Gearbox Construction
 

Looking at somethings online I was thinking about how to make a really good gear box, and especially after looking at those pics of the segway gearbox in thread I was wondering about a few things.

1.) What is the biggest difference between 14.5 and 20 degree pressure angle gears, and how do you determine what the angle on the gear is?

2.) Significance of helictal gears. Why use a gear with angled teeth?

3.) How do you make a gearbox with a high efficency?

4.) How do you determine the efficency?

5.) How do you make it run quitely?

6.) Correct spacing for gears?
6.1.)Is the paper spacing method sufficient for a sliding mount?

7.) Best type of bearing to use?
7.1.) how does a sealed differ from the rest?
7.2.) how does a shielded differ from the rest?
7.3.) How does a unshielded differ from the rest?
(eviroment of use and characteristics of use, which one is best for speed o rlow friction, ect...)

I have found some of the answers else where but the sources differed in the information for alot of these questions, so I decided to ask the experts.

Note: I may end up creating a white paper out of this information.

Re: Gearbox Construction
 

The pressure angle is the angle in which teeth between two gear contact each other. 14.5 pressure angle gears are smoother but have a lower load capacity. 20 pressure angle gears are less smooth but have a higher load capacity.
With the angled teeth helical gears have higher load capacity because there is more contact area between teeth. Also since the angled teeth gradually engage each other, helical gears run smoother and quieter. However since there is some sliding action between the teeth, there is some loss in efficiency. The efficiency also decreases as the angle of the gear increases. In addition to the loss of friction, the angled teeth also create axial thrust forces so you have to use thrust bearings with them and it also creates more efficiency losses.
Minimize the number of stages of reduction through gears and chain because you lose efficiency with each interaction between gears or between chains and sprockets.
You can use benchmarks to determine the efficiency. I know that some people use an overall gearbox efficiency of 81%. Others prefer to determine the efficiency through each stage of reduction. Some benchmarks I've found are:
Spur gears - 95%-98%
Chain/Belt - 95%-98%
Bevel gear - 90%-95%
Lower pressure angle and lower diametral pitched gears tend to run smoother and quieter. I've also read that adding .002" to the center-to-center distance between gears allows the gears to run smoother and quieter. Lubrication also helps.

Sorry, but I don't know the answers to your last two questions, but I hope this helps you. I'm also working on designing a custom gearbox for my team and its great to see other people interested in this type of thing too. Good luck with your gearbox project!!

Re: Gearbox Construction
 

Really quick answers to what I know. If Joe Johnson happens across here, I hope I'm not embarassed too badly.

1. Without getting too complicated, 20 PA teeth are thicker and somewhat stronger. Noisier and more prone to backlash as well.

2. Helical gears are stronger, quieter, and more efficient when well lubed because the gear gradually engages. They're also more expensive, and develop a thrust load that you have to counter with good bearings.

3. High efficiency comes from using the fewest gear downs possible, mostly. You lose efficiency from the gear teeth meshing, bearings, etc. You lose a LOT if the gears are too close and bind. Of course if they're too far, it's noisier and you have more backlash.

4. After it's built, you can measure power input vs power output. There's a team I can't remember that will test your drivetrain for you.

5. Tolerances. The tighter your tolerances, the less backlash, the less noise, the higher the efficiency. But it's tolerances that matter. If you have bad tolerance control and try to cut the center distances too close, you can end up binding the gears. Why do you care how quiet it is? It will scarcely drown out the music on the field...

6. Assuming 0 tolerance, add the pitch diameters and divide by 2. In practice, you can add .002" to allow for machining and it should cover you well.

6.1 I've no idea what you're asking... if you want to know a sliding shaft fit... ISO hole/shaft fits are metric fits. I highly recommend getting your hands on any copy of Machinery's Handbook. A paper thickness would leave you with a whole lot of play.

7. Ball bearings? Flanged might make assembly easier, but the choices are limited. Unflanged means you have to worry about press fitting or otherwise capturing the bearings.

7.1 Sealed are well sealed againt water, dust, etc. with rubber typically. The seal holds grease in the bearing so they don't need maintenence.

7.2 Sheiled have metal shields to hold in grease and keep out the worst of any debris. Also maintenence free.

7.3 Open are just that. the balls and races are exposed to air. you'll probably need to grease them occasionally, and the balls and races could get dirty. If you're running very high RPMs you could use oil in open bearings instead of grease.

7.1,2,3 Other than that, they're effectively the same for FIRST purposes in terms of torque, etc.

Again, you may want to invest in a copy of Machinery's Handbook either hardcopy or on CD. It's terribly useful for all sorts of things. Anyways, I'm gonna give my poor fingers a rest now....

Re: Gearbox Construction
 

I believe its 494 and their dynomomiter

Re: Gearbox Construction
 

Open bearings are traditionally not as wide as shielded bearings. A bonus for some applications such as dog gears in a shifting transmission.

Re: Gearbox Construction
 

Traditionally, yes. I've run across some bearings that use the same races for open and shielded bearings, though. So there's no difference at all. Just make sure you look before you leap. :)

Re: Gearbox Construction
 

A simple test is to measure the motor current required to spin the gearbox, and compare it to the motor's free-spin current. Be sure you are using the same voltage for both measurements. This test will give you a idea of the (unloaded) frictional losses; just multiply the difference in currents by the voltage you used to get the frictional losses in units of Watts.Sealed bearings may also add significantly to the frictional losses, esp. when the bearing diameter is large.