Light Weight - 0.01 Digital Protractor for Prop Pitching

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

TiPi

Well-Known Member
Log Member
Joined
Aug 25, 2019
Messages
421
Location
Peeramon (AUS)
this is the fixture I built. a V-channel that fits over the trailing edge and an adjustable contact point that rests on the blade backside.
each blade needs to be very accurately marked for a certain distance from the hub (actual distance doesn't matter, just needs to be the same for all blades). If you have verified that every blade is the same length, can also measure back from the tip.
then place the bracket at the correct location, move the blade until until the spirit level is dead level and take your measurement (digital level or laser pointer).
Move the whole setup to the next blade and repeat the process.
Extremely accurate and repeatable (keep your noseweel/maingear securely locked, removing a set of sparkplugs makes it a lot easier to turn over the engine).
 

Attachments

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
Fair enough. I once rented a Duchess that had visibly different blade chords on each blade of one of its props. Not suggesting that a ground adjustable prop is likely to have that kind of disparity but if one is measuring/adjusting more than one prop your device becomes a manufacturing exercise for each prop.
An extreme example obviously and why we raised the question around residual (typiclaly 0.5 - 0.3 IPS) first order vibes post dynamic balancing being potentially aero. Essentially due to much of prop work being somewhat of an art. Thus the inception of the prop sock laser tool or equivelent, to allow accurate relative blade adjustments.

So far it locks on to within a mm on the four DUC blades I have tested.
 

davidjgall

Well-Known Member
Joined
Oct 7, 2010
Messages
117
Location
Northern California, USA
Sketch please.
Here's the bare minimum of what I was trying to describe. I left out the second laser. As I was drawing this I thought of so many "better" ways to do it; I'm sure that by aggregating several of the suggestions on this thread a robust instrument could be made. I apologize for sending a rendering instead of a photo. It would have taken less time to build the device than to draw and render it but I'm using this also as an exercise while learning some new CAD skillz.

Render 1.png
 

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
OK so after dialing in the track to ~0.1mm at the tips and using the aser pointer prop sleeve to match the pitch between blades, I managed to get the dynamic balancing down to 0.005 IPS at 2820 RPM.
 

Attachments

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
Now this might need a different thread but just in case billski wsimpso1 et al. are lurking! :)

This is the spectrum analysis at 2856RPM (pre dynamic prop balancing) I can understand that these first order vibes and harmonics are from the prop... (The cross hair is on the 2856RPM peak and the empty circles are on peaks matching the harmonics.)
Spectrum_Analysis_2856RPM_First_Order_Vibes_PreBalance_20210421_091537.jpg

And these second order vibes and harmonics are from the power pulses...
Spectrum_Analysis_2856RPM_Second_Order_Vibes_PreBalance_20210421_091537.jpg


... But what are these 1/2 order vibes? They feel like a bit like a bad cylinder even though all of the compressions are even and the engine is in very good shape.
Spectrum_Analysis_2856RPM_Half_Order_Vibes_PreBalance_20210421_091537.jpg

Or even these 1/4 order ones?? Really confused about what could be the origin could be here.
Spectrum_Analysis_2856RPM_Quarter_Order_Vibes_PreBalance_20210421_091537.jpg



 

Geraldc

Well-Known Member
Joined
Nov 12, 2011
Messages
583
Location
nz
Better off with one of these.Has a bubble to get the blade horizontal and a protractor function to set pitch.https://www.ebay.com/itm/E-Props-Digital-Protractor-for-adjusting-NG-D-Adjustable-Carbon-Fiber-Propellers-/152402744977
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,949
Location
Saline Michigan
I need more detail to give a satisfying answer... is this an airplane flat four? How did you obtain this data? What is the rev range of the engine? Isn't 2856 rpm a tad high to be worried over. What was being measured?

Since you are talking ips, I am assuming you are using a set of linear accelerometers. If you have torsional accel data, it would be in some rotational units.

As to the breakdown of your data, usually the data is broken down per some sort of specified windows, like whole orders and half orders.

A Single cylinder piston engine with conventional crankshaft and connecting rod has accelerations at firing order, twice firing order, four times firing order. Firing order is simply due to firing acceleration on each piston once every other turn. Twice firing order is due to the pistons going up and down in the bores - the piston accels are different in the upper and lower halves of the stroke due to the crank and connecting rod geometry, and that drives crank acceleration.

Combining cylinders can cancel some modes but not others, and the methods of combination dictate what is left over, assuming all cylinders have the same firing cycles. Consulting the standard engine texts will help with understanding the sources. Then, I have never seen an engine on test that did not have variations. You will see bank variations, variations between cylinders, and even (in pushrod V8's) cylinders in a pattern of variation following the firing (there is only one part number for the head assembly of pushrods V8), with variations in combustion chamber volume, shape, intake and exhaust flows and interactions, piston and rod mass/balance variation, etc. Oh, and the intensity and duration can even vary with how far away PTO is from the cylinder along the crank (cylinders at the other end of inline engines can twist the crank measurably more than the front cylinders).

All engines have non-firing order vibe. Some can coincide with other items, and are wise to correct for. Mostly however, when firing order is covered, folks quit worrying over them as the remaining ones are usually substantially smaller, except folks who worry exciting prop blades and engine accessories...

One of the major things that happens with boxer twin sets, like we use in opposed twins, is that the masses of pistons are balanced side to side but they produce a yaw axis couple. Set up a four with a crank symmetric about its middle, and the yaw axis accel is cut down, but still present. Go flat 12, and you give up boxer for paired crank throws and a crank symmetric about its middle.

All of this can drive use of balance shafts if you want smoothness more and can tolerate the weight. The use of bifilar torsional pendulums on airplane engine crankshafts started in pre WWII radial engines and is applied in horizontally opposed engines in what at first glance appear to be odd orders of engine rotation. They have also been applied to make car/truck Diesel engines acceptably smooth - they are applied at the flywheel or torque converter and are exclusively tuned to firing order here.

Taylor's two book set is the classic work on the topic of engine layout, etc. there are others.

Enough for one night from an dude with freshly broken bone driven insomnia. Maybe I can help with insights tomorrow.

Billski
 
Last edited:

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
Massive as always Bill, I am amazed you are able to still type!
Yes I think the half order vibes are yaw, not sure about the quarter order ones though. Might be an airflow swirl pattern and by the looks of the photos has a very small origin and most likely the half order harmonics are what we are seeing up frequency.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,949
Location
Saline Michigan
Massive as always Bill, I am amazed you are able to still type!
Yes I think the half order vibes are yaw, not sure about the quarter order ones though. Might be an airflow swirl pattern and by the looks of the photos has a very small origin and most likely the half order harmonics are what we are seeing up frequency.
Thanks. Where and what orientation are the sensors you are reading?
 

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
One piezo electric accelerometer mounted on the front top of the engine with the primary sensitivity on the normal axis. It sure does seem that half order vibrations of over 0.7 IPS are quite a bit higher than the usual 0.1 to 0.3 IPS seen on your average flat four with even cyclinder power pulses. 1.0 IPS is typical for a full missfire for instance.
 

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
I ran a cylinder balance check today running at 1900RPM and disconecting one injector at a time to verify no uneven power pulses. All cylinders were within 3% of each other so no issue there... Curious..
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
14,692
Location
Memphis, TN
That’s a nice box. It’s the kind I have access to. I have never done spectrum analysis on an engine though.
 

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
Agree, the Microvib is a very solid unit and looks to be accurate to another 10th of an IPS. I wonder if this 1/2 order vibe is something to do with exhaust resonances. I know some Subarus have a 1/2 order exhaust note. Seems unlikely TBH. Does anyone know if a harmonic can be excited down the orders as well as up? Apologies in advance for the probably very silly question...
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,949
Location
Saline Michigan
I ran a cylinder balance check today running at 1900RPM and disconecting one injector at a time to verify no uneven power pulses. All cylinders were within 3% of each other so no issue there... Curious..
You only need a forcing function and an Eigen mode to be within half an octave to have them amplify a lot.
 

Flow

Well-Known Member
Joined
Apr 27, 2016
Messages
110
Location
Auckland
Thanks but what could be the other part of the system besides one cylinder being 3% higher than the rest?
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,949
Location
Saline Michigan
Thanks but what could be the other part of the system besides one cylinder being 3% higher than the rest?
In car and truck engine crank output, if all of the cylinders were within 3% of each other, we would have considered that fabulous.

There are several reasons why. Combustion chamber volume, compression ratio, fill and exhaust, scavenging, even flame quench all vary from cylinder to cylinder. You have all that at the cases. Crank stiffness is much lower for the throw the furthest away from the flange, and stiffnesses are higher as you get closer, and even that will show up on the cases as lower stiffness raises peak forces and spreads out the pulses. Well designed crankshaft stiffnesses have to be high enough to put 1st torsional resonance 2.5x highest firing frequency, so this particular effect is likely kind of small, but it is in there as well as variations in combustion chamber volume, fill and exhaust, etc.

3% differences? Even blueprinting the engine won't get firing variations much much smaller than that.

Billski
 
Top