a tool addition to our shops.

David for what youre doing you gotta get all the way in it. If youre going to continue with motor design you may as well get with Jim Allen, Marty Davis and the like on fixtures and tools dedicated to small 2 stroke engines.Theres just no way around it. To pay someone every time you want to play with a crank might get costly as you clearly like to tinker alot.

Hugh

 

[Jim Allen]

In my early days of engine building, I attempted to balance engines using Sir Henry Ricardo's formula for balancing single cylinder engines. Ed Kalfus used this formula to balance his .90 cu in nitro tether boat engine. This worked extremely well for him because the engine used a "hanger style piston". There were no piston bosses, the wrist pin was as short as possible & the piston had two large intake windows located where the piston bosses would normally be. The engine also had two exhausts.

The fact that there were no piston bosses, two large windows in the piston & a very short hollow wrist pin made the reciprocating mass extremely low. The hanger piston also allowed the engine to have the longest connecting rod possible. The connecting rod was case hardened steel with rollers at both ends.

Ricardo's formula states that 1/2 of the total reciprocating weight plus the total rotating weight should be on the crankshafts counterbalance. The reciprocating weight equals the total weight of the piston, wrist pin, circle clips, roller assembly or bushing, thrust washers & wrist pin plugs. The rotating weight equals the total weight of the lower half of the rod, roller assembly or bushing & crankpin within the rod. This formula does not work in engines with piston bosses, wrist pins which span the piston's diameter & pistons without intake windows. Through many engine dyno tests, I found that 1/3 of the total piston weight & the total connecting rod's assembly weight gives a smooth running engine in the 20,000 to 35,000 rpm range. Engine rigidity has a great influence on shaking & vibration problems. The use of a full hard steel front end makes engines run smoother than any type of aluminum front end. Engine motions at WOT were examined with a strobe during dyno pulls. You would be very surprised what a solidly mounted .90 cu in engine with a bar stock case & a hardened steel front end, running at 30,000+ RPM & making 7+ HP is doing!!

Jim Allen

 

[hugh hargett]

Im curious as to the frequency of the strobe. Sorry fellas I will tangent off to electronics in this thread from time to time.Im still with you though.I guess having a variable strobe frequency will aid in what Im trying to do.

http://www.youtube.com/watch?v=eNymJ9LaMrw&feature=related

ok they tie the strobe frequency into the rpm . That makes sense. So 25,000 rpm is about 416 hz flash rate?????

David, that may be a useful conversion for a audible pipe tach sensor hz times 60 give you rpm. In you controller youd put a value of 60 in a register based on your internal clock speed you have calculate a second measure the amount of pulses in a second and store that number in another register and have the mcu perform a math function that mutliplys the two values you have store in the registers then outputs the results to another register which can be output to a lcd or desktop. Its crude psuedo but I dont know the syntax so bear with me. im still thinking there has to be a way to pick up a pulse from a revolving nitro motor that would allow you guys to spin balance you prop while on the boat. im looking for a way to do the entire shaft out of the boat then return it to the user balanced. It will mean do not alter it or even think about removing the prop unless you know whats going on after you got it back.

Hugh

 

[Mark Bullard]

Im curious as to the frequency of the strobe. Sorry fellas I will tangent off to electronics in this thread from time to time.Im still with you though.I guess having a variable strobe frequency will aid in what Im trying to do.

http://www.youtube.com/watch?v=eNymJ9LaMrw&feature=related

ok they tie the strobe frequency into the rpm . That makes sense. So 25,000 rpm is about 416 hz flash rate?????

David, that may be a useful conversion for a audible pipe tach sensor hz times 60 give you rpm. In you controller youd put a value of 60 in a register based on your internal clock speed you have calculate a second measure the amount of pulses in a second and store that number in another register and have the mcu perform a math function that mutliplys the two values you have store in the registers then outputs the results to another register which can be output to a lcd or desktop. Its crude psuedo but I dont know the syntax so bear with me. im still thinking there has to be a way to pick up a pulse from a revolving nitro motor that would allow you guys to spin balance you prop while on the boat. im looking for a way to do the entire shaft out of the boat then return it to the user balanced. It will mean do not alter it or even think about removing the prop unless you know whats going on after you got it back.

Hugh

There is a company that used to make good strobe machine. The name was Dynabal. They are still in bussiness today but all of their machines use computers. There maybe one of these out there cheap. I have one but it is not for sell. The only problem is the pickup are of the seismic type and do not work very well for small weights.

 

[hugh hargett]

What we really want to do Mr Bullard is come up with something scaled down for props. A small device. It can still be run by the computer or laptop.

 

[Mark Bullard]

What we really want to do Mr Bullard is come up with something scaled down for props. A small device. It can still be run by the computer or laptop.

I understand that and they do make them. You just have to get off your pocket book.

But the way the strobe balancer works is the seismic pickup measures the movement of the shaft in the plane that the pickup is mounted. This pickup floats on a set of waves on each plane. This is know as a softbearing machine. On a hard bearing machine the seismic pickup will not work as well. It makes no different on what speed you balance at. Most machines balance between 300 and 600 rpm. Your system must be able to tune the frequency that the seismic pickup sees on both left and right planes. This is how you lock or freeze the part to be balanced that make it look like it is not turning. It is also matching the RPM of the turning part. Now you have to measure the amplitude that the seismic pickup read. The larger the number the more out of balance. The goal is to add or remove material to bring the pickup reading to the lowest level on the left and right plane. One other problem is the speed of balance will move the out of balance position. So you must know where to work your degree of balance position. The new machines do all of this because you enter the position of part from waves, the diameter, weight, and max speed the you want to balance to.

I learned how to balance rotating equipment on a strobe machines many years ago and it take years to master it.

The machines that I have now will allow me to balance something like the picture in two runs to a G 1 spec. But these two machines were a $140,000 dollars. Good luck with your project.

 

[hugh hargett]

So, Mr bullard are you saying that the strobe frequency is in tune with the vibrations picked up by the piezo device and the rpm ? How so? I can understand the theory behind the floating fixture I kind of alluded to that earlier in the thread back a few post. let me ask you this. Once I found the heavy side of the prop could I simply balance it by removing material until the said harmonics are removed? Is there a piezo element or knock sensor that measures finitely enough to do what we want to do here?Is it a possiblity or is it just foolish?we can isolate or exclude a certain frequency by way of a notch filter or bandpass filter.Bear with me. You are obviously very seasoned to this kinda thing Im trying to keep up so please go slow.Once I had a piezo device that could handle it I could scope it and that would tell me quite easily what the dominant frequency is and tune it accordingly. Auto tune is quite possible too with the right code in firmware via appropriate register refrencing and arithmatic processing. I have a few pic 28 and 40 pin dsp chips that get down with it if need be. We may be wandering outside of an aurduinos capabilities but David the sytax is still very valuable . You can always upgrade to an atmel 32 bit dsp device and if you need it in a developmental package all you need is a crystal for clock and a voltage regulator for the auxillary. Theres always ebay too.

devices.

 

[Mark Bullard]

So, Mr bullard are you saying that the strobe frequency is in tune with the vibrations picked up by the piezo device and the rpm ? How so? I can understand the theory behind the floating fixture I kind of alluded to that earlier in the thread back a few post. let me ask you this. Once I found the heavy side of the prop could I simply balance it by removing material until the said harmonics are removed? Is there a piezo element or knock sensor that measures finitely enough to do what we want to do here?Is it a possiblity or is it just foolish?we can isolate or exclude a certain frequency by way of a notch filter or bandpass filter.Bear with me. You are obviously very seasoned to this kinda thing Im trying to keep up so please go slow.Once I had a piezo device that could handle it I could scope it and that would tell me quite easily what the dominant frequency is and tune it accordingly. Auto tune is quite possible too with the right code in firmware via appropriate register refrencing and arithmatic processing. I have a few pic 28 and 40 pin dsp chips that get down with it if need be. We may be wandering outside of an aurduinos capabilities but David the sytax is still very valuable . You can always upgrade to an atmel 32 bit dsp device and if you need it in a developmental package all you need is a crystal for clock and a voltage regulator for the auxillary. Theres always ebay too.

devices.

The simple strobe machines are in tune to the RPM by the movement of the vibrations. The data recievers were nothing more than voltage anlog meters with switches, filters and trim pots to tune the signal for differant amplitudes. I think the seismic device on the old machines I had were 0 to 5vdc. And the base speed to balance at was 300rpm and the reason was the pickups were mounted on the back side of the waves.. The waves moved from front to rear on balls slides. At 300 rpm the lag time in the electrical signal to be process and read on the meter made the unbalance directly in the front of the machine in the light of the strobe. We would always write number on the part to know where it was at. We would add weight in the rear and remove weight in the front. Also each plane had to be done separate from left to right. This machine had a switch to go from left to right plane.

The new machines that I have use piezos for the pickup. And I am not sure that a piezo will be able to work on a softbearing machine because of to much movement. My machines are hardbearing machines. The waves have two flat bars that carry the wave bearings and the piezo is mounted in a way as to be preloaded to a voltage level on one of the bars. This machine still has a left and right pickup. The rpm and angle position is done with a encoder. All of these signals are recieved into their measuring module. These modules have their firmware and design of many years of deployment. The computer is nothing more that a dum front end. It can not process the information fast enough with all the signals connected straight to it. But the dum front end does have a nice good working software to run the machine. It will display the position of unbalance in add or takeaway with the position of the ball in the large circle. The little circle follows the angle position of part. It will use any format of measure that you want to use. And you can see that the left and right side are displayed.



Display and Measuring module