[youthreewire]

HI,I am new to mechanical design.I want to figure out the match required for making a four stage spur gear box.(I read basic spur gear terminology).I require a gear reduction of 254.How shall I figure out the the individual spur gear dimensions?Please guide me.(Thanks in Advance).Also the final o/p torque is 1.6 Nm.

(Thanks in advance)

[greg b]

Its a bit more complicated than that as you have probably found out,I guess we will need to know what size you want to make the gearbox as it may need to fit into a tight space ,or awkard position which will limit the size of the gears and there shafts.Do you want to work in the metric or imperial system,254/1 is quite a reduction if i am understanding you properly.

[LeeWay]

There are lots of pictures of lathe gear setups floating around here. I can't give you any specific gear info except that such a reduction is complicated as you already know.

I think the way it should be setup is a small gear drives a large gear. another small gear is attached to that large gear and drives another large gear. That is a pretty good reduction right there, but you will need to figure out tooth count to get it precise.

[youthreewire]

I want to make a compound spur gear train which would fit into 5 x 2.5 area and height can be minimal not exceeding 1 cm.I was looking at this page of compound gear trains.
http://www.roymech.co.uk/Useful_Tabl...ve_trains.html
But tricky part is finding the no.of teeth on the o/p gear circle and i/p gear circle of a compound spur gear.Also as the gears move towards the o/p shaft in increasing order the Diametrical pitch also should increase to give tooth strength (because I think though one might get a larger reduction by increasing the no.of teeth on the gear with same diameter/radius the tooth size decreases and and its tendency to fracture /ware out also increases).

[One of Many]

You3,

Is this for the R/C servo you wish to design in another thread?

I am involved a bit in R/C myself and have thought about the gear train in many of the designs I have come across. Especially the smaller servo cases. Such tiny gears for what they do.

A few thoughts I had were; worm gear ratios are both high torque and compact, but the problem is high speed and plastics don't mix. Then I had some older servos that were large cased, in which they used gearing to a screw and true linear actuation. Then I thought about planetery gearing which needs very high speed to get torque. Each trade off brought in other negative issues.

I keep going back to the screw idea if there were a better micro stepper controller with feedback, but torque requirments are doubtful in small stepper motors. The mechanism that drives a DVD or CD laser looks inviting, only if it could handle more load.

The old tried and true systems still prevail. Replacing the plastic gearing with metal offers the least amount of re-engineering.

DC

[HuFlungDung]

The overall ratio is 1/254

You can use that fraction as a basis to create a gear set by using common multiples top and bottom.

To start with, a 1 tooth gear is basically a worm, and that is actually the most likely option.

For spur gearing a minimum tooth number might be 10 tooth, so that means that a single reduction would have to be a 10 tooth driving a 2540 tooth gear.

For a double reduction, the smallest pair of drivers would be two 10 tooth gears:
1*10*10 / 254*10*10

Since 254 has only 127 as a factor, you can write it out thus:
1*10*10 / 2*127*10*10

You can then combine the factors in various combinations: let's say 127 and 200 would be the driven gear tooth numbers, with two 10 tooth drivers.

A 200 tooth gear fitting in a 2.5" space is 200/2.5 = 80 Diametral pitch! That is a tiny gear tooth.

You take it from there

[One of Many]

Good info HUflung!

But the posted envelope is 5cm x 2.5cm!


DC

[HuFlungDung]

Yup, I see that now, but he wanted a 4 stage reduction anyways, so maybe he can make it smaller with enough gears in it......something like clockwork.

[One of Many]

Yup, that is true. Also at 254:1 at least the input will only need to be 1/254th of the output, not counting the loss in the gearing etc. provided the teeth are strong enough to handle it.

DC

[youthreewire]

The overall ratio is 1/254

You can use that fraction as a basis to create a gear set by using common multiples top and bottom.

To start with, a 1 tooth gear is basically a worm, and that is actually the most likely option.

For spur gearing a minimum tooth number might be 10 tooth, so that means that a single reduction would have to be a 10 tooth driving a 2540 tooth gear.

For a double reduction, the smallest pair of drivers would be two 10 tooth gears:
1*10*10 / 254*10*10

Since 254 has only 127 as a factor, you can write it out thus:
1*10*10 / 2*127*10*10

You can then combine the factors in various combinations: let's say 127 and 200 would be the driven gear tooth numbers, with two 10 tooth drivers.

A 200 tooth gear fitting in a 2.5" space is 200/2.5 = 80 Diametral pitch! That is a tiny gear tooth.

You take it from there

I understood this idea.But the RC gear servos are not simple spur gears.They are like some sort of compound spur gears (where each compound spur gear is made of two spur gears.One larger diameter and the other smaller diameter.The smaller diameter spur part drives the next compound spur gear in the train.The larger diameter spur part is driven by the previous compound spur gear.And the pitch of the spur gear increases as it goes closer in the chain to the output shaft.

(So I think the calculation is not as simple as finding factors of 254 and dividing it up into stages).

For example consider the Hitec -5995-TG here (or any other servo gear set).
http://servocity.com/html/hsr-5995tg_ultra_torque.html

I want to learn how do these people do the gear math for such RC servo systems.I would like to machine my own servo then.

[youthreewire]

You3,

Is this for the R/C servo you wish to design in another thread?

I am involved a bit in R/C myself and have thought about the gear train in many of the designs I have come across. Especially the smaller servo cases. Such tiny gears for what they do.

A few thoughts I had were; worm gear ratios are both high torque and compact, but the problem is high speed and plastics don't mix. Then I had some older servos that were large cased, in which they used gearing to a screw and true linear actuation. Then I thought about planetery gearing which needs very high speed to get torque. Each trade off brought in other negative issues.

I keep going back to the screw idea if there were a better micro stepper controller with feedback, but torque requirments are doubtful in small stepper motors. The mechanism that drives a DVD or CD laser looks inviting, only if it could handle more load.

The old tried and true systems still prevail. Replacing the plastic gearing with metal offers the least amount of re-engineering.

DC

I want to make my own RC servo.I do not want to replace the plastic with metal ones because this would fail if I have to make my own specific servo.I want to learn the geat train math of an RC servo (How do they figure out such gear ratios and DPs)

[HuFlungDung]

Youthreewire,
What I wrote is exactly relevant. A spur gear is just a toothed wheel, it does not designate anything about ratios.

A compound gear setup is one in which a driven gear is coupled to another driver on a common hub so that they turn at the same rate. So with a 2 stage reduction you begin with a single 10 tooth driver, engaging a 200 tooth driven which shares a common hub with the second 10 tooth driver which drives the final 127 tooth driven gear which is affixed to the output shaft.

That is what the formula Product of Drivers/Product of Driven = Ratio means.

You can expand the ratio with whatever multiples you want, so long as you do the same thing in both numerator and denominator.

Working with the prime factors of the ratio gives you the most options to build custom gear sizes that are allowable: for example the combining of 2*10*10 gives me a 200 tooth gear but it would also allow for a 20 tooth and a 10 tooth gear (along with the 127). Or you could use a 256 tooth gear and a 100 tooth, mix and match.

Note I did not work with the prime factors of 10, but one certainly can do so because the 5 and the 2 permit you to imagine more gear choices. But in practical use, you would need to make sure that no gear has less than 10 teeth for sake of smooth running.

Get an old gear clock and open it up and have a look.

[flick]

Working with the prime factors of the ratio gives you the most options to build custom gear sizes that are allowable: for example the combining of 2*10*10 gives me a 200 tooth gear but it would also allow for a 20 tooth and a 10 tooth gear (along with the 127). Or you could use a 256 tooth gear and a 100 tooth, mix and match.

Just to expand on this, is the exact ratio 254:1 required? (perhaps a metric/imperial conversion geartrain?) or would something close to that ratio be acceptable? As Hu pointed, working with the prime factors is the best method of coming up ratio combinations, and it so happens that 256 (2^8) has a lot more prime factors than 254.

Also, this may help to optimize the gears for your power requirements: http://www.rushgears.com/Tech_Tools/horsepower.php

[youthreewire]

I contacted Rush gear few days ago and sent few mails.But I did not get a reply from them till now.I see they have a great website and their webservice is also customer friendly.I sent a mail to their sale staff.I did not get a reply still.

Can anyone tell me why?(Is it common with them?).Or anyone who had any prior experience with Rush gears..Please let me know.

[youthreewire]

Just to expand on this, is the exact ratio 254:1 required? (perhaps a metric/imperial conversion geartrain?) or would something close to that ratio be acceptable? As Hu pointed, working with the prime factors is the best method of coming up ratio combinations, and it so happens that 256 (2^8) has a lot more prime factors than 254.

Also, this may help to optimize the gears for your power requirements: http://www.rushgears.com/Tech_Tools/horsepower.php

Even a closer number will help.In the future I want to learn making gear stages with different reduction ratios.