Weight and Balance Aircraft attitude

[Nev]

I'm going to remeasure all the station points on my 4-place for the spreadsheet I've made.
When putting the aircraft in the level flight attitude to make the measurements, what is considered to be the level reference?
Is it the horizontal upper door frame?

Thanks.

[Nev]

Actually I found it.

DA876837-D4AD-4C57-9C33-31CF2E1368DB.jpg

[Battson]

I suggest checking with Mark. The factory jigs and construction methods are what's of most interest here.
Noting that Bob's plans are owned by Bob (R&B Aircraft LLC), and the majority of our airframes are built by Bearhawk Aircraft LLC, which is affiliated but not the same - there may be some differences.
Previous advice has been top of the window frame - but things change over time, construction processes evolve etc.

Engineering tolerances should also be considered throughout the analysis. Measurement, tolerance, and uncertainty analysis is most easily done in parallel with the underlying calculation, it's a lot less work that way. I suspect +/- 2% would be a best-case outcome here.

[Battson]

In case anyone wants further reading on calculating errors when you add numbers with tolerance / error (I doubt many will be interested in this):

https://www.geol.lsu.edu/jlorenzo/geophysics/uncertainties/Uncertaintiespart2.html

[Brett795]

In my notes for the BH5, I had it copied down as Fore & Aft- top of front door Frame and Rt-Lf- holes in front wing fittings
Following this to see if this changes.

[Nev]

I checked today with a laser level and found that the top and bottom of the doorframe both match the alignment of the datum.

[rodsmith]

On my scratch built fuselage I leveled the fuselage using the thrust line marked on the plans and a water level. I then used a digital level to check the lower longeron just behind the rear gear leg fitting on both sides. One side was perfectly level, the other .1 degree off, so that is where I go to quickly level the fuselage, have never checked the upper door frame.

[whee]

Battson has the best answer IMO.

Years ago I was checking the balance sheet of a BH I intended to buy. I didn’t have plans so call Bob and asked him. He was very clear that the level datum is as Nev depicted. I asked about the upper door frame, because I read it somewhere, and he said something to the effect of “it could be the same, or not, just use the designated location.” But that’s for his plans. For the kit sets I’d use whatever Mark says to use.

[Nev]

Mark Goldberg

Mark, would you like to weigh in here with your thoughts? Which datum do you recommend ?

[AKKen07]

The datum Mark told me to use is the one indicated in your plans. Under the door.

[Nev]

I re-checked all the arm positions on mine yesterday using a laser plumb bob. The most notable observation (as we all know) is that a very small change in pitch angle during the setup leads to a large change in the arm of the main landing gear, due to its position vertically below the datum. The other arms are barely affected, although the front seat arms are also sensitive because they also lie vertically close under the leading edge. I spent a lot of time transferring the arm positions to the floor and measuring them. Then I completely re-did it by putting the arm positions on the fuselage to remove any parallax error (from the laser on the floor).

Before marking any arm positions I used a spirit level to check the laser level, and then checked the levels of the upper door frame, lower door frame, and the lower longeron (datum) against each other. I was unable to detect any measurable difference in alignment between them. However a very small change in attitude (tailwheel height) that was difficult to perceive resulted in a disproportionate change in empty CG position. On my aircraft, there was up to a 1.8 inch change in empty CG position between the first measurement (checked by an engineer and another pilot when new) and yesterday's measurements with the laser.

I propose that once we get Marks thoughts to verify this, that for kitsets we use a vertical measurement from the floor to the datum, and another vertical measurement from the floor (axle) to a point on the rear fuselage. This will remove any differences in pitch angle and ensure that all Bearhawk's of a type are being compared equally. Once we agree on a datum and the correct arm lengths, then there is no real need to measure these individually (unless modified), though you may still wish to do so and it will much easier to set up knowing the exact angle to use.

The problem with not doing this correctly is that the empty CG results can vary quite significantly, and the forward and aft limits can be out by the same amount.

I observed several other interesting points during this exercise:

My propeller appears to be located 1.6" forward ( at -61.6") of some A models.
My tailwheel is located approximately 10 inches further aft of those with leaf springs.
The change in front seat position on their rails complicates things slightly, but overall the effect in minimal because it's very close to the aircraft CG.
The difference in engine CG position between an IO360 and an IO540 is quite small - presumably the forward 4 cylinders all lie in the same position (but the weight difference quite large).
My propeller arc was 15.5" from the floor in this position.

7EDC7AE7-FF36-4FAC-9712-2DF93EFF9E8A.jpg
A552A0FE-8CCA-48C0-862E-C15CC392D772.jpg

A6EE44E3-D542-4D46-A900-1C5929204DEE.jpg

4D515082-D207-4407-8258-B379FDB2FA91.jpg

AE84E622-6D3A-45A4-AECB-71683502CD63.jpg

0FE5660C-7033-450C-8758-988CF842C0BF.jpg

[Battson]

Hold on - what's supporting the tail

[Nev]

I've trained it to do that. Does yours do that ?

[Nev]

Mark emailed and suggests that Bobs datum is a good one to use, and that the top and bottom door sills are also parallel with the datum. I'll attach a screenshot of the spreadsheet below. Currently I've done it in numbers so it works on an iPad or iPhone, and this one is imperial (although you can see I entered the data in metric !). I'll do a metric one in a while.

Feel free to look over the arm measurements in the yellow column, and let me know if there's any errors. It's easy to fix.

Three arms will likely differ:
1. I'm told that many of the NZ Bearhawks have a longer prop spacer (for some reason) so the arm on other props may be different.
2. Aircraft equipped with a stinger appear to have significantly longer tailwheel arm than those with leaf springs (in the fully extended position).
3. There was a very early weight and balance chart published by Bob for his prototype (it's on this forum). The landing gear position on the prototype appears to be aft of current kits. If anyone used the prototype arm measurements to calculate main gear position for a kitset, it may be worth double-checking as it could lead to an error of around 1.5 inches. (The real CG would be forward of the calculated CG).

In a nutshell, if you know the wheel weights they would be entered into the magenta boxes. If you know the wheel weights of a friends aircraft you can use those as a starting point. Green boxes are for normal weight and CG calculation for everyday flights.

The blue boxes are to explore different scenarios. You can add or subtract weight from the various stations and see the result in the grey boxes at the top.

3A5DBDBC-FD61-41BC-B4B5-7BD147380919.jpg

[Nev]

Thanks for that. The main wheels are the most difficult to measure and where we all seem to be seeing the most variation. Actually my final measurement was not the one pictured above, rather 65mm, so about 2.5". We're getting close.

[Nev]

And there's always the possibility that yours are more accurate too! Mine were simply my best effort.

[AKKen07]

Well I like 2.5” much better! Hah 1/2” makes me much less panicky. Not that I panic of course… panicky-ish. Our other measurements, except fuel tanks, were within about 1/2”.

[rodsmith]

My propeller arc was 15.5" from the floor in this position.

Very interesting post. I'm wondering how long your prop is? I believe you are on 29" bushwheels. I have the aeroclassic 8.50s and when level I have the same prop clearance (15.5") with an 82.6" prop.

[Nev]

82" prop, and yep 29" ABW's. I'll double check the prop length just to make sure what is installed is what was ordered.
The 29" wheels measured from memory at about 28" at 16 psi, and I currently have them at 8psi so that might account for it.

[Battson]

Did you happen to check how parallel the level parts of the fuselage (top window, bottom longeron, etc) are with each other?

[Nev]

Did you happen to check how parallel the level parts of the fuselage (top window, bottom longeron, etc) are with each other?

Yes, they're all very parallel with each other. I couldn't discern any difference.

[Nev]

What I found after leveling it 3 different times (on different days), is that the results are difficult to repeat with accuracy. From looking at the results of others on the forum it appears that this may be a common issue when lifting the aircraft tail to level it for a weight and balance.

The significance of this is that a one inch difference in tail height leads to approximately 1/2 inch difference in the distance measured between the main landing gear axles and the leading edge of the wing, and also about a 1/2 inch difference in empty CG, and both CG limits. One inch may sound like a lot, but apparently it's not. That's using a laser level, a plumb bob, and a spirit level.
The main gear axle position in relation to the leading edge is critical measurement to get right. The other arms don't vary much with a change in tail height.

Once I got a tail level that I was able to repeat with the laser, I then used the laser level to measure the height of the tail (where the lower longerons meet the vertical tail post) above the height of the datum. The height difference is 510mm. This can then easily be converted to a height above the hanger floor to get a very repeatable, level aircraft.

This gives a measurement of 65mm between the leading edge and the main gear axle. This seems to be in line with what others have posted.

Using this measurement (tail height above datum) - if we all agree on it - would mean that every 4-place B model is being measured in exactly the same fashion, eliminating errors, and the CG limits will all be the same.

I would welcome constructive feedback and input. I have a spreadsheet ready to go, and if it simplifies things then that's great.


DAA742D0-1D29-4A31-9FBE-D7CD106289B5.jpg

66D10C07-A67E-49A7-AD2E-885716A1F371.jpg

5E21F13F-0AD2-4B65-910C-C65BDB01DB2E.jpg

DDE47203-2274-45F2-BDB2-1BA807FEA3FB.jpg

[Bcone1381]

Feedback:    The cement floor in my shop is not level.  Neither is my Garage, nor the floor in my basement.  These all have high and low spots that risk inducing measurement error.  

It doesn't take much to move that string, or for fabrication standards to exceed your required tolerances.  

I like the repeatablity and accuracy of a water level read with high power magnifier reading glasses. Although the bottom of the two wing spars has nothing to do with Bob's level datum, Bob really defines the CG according to the Wings Mean Aerodynamic Cord.  The Patrol has a +2 degree incidence.  Yours might be different.

I'm a poor designer and no engineer.  But I think to get tight tolerances on your CG markings this will comply with Bob's Drawing on Plans P.1 with tight tolerances.  

Attach a long straight piece of stock the bottom of the wing.  Drape a Plum Line over the wing's leading edge.  Adjust the tail up and down to set the bottom of the wing to the Angle of Incidence Bob calls out on P.1.  (maybe its +2 degrees)

To get that angle, I'd attach a big carpenters square to the straight edge.   Trigonometry says 18 inches below my wing, a 2 degree pitch will position the square 0.628"  ahead of the Plumb Line.  So adjust your tail so its just so.......Now your Level Datum is where it should be.  Mark your spot on the Longeron abeam the plumb line. Mark your floor boards and tell us what you got.

[Nev]

Brooks thanks for the feedback, exactly the type of thing I was after. I did try a straight edge under the wing to see if I could use incidence, but the B model riblett aerofoil isn't straight on the lower surface, so I only used Bobs datum. To eliminate the effect of uneven floor surfaces I used the laser level, and hence measuring the height of the tail above the datum. I like your idea of using water in a hose, also easy to do and very repeatable.

With regard to using Mean Aerodynamic Chord - I avoided referring to it - because MAC is normally used as a reference on swept wing aircraft where the chord is not consistent (hence Mean). I've only referenced the measurements 10.5 - 22.5 to keep it simple. It avoids doing further calculations. But those are only my thoughts and if people felt that MAC was a better reference then there's nothing stopping us from adopting it, though I do think we should just choose one method.

[Bcone1381]

I agree Nev! You do great work