backdrivable gears

student_ucsd

Member
Join Date
Dec 2006
Location
san diego
Posts
9
hi everyone i was just wondering if anyone knows about how backdrivables gears work. what i mean is when something is backdrivable and when it is not based on a equation or the amount of friction created by the gears. thanks for your time and i hope to hear from you guys soon
 
I may be wrong on this but I don't think this is very easy to quantify. First of all the gearing type has a big effect on this. A 20:1 helical bevel gearbox will be much easier to backdrive than a 20:1 worm gearbox. Internal to any given gearbox type, gear face finish, lubrication type and tooth clearance will all have an effect. In addition ANY vibration will significantly reduce the required backdrive force in any type of gearbox. These are just the first factors that come to mind. I'm sure there are many others.

I don't know of any straightforward equation that will give you a practically usable value.

Keith
 
backdrivable gear boxes "DONT WORK" when driven backwards. I assume I understand what you are asking. The gear boxes that can be driven backwards due to tension etc are not exactly "driven" backwards but sustain "energy" like an electrical coil; so the force it has sustained is released in the only direction that has no "load" i.e. backward.

It is xmas, think of a toy soldier, as you wind it you WORK and expend energy to wind it, when released it runs around. This is a direct example, sometimes in a system you "drive" something by turning gears and they also build up energy that will be released "opposite" of or backwards from the original motion.
 
Last edited:
Just think of it like a spring mattress, lay on it and it attempts to conform to your body, get up and it goes back to original position. JUST REMEMBER, for every action there is an equal opposite reaction.
 
Last edited:
i think your answering a different question. when a geared motor is backdrivable then it is able to go in both directions without the motor running. if it is not backdrivable and you try and turn the shaft it wont turn because the gears cause soo much friction that it wont move. im asking eqautions govern this.
 
Let me explain the application which is the source of my question. We have to design a robotic arm for an industry sponsored project and our mentor has asked us to research backdrivable servo systems. The example he gave us was the CD-Rom tray in your PC. If its open and you push on it, you get initial resistance and then it goes back in on its own. The portion of that we are interested in is that initial resistance, not so much the fact that it goes back in.

From what we have gathered via literature on the net, backdrivable simply refers to any gear set that can be forcedi n the opposite direction. The opposite of this being something like a worm gear. So, we are asking if someone can explain, or at least point us in the right direction, on how we could design our own backdrivable servo that could be applied to rotating an arm in 1-D and would be able to survive forces such as children grabbing and pulling on it often.

Sorry for any spelling errors and thanks again!
 
Your instructor should explain "what he wants to hear" if it is not specifically in the books provided for the course.

The CD rom example is not an example of backdriven gears!

I should not but will explain, backdriven is applied to situations where devices go one direction but "forces" drive them backward.

A CD rom "tray" goes in or out, it uses "technology" to "sense" when the tray has been pushed "in".
 
Last edited:
You are asking for something that, to the best of my knowledge, gearing manufacturers don't design or test for and don't publish data for. All you can really do is speak in generalities.

I would be very surprized if you find what you are looking for. You may need to develop these numbers on your own if you really need to know. If you want it to be 'easy' to backdrive stick with helical bevel or planetary gearboxes at ratios of 10:1 or less. You can backdrive 20:1 to 50:1 but that can take some effort.

Your CD tray example may be a bit misleading. I've never really dug into it but I think that initial relatively high force isn't necessarily due to anything inherent in the gearing. I think it is a lay-out artifact. Unless you specifically design for this trait you probably won't get it.

Keith
 
I should have been a bit more clear on my definition of back drivable, but yes I understand that they want to go one way and the force is applied in the opposite direction.

It sounds like we're in trouble on this. We need to move an arm and allow for random forces to be applied on the mechanism without it failing, and it needs to be very safe. Any sugestions on how to drive such arm? So far we have explored the basics of servos and some micro-pneumatics. The servos used failed after two days due to too many of these resistant forces applied on the arm, and the pneumatics arent safe enough for our project.

Thanks for the help.
 
student_ucsd said:
I should have been a bit more clear on my definition of back drivable, but yes I understand that they want to go one way and the force is applied in the opposite direction.

It sounds like we're in trouble on this. We need to move an arm and allow for random forces to be applied on the mechanism without it failing, and it needs to be very safe. Any sugestions on how to drive such arm? So far we have explored the basics of servos and some micro-pneumatics. The servos used failed after two days due to too many of these resistant forces applied on the arm, and the pneumatics arent safe enough for our project.

Thanks for the help.

WHat you have to determine is "backlash", different subject.
 
Its funny that when I post on forums similar to this yet intended for video game discussion I get similar non-informant, seeminly childish responses except on those boards they are actually from children. I find it odd that a person, who I assume is some sort of professional, would take the time to read my post and respond numorous times without saying more then one bit of usefull information, all the while trying as hard as possible to act smarter than a student.
 
Last edited:
student_ucsd said:
Its funny that when I post on forums similar to this yet intended for video game discussion I get similar non-informant, seeminly childish responses except on those boards they are actually from children. I find it odd that a person, who I assume is some sort of professional, would take the time to read my post and respond numorous times without saying more then one bit of usefull information, all the while trying as hard as possible to act smarter than a student.

There is no answer because there has not been a viable question yet.

I reckon I be childish now and not answer any more, cuz I am good at that.

BTW: YOU have a chance to obtain something I never did, do not be assinine in the process. So far you asked a question about something that had no answer but when answered best as possible got a "tude" cuz did not like the answer. Take what you get and expound on it, if not capable then ask the teacher, best I can offer.
 
What is the question again?

Can you please just explain what you want the arm to do when it's pulled at? You want it to move? Or stand still? Do you have mechanical breaks? Electrical break?

My first idea for a childsafe 1-Dimensional robotic arm would be to have a very low torque-limit on the servomotor, and possibly either apply or release all breaks when that torquelimit is reached (depending on what you want the arm to do).

 
studentuscd, in the motor-gearbox world that I live in, backdriveable simply means that the load side of the powertrain can transmit torque backward thru it to the motor. This commonly occurs when using a motor or motor-brake to decelerate a high inertia load. It can also occur when the load side of the system is being "forced" mechanically, as in taking a robot arm and pushing it with your hand into a position with the motor power off or, as in pushing a manual transmission car to start the engine. I think this is the meaning you are presenting here.

In that case, most common gearing, spur, helical, planetary, will backdrive. In fact, occasionally such gearing is installed backward to get speed increases rather than decreases.

The big exception to the above statement is worm gearing. The general rule that we follow is that a worm gearbox is backdriveable at a ratio of 25/1 or less. Above that ratio, the gearbox simply locks up and, if forced, will break either the shaft or the casing.

You will note in some of the above posts that there are a lot of variables that affect actual field performance---lubrication, temperature, gear tooth finish, vibration, shock loading, etc.

Bottom line, the gearbox manufacturer is going to be your best source for advice on backdriving since it their warranty to protect.

As a student, you will find, as you grow into your chosen field, that the manufacturer is often the first place to turn.

Good luck on your studies.
 

Similar Topics

What happened to the gears/ladder on Studio 5000?
Replies
1
Views
1,908
Hello, everyone I'm new to this put from what i've seen here you guys know your stuff. I'm in college learning about Automation systems, and I...
Replies
3
Views
1,816
Back
Top Bottom