Pull-pull setup
12-20-2014, 04:07 PM
#51
Senior Member
Roo Man
Picked up the grands. One is getting over a cold and is resting. I went ahead and did a little more work on the drawing while they sleep.
As you can see, the straight arms will generate about 3/16" of slack in the cables at a 45º servo deflection. This is probably tolerable in most sport planes.
However, if you offset the cable connection point at the servo approximately 1/4" toward the rudder, you will get less than 3/64" of slack when deflected. You could probably increase the offset a little bit more and get rid of it entirely, but going too far and you will have a cable that gets too tight when deflected. It's a fun exercise anyway you look at it.
One issue that I did not address is the exit point of the cables. They exit considerably back from the slots provided. You can install cable guides as you originally suggested, but if you do, we should really determine their impact on the lengths of the cables.
In any event, check out the dimensions on the fuselage. Make sure that is drawn accurately. I can dimension the slot requirement if you like, but let's make sure that the fuselage is drawn correctly first.
They're awake now. Gotta go play.
Charles
Picked up the grands. One is getting over a cold and is resting. I went ahead and did a little more work on the drawing while they sleep.
As you can see, the straight arms will generate about 3/16" of slack in the cables at a 45º servo deflection. This is probably tolerable in most sport planes.
However, if you offset the cable connection point at the servo approximately 1/4" toward the rudder, you will get less than 3/64" of slack when deflected. You could probably increase the offset a little bit more and get rid of it entirely, but going too far and you will have a cable that gets too tight when deflected. It's a fun exercise anyway you look at it.
One issue that I did not address is the exit point of the cables. They exit considerably back from the slots provided. You can install cable guides as you originally suggested, but if you do, we should really determine their impact on the lengths of the cables.
In any event, check out the dimensions on the fuselage. Make sure that is drawn accurately. I can dimension the slot requirement if you like, but let's make sure that the fuselage is drawn correctly first.
They're awake now. Gotta go play.
Charles
12-21-2014, 09:47 AM
#52
Great info guys and my sincere thanks for taking the time to do the drawings. I'm so not a math guy but through trial and error I " Forrest Gumped " my way to a good setup. As far as where to locate the exits, I always made a full size drawing. Usually I would get the exit in the correct location but would have to start small and increase slot length as required. Most of this was done on fiberglass fuselages which made it easier. I did a couple years ago set up an airplane will pull pull on both rudder and elevator. I had to use guides on that one and found the effect of them we so minimal as to not consider it an issue. The guides in this case were further forward from where the cables ideally would have exited.
12-21-2014, 10:43 AM
#53
Senior Member
Pull-Pull
An easy way to not have to worry about arm lengths, Ackerman at the servo end and relative lengths of servo arm versus the length of the surface tab is to use pulleys at the servo end. A pulley will always pay out at one side the amount it pulls in on the other line. This eliminates the problem with the unequal changes you get with a straight arm when you have unequal lengths at either end of the pull-pull system. If you use pulleys, you just have to make sure that you do not have the attachpoints at the surface being moved forward of the pivot point. It can be a bit aft with no problem as you will get a minor bit of slack on the line not being pulled as you move off center. If you can get ahold of an old Model Airplane news (May 1963) you will find an article on making your home made pulleys an easy task.
12-21-2014, 03:10 PM
#54
Moderator
I've often wondered why no one has commercially produced a pulley system for pull pulls. A pulley on both ends would completely eliminate the geometry issues that we have, and would allow us to multiply torque by having a smaller pulley at the servo. It would also allow cables to be routed through guides as needed, eliminating the big slots that we have to add.
12-22-2014, 03:49 PM
#57
Charles,
I really appreciate all the work you have done.
The only way I can get an off set at the servo is to use an aluminum arm, the one I planned to use is a DuBro twin arm that the swivels are mounted between, can't have any off set with this one. Before I drill the alum arm at 3 1/8" wide with a 1/4" off set, I am going to try using it as it came with a 3 15/16" width and an off set of 3/16". This wider arm may fit the slots, time will tell, hope to work on it tomorrow. I purchased some 30lb monofilament fishing line to use for testing the geometry only.
By the way how old are the grand kids? Mine are 17 [Noah] and Josh 21. Josh is engaged to a wonderful girl with marriage planned next year as Joss is on military assignment till then.
Also where is Montgomery TX as related to Houston?
Again, thank you Charles for your assistance.
Gary
I really appreciate all the work you have done.
The only way I can get an off set at the servo is to use an aluminum arm, the one I planned to use is a DuBro twin arm that the swivels are mounted between, can't have any off set with this one. Before I drill the alum arm at 3 1/8" wide with a 1/4" off set, I am going to try using it as it came with a 3 15/16" width and an off set of 3/16". This wider arm may fit the slots, time will tell, hope to work on it tomorrow. I purchased some 30lb monofilament fishing line to use for testing the geometry only.
By the way how old are the grand kids? Mine are 17 [Noah] and Josh 21. Josh is engaged to a wonderful girl with marriage planned next year as Joss is on military assignment till then.
Also where is Montgomery TX as related to Houston?
Again, thank you Charles for your assistance.
Gary
12-23-2014, 12:42 PM
#58
Senior Member
Roo Man
I'm not really familiar with the DuBro twin arm, but there are many servo arms out there that cannot be offset. I've never used this one, but it is close to what you want and I have heard good things about their product. http://www.wrongwayrc.com/index.php?...&product_id=96 Once you decide on the parts you are going to use, let me know what they are as well as their dimensions and I will correct the drawing to accommodate them.
The grands are 13, 11, 9, & 8. Quite excited to spend time with Grandpa this close to Christmas.
Montgomery is about 50 miles northwest of Houston. By the way, I am originally from Sparta, TN.
Merry Christmas
Charles
I'm not really familiar with the DuBro twin arm, but there are many servo arms out there that cannot be offset. I've never used this one, but it is close to what you want and I have heard good things about their product. http://www.wrongwayrc.com/index.php?...&product_id=96 Once you decide on the parts you are going to use, let me know what they are as well as their dimensions and I will correct the drawing to accommodate them.
The grands are 13, 11, 9, & 8. Quite excited to spend time with Grandpa this close to Christmas.
Montgomery is about 50 miles northwest of Houston. By the way, I am originally from Sparta, TN.
Merry Christmas
Charles
12-23-2014, 01:58 PM
#59
Charles,
I just installed the rudder permanently and may have some unplanned off set at the rudder. The DuBro arm [pictured] has various center widths available, 3 1/8", 2 5/8", 2 1/8",
and 1 5/8". Will try some of the options at the servo arm and will let you know. Will also try matching the rudder centers too. Also the narrower centers may allow the crossed cables to pass through the slot without rubbing,
Merry Christmas to you and yours,
Gary
I just installed the rudder permanently and may have some unplanned off set at the rudder. The DuBro arm [pictured] has various center widths available, 3 1/8", 2 5/8", 2 1/8",
and 1 5/8". Will try some of the options at the servo arm and will let you know. Will also try matching the rudder centers too. Also the narrower centers may allow the crossed cables to pass through the slot without rubbing,
Merry Christmas to you and yours,
Gary
12-26-2014, 01:07 PM
#60
Moderator
That arm will work fine, and you will make life very easy on yourself if you use a DuBro threaded double control horn on the rudder. With the threaded horn, you can mount the horn back into the strong wood of the rudder and easily get the attachment points exactly even with the hinge line. Then you can tweak the length to get it exactly right. If you use any kind of non-adjustable control horn, it will be a guessing game to get the geometry right.
01-08-2015, 10:23 AM
#61
An easy way to not have to worry about arm lengths, Ackerman at the servo end and relative lengths of servo arm versus the length of the surface tab is to use pulleys at the servo end. A pulley will always pay out at one side the amount it pulls in on the other line. This eliminates the problem with the unequal changes you get with a straight arm when you have unequal lengths at either end of the pull-pull system. If you use pulleys, you just have to make sure that you do not have the attachpoints at the surface being moved forward of the pivot point. It can be a bit aft with no problem as you will get a minor bit of slack on the line not being pulled as you move off center. If you can get ahold of an old Model Airplane news (May 1963) you will find an article on making your home made pulleys an easy task.
This is how I do my pull/pull setups. Never had any problems. Another advantage of the pulley setup is you don't have to worry about putting a load on the servo due to the cable tension.
carl
01-08-2015, 10:49 AM
#62
Gentlemen,
I am very sorry that I have not been active on this thread but my wife has just gotten over a bad case of chest congestion and I have been having problems with my artificial hip and God given knee just below it. Saw Dr. Tuesday and he said nothing critical just a combination of too much wood cutting, cold weather and good old arthritis.
Hope to get back to the plane after my wife and I return from FL the end of next week. We are planning to relocate to Lakeland FL first half of this year.
Can any one give me any info on the RC plane community there?
By the way I want to thank all of you that have contributed in helping me set up my first Pull-Pull system. When I return I am going to try Lone Stars last suggestion of using a small offset on the servo arm at 3" centers. I have an aluminum arm that I can modify. I think with a little tinkering I can get it to work just fine for my simple flying style.
I will let you all know how it works out.
Gary
I am very sorry that I have not been active on this thread but my wife has just gotten over a bad case of chest congestion and I have been having problems with my artificial hip and God given knee just below it. Saw Dr. Tuesday and he said nothing critical just a combination of too much wood cutting, cold weather and good old arthritis.
Hope to get back to the plane after my wife and I return from FL the end of next week. We are planning to relocate to Lakeland FL first half of this year.
Can any one give me any info on the RC plane community there?
By the way I want to thank all of you that have contributed in helping me set up my first Pull-Pull system. When I return I am going to try Lone Stars last suggestion of using a small offset on the servo arm at 3" centers. I have an aluminum arm that I can modify. I think with a little tinkering I can get it to work just fine for my simple flying style.
I will let you all know how it works out.
Gary
01-08-2015, 11:29 AM
#63
Senior Member
One way to eliminate the load altogether is to put a pulley on the rudder as well as the servo. This will assure that the cable lengths remain constant. In my opinion; however, it's more trouble than it is worth.
01-08-2015, 11:49 AM
#64
Gentlemen,
I am very sorry that I have not been active on this thread but my wife has just gotten over a bad case of chest congestion and I have been having problems with my artificial hip and God given knee just below it. Saw Dr. Tuesday and he said nothing critical just a combination of too much wood cutting, cold weather and good old arthritis.
Hope to get back to the plane after my wife and I return from FL the end of next week. We are planning to relocate to Lakeland FL first half of this year.
Can any one give me any info on the RC plane community there?
By the way I want to thank all of you that have contributed in helping me set up my first Pull-Pull system. When I return I am going to try Lone Stars last suggestion of using a small offset on the servo arm at 3" centers. I have an aluminum arm that I can modify. I think with a little tinkering I can get it to work just fine for my simple flying style.
I will let you all know how it works out.
Gary
I am very sorry that I have not been active on this thread but my wife has just gotten over a bad case of chest congestion and I have been having problems with my artificial hip and God given knee just below it. Saw Dr. Tuesday and he said nothing critical just a combination of too much wood cutting, cold weather and good old arthritis.
Hope to get back to the plane after my wife and I return from FL the end of next week. We are planning to relocate to Lakeland FL first half of this year.
Can any one give me any info on the RC plane community there?
By the way I want to thank all of you that have contributed in helping me set up my first Pull-Pull system. When I return I am going to try Lone Stars last suggestion of using a small offset on the servo arm at 3" centers. I have an aluminum arm that I can modify. I think with a little tinkering I can get it to work just fine for my simple flying style.
I will let you all know how it works out.
Gary
There's a lot of events within a 2 hour drive between Tampa, Orlando, Melbourne, Jacksonville, Ocala etc
01-12-2015, 09:25 AM
#68
Senior Member
Pull-pull set ups are easy. Firstly: the distance between the openings on the horns for cables, uncrossed or crossed, MUST equal that of the servo horn. Period. No 'ifs', 'ands' or 'buts'! Secondly: the pivot point for the horn has to lie within the same line of the cable attach points in order to maintain constant tension on cables throughout range of motion. For uncrossed cable systems, if the distance between cable openings each side is 'X' then the distance between pivot points will be 'X'. The distance 'X' will remain constant throughout the range of servo motion. For Ackerman differential in uncrossed cable systems the pivot point of the servo and/or control horn point MUST lie at a distance of '>X'. It cannot be '<X' or binding will occur. Personally, I like to use a 'tiller' (walking beam) driven by the servo. A servo always has more range of motion than a control surface needs. To me, it is easier to set up and any time I need to remove a servo , de-rigging isn't needed.
Crossed cable systems are just a bit more complicated. Without getting too 'wordy' as in above (sorry) the pivot points cannot be in the same line as the cable attach points or binding WILL occur. An uncrossed cable emanates from its attachment point at an angle of 90 degrees at neutral setting. A crossed setting will have an angle of less than 90 if the horns' pivot set up is like one for uncrossed cables and no Ackerman differential. The easiest thing to do is to draw your full scale model's crossed set up on paper and then take measurements from there. At neutral settings the cable route line and horn pivot point to cable attach point line should form an angle of 90 degrees for 'no Ackerman' or slightly greater than 90 degrees for 'Ackerman'. You will see that in both cases the pivot point will be '>X'. Draw it out on paper and it will be easier to visualize. Good luck!
FWIW, I use cables almost exclusively in my models. See Gallery for examples.
Crossed cable systems are just a bit more complicated. Without getting too 'wordy' as in above (sorry) the pivot points cannot be in the same line as the cable attach points or binding WILL occur. An uncrossed cable emanates from its attachment point at an angle of 90 degrees at neutral setting. A crossed setting will have an angle of less than 90 if the horns' pivot set up is like one for uncrossed cables and no Ackerman differential. The easiest thing to do is to draw your full scale model's crossed set up on paper and then take measurements from there. At neutral settings the cable route line and horn pivot point to cable attach point line should form an angle of 90 degrees for 'no Ackerman' or slightly greater than 90 degrees for 'Ackerman'. You will see that in both cases the pivot point will be '>X'. Draw it out on paper and it will be easier to visualize. Good luck!
FWIW, I use cables almost exclusively in my models. See Gallery for examples.
I also disagree with your analysis of crossed cable systems. I drew it out on paper and the trigonometry just doesn't work.
Ackerman - Smackerman. I have never understood this reference, maybe that is my problem. Perhaps you could explain his/her theory as well.
Maybe I just wasn't able to follow your description. Perhaps you could use 'gt' for greater than, 'lt' for less than, 'ge' for greater than or equal, 'le' for less than or equal, and 'eq' for equal. That would avoid the issues with the lazy V. Also, if you could post some drawings, it might help.
Thanks
Charles
01-13-2015, 09:13 AM
#70
Senior Member
Look up some of Jim Newman's drawings and guides. He was a regular contributor to the 'Model Airplane News' magazine. They are very straight forward. So long as you adhere to basic principles as I have outlined pull-pull cables set ups are... easy! I'm sorry you don't understand what I've posted but it really is correct. Whether you agree or disagree with me doesn't change the accuracy! Good luck!
I suggest that you try the very analysis that you suggested on the crossed cables. See if the cable lengths stay constant over a typical range of motion. If you would, please present your findings here. I have tried your analysis (as I understand it) and can't get it to work out. I would like to know where I might have gone wrong ---- or not. I can graphically post my findings if you like, but since I may not have understood your premise, I would prefer to see your example.
Thanks in advance for your help.
01-14-2015, 07:30 AM
#72
Moderator
Ackerman geometry was originally about solving the problem of building a 4 wheeled car that wouldn't drag one tire in a turn. Since the two front wheels actually follow a curved patch with different radii, it creates a geometric problem to make both of them point in the right direction throughout the steering wheel's range of motion. Essentially, the inner wheel has to turn more than the outer wheel does both right and left. The angles in the linkages for the front wheels cause that to happen, and getting them exactly right is the key to building a good handling vehicle that doesn't wear out front tires quickly.
01-14-2015, 01:03 PM
#73
Senior Member
Jim Newman was a pretty awesome illustrator... too bad you missed his works! This applies for crossed and uncrossed cables: when controls at neutral setting the line formed by the pivot point of the horn to the cable attach point and cable run must make an angle equal to or greater than 90 degrees otherwise it will bind (or be immovable, theoretically, for a rigid set up). If you sum up the four inside angles of your set up they have to be equal to or greater than 360 degrees. If the angles sum to 360 degrees then the cables stay taught throughout their range of motion. If more than 360 then one cable starts going slack as the other 'pulls'. That is where the 'Ackermann' differential reference originates although the definition of the term was originally applied to autos.
One thing that is helpful is to use scrap wood and pins to make 'models' (pun, lol) of your set up desired. Use scrap wood for the 'cables', too. I often do testing empirically to 'flesh out' my ideas! I'm not as smart as most folks think... even though I have my doctorate from Texas!
Hope this helps!
One thing that is helpful is to use scrap wood and pins to make 'models' (pun, lol) of your set up desired. Use scrap wood for the 'cables', too. I often do testing empirically to 'flesh out' my ideas! I'm not as smart as most folks think... even though I have my doctorate from Texas!
Hope this helps!I have used AutoCAD to prepare these drawings because of its ability to do the trigonometry. It not only simplifies the effort but also enhances the accuracy of the drawing. Also, I have dimensioned the drawings to four or five significant digits in order to illustrate a point, not to suggest that any of us will build to that degree of accuracy.
Let’s get started with your first statement:
As I see it, this is just not true. It is true that on parallel (uncrossed) cable systems it is probably the easiest to set up and usually works well; however, I have seen many systems that do not follow this rule and they seem to work just fine. Your rule is not valid on crossed cable systems either.
This is a very simple Parallelogram, Pull-Pull set-up. At neutral, all of the arms are of the same length and all angles between the cables and the arms are 90º.
When deflected 45º, the cables remain taut, just as expected. Everybody is happy.
This is the same drawing as above, except that I have offset the upper control horns 0.5000 units aft. If you will look closely, you will notice that not only are the arms of different lengths but also that the pivot point does not lie on a line between the two cable connect points nor is it equidistant from those points. You should also notice that the offset in the upper servo arm is matched by an equal offset in the rudder control horn.
Again, when deflected 45º, the cables remain taut. Maybe this not the result expected, but everyone is still happy.
In this drawing, I have created a parallel (uncrossed) cable system in which the distance between the pivot points (24.0000) is greater than the cable length (23.5000). As you can see, when the servo is rotated 45º, the lower cable will be 0.2994 units too short. If there is no flex in the system, binding is liable to occur. This example does not agree with your theory; therefore, theory is not valid.
Obviously, this analysis could go ‘on and on’, but suffice it to say that I think your conclusions about crossed cable, pull-pull systems are worse than your uncrossed analysis.
Of course, I could have misinterpreted your comments or I could be wrong. I look forward to your further comments and analysis.
Charles
Last edited by Lone Star Charles; 01-14-2015 at 01:13 PM.
