# Heavy duty door, automatically operated

## Main Question or Discussion Point

Hi, I am trying to design a 2000kg door that has dimensions of 3m wide X 4m high and will be automatically swing door operated, for a college project. I have come up with a relatively straightforward design using a motor and gears to open the door, and to close it again with a spring. I have recently moved on to the calculations section which will help me to choose my motor and its operating speed etc.

This is where I am stuck. I am finding it very difficult to come up with the torque and overall force needed to be exerted by the motor. I want to include a factor of safety also to provide for external forces such as people pushing or pulling or hanging off the door. So far I have come to the conclusion that I will use two extra duty hinges for the door.

Each hinge will have a vertical and horizontal force component. I am wondering if I should take the top hinge to bear all of the weight, or both hinges take half. And how do I calculate the torque and horizontal forces. I know torque is F.d.sin(angle), but is that all I need to do? and I know the horizontal forces will be calculated by taking moments about each hinge, but I'm not sure if the answer i'm getting is correct.

Could somebody please point me in the right direction, I have spent hours on this and made little progress. Thanks you

Related Mechanical Engineering News on Phys.org
AlephZero
Homework Helper
For something that massive, I would stop thinking about "hinges" and think more about "bearings". In that case, you need a thrust bearing to take out 20kN weight, and a roller bearing to take the side load. Whether you put the thrust bearing at the top or the bottom depends on the practical details of how you are going to install it, for example how you would get access for some heavy lifting equipment.

With ball or roller bearings you could easily make the friction low enough (typical friction coefficient = 0.001) so the size of the motor depends on the inertia of the door, not the friction at the hinges.

BTW, a spring strong enough to close a 2-tonne door is going to be one very impressive looking spring. And there may be one very impressive bang when the door hits the door frame...

First of all thanks you very much for the reply. I really appreciate the help.
So basically ur saying the ball bearings are going to "bear" the full brunt off the doors load right? So one hinge will support the load of the door. One thing I am stuck on is how I am going to open and close the door. I think that the mechanism for opening it will vary greatly from that of the normally automated swing doors. If i used a hydraulic pump (piston-cylinder) attatched to the motor, and exerted the opening force on the furthest edge of the door, that would open it easily enough. But I still would have to regulate the speed of the doors movement upon opening, bearing in mind that I have to keep my design as simple as possible. I assume there is a way to regulate the speed with latches and certain hinges, etc. Finally to close the door, i think u hit the nail right on the head, in saying that the spring is a bad idea. I need a new concept for the closing stage. Possibly having a second motor to close the door, and a dampener on the frame to avoid banging and unnecessary friction? Or else i could have a linking mechanism on the door, hooked up to the motor, that could open and close the door, using just the single motor.

I'm pretty confused though on this matter. I know that once I get over this hurdle of the design phase, I will be able to dive into the calculations and work out my speed-torque characteristic, gear ratio, angular velocity, etc.

Why isn't all the movement being done with hydraulics, or at least hydraulically assisted?

Also how wide is the door opening? How fast does it have to open? How is it opening, is it by someone pushing and the system assists it or totally automatic? Also how large can the mechanism be to open it, is there a space, size, etc operating constraint?

Ooh, and also, is this purely a paper excercise? (probably the most important)

EDIT: One thing you'll learn with engineering is that the first thing many will do is ask you a shed load more questions.

Last edited:
AlephZero
Homework Helper
If it helps you get started "thinking like an engineer" about this, the first thing I did was get some idea what the door that size might look like. A quick calculatiion said a solid steel plate 3m x 4m x about 23mm thick would have about the right mass. Now I have that picture of something "definte" in my mind (not just the 3 numbers in your specification), I can start thinking about what to do to make it work like a door....

First of all thanks you very much for the reply. I really appreciate the help.
So basically ur saying the ball bearings are going to "bear" the full brunt off the doors load right? So one hinge will support the load of the door.
In your OP you talked about whether to split the weight between "both" hinges or take it all on one hinge. A good general principle is "design something that is easy to analyse". That principle suggests making the loads on the hinge system statically determinate, rather than dependent on the manufacturing tolerances, the flexibility of the door and the frame, etc.

Of course you may find you can't do that in practice. For example you may need more than two hinges, because just two would be too big to fit in the available space, but starting with a simple idea and making it more complicated if you have to, is usually better than starting with a complcated idea without a good reason why it is complicated.

One thing I am stuck on is how I am going to open and close the door.
...snip...
I'm pretty confused though on this matter. I know that once I get over this hurdle of the design phase, I will be able to dive into the calculations and work out my speed-torque characteristic, gear ratio, angular velocity, etc.
You could work this backwards. Decide how fast you want the door to open/close. Then find the kinetic energy in the door when it is moving. Then you have an idea of how much power you need to start it moving and stop it again. So you can estimate the maximum power of your motor (whatever type you choose), and what force or torque it needs to apply to the door. Ignore friction to start with, because you know at least one way to build the door that will make all the friction forces negligible (i.e. use ball and roller bearings for the hinges). Again, it may turns out you can't ignore friction in your final design concept, but by the time you get to the final design, you will have something definite to analyse, not just some vaguely worrying thoughts.

Most likely your first "guess" will come up with numbers that are not reasonable in practice so you will have to try again. That is quite normal. Another general principle of designing things is that "at least 90% of the work goes into the trash can, not into the final design".

It's your project not mine, so I'm deliberately not telling you what my first thoughts about a design would be - but I hope this gets you started.

How about using a linear actuator to both open and close the door? Linear actuators have a good price to force ratio and would solve the problem of the door slamming shut.

This is purely a paper concept, so i need not worry about space, cost, etc. that much. I think thats a good idea for working backwards to find the angular speed. It's just difficult to picture how i will open it, and where the force will be exerted on the door. I was thinking of exerting a force on the furthermost edge of the door, so as, the least force will be needed there to drive the door closed.
As for the hinges, yes just having two hinges may mean that both hinges will have to be massive, however i found that when there is 3 or more hinges, the system is statically indeterminate and didn't really know what to do from there.
A linear actuator could be a good idea. I do not have any experience at all with them but i could certainly give it a go.
I think that the actual physical design is the hardest part, because i cannot really do many calculations unless i know what I am using. However some of the ideas I have been getting are extremely useful, it's helping me think "outside the box" a bit

would a electro-mechanical actuator connected to a power screw be able to open the door? An arm attached from the shaft to the door could rotate the door and the power screw would allow it to stop. Not sure yet how I will connect the motor to the power screw yet, possibly gears?

Steve,

I don't completely understand your concept. What I had envisioned looks like the attached image. The linear actuator will keep pulling/pushing for as long as a current is applied. The direction of the actuator (push/pull) is changed by swapping the voltage polarity which is usually done with a double pole double throw switch.

Once you know the weight of the door, you can select a linear actuator based on load capacity and stroke length to match to your mounting point and the weight of the door (lever arm, moment).

#### Attachments

• 5.4 KB Views: 1,254
yes that looks like a very good idea alright. A huge force will be required from that linear actuator to open the door but I think it could work. It is also very simple. Is the linear actuator mounted on some sort of pivoting system so that it can open the door to angles of 90 degrees and over? Also Any ideas for the inclusion of a locking system for this design?

My previous idea included a large shaft going parallel alongside the door, connected to an arm which had one end welded to the door and the other to the shaft. A rotary actuator would turn this shaft, thus turning the arm and opening the door. To close the door, this process would simply be reversed. This idea seems more simple though.

In the concept I described, the actuator would need to be mounted with brackets like those used for screen door closers.

A good place to look for available push/pull forces and stroke lengths is http://www.firgelliauto.com/" [Broken]. I've worked with them in the past and found them to be very helpful.

Something that might work for you is the 450 lbf, 30" stroke length actuator. They even sell matching brackets that allow for pivoting.

Unfortunately, you have a lot of things to still consider. The minimum force of the actuator is determined by the point on the door where it will be attached (Force x Distance = Torque). The torque must overcome the moment created by the weight of the door revolving about the hinges. The retracted length of the actuator and the mounting point on the wall are affected by both the torque and the distance the door must open.

#### Attachments

• 4.1 KB Views: 608
• 22.9 KB Views: 605
Last edited by a moderator:
I think i understand it better now after seeing those images you attached. I was also looking at doors powered by a linear actuator, that were connected by means of a linkage system, however it looks far too complicated to pursue. I calculated the torque by multiplying the distance from the edge of the door to the center of gravity by the weight of the door. I am using two large strap hinges on the door. However when calculating the forces on the hinges and the moments, it seemed to be statically indeterminate. So I assumed that both of the horizontal forces acting on each of the hinges were equal to each other.

it looked something like this. Each hinge has a vertical force going upwards, Va and Vb.
so Va + Vb = weight of door. there are also two horizontal forces on the hinges, Ha and Hb. Ha is the force on the top hinge. Ha is going to the right and Hb is going to the left as far as I know. So they result in Ha=Hb. Taking moments about A and B just resulted me in getting the following two equations. weight of door = -Ha, and weight of door = -Hb. I still amnt quite sure if that is correct. I could be leaving out some key force. It's just that this is the the fundamental analysis of the door and I must get this right before even considering torque or angular velocity. Sorry for asking so many questions, you have been so helpful so far, over and above what I though I would get.