Should I avoid using high ratio spur gears?

[Jarfi]

So I have been looking into efficiency, My system is high radius with the motor located outside. This motor is small, and thus the gear being driven by the motor, is also small for space requirements.

I have wanted to use a ratio of about 22:1 to make this system, and everything seems to work fine, however it seems to be inefficient to use gear ratios of spur-gears above 1:6 according to some sources
http://www.meadinfo.org/2008/11/gear-efficiency-spur-helical-bevel-worm.html
http://www.roymech.co.uk/Useful_Tables/Drive/Gear_Efficiency.html

But according to other sources, a rack and pinion drive is about 97% efficiency. While at the same time, a rack and pinion can be seen as a spur gear with infinite ratio.

1:infinite, efficiency is 97% while 1:22, efficiency is bad as far as I understand. Why is this?

And what should I do, should I continue using small teeth and a 1:22 spur gears, or should I replace my small drive gear with something 2 times bigger, to create a ratio of f.ex 2:11, and then another 1:2 gear, this obviously weight much more, and I want to avoid having to do this.

 

[Dr.D]

If you don't like the result of 22:1 in a single step, consider using a multistep gear reduction, such as 3:1 and 7:1 which would get you 21:1.

 

[Jarfi]

I guess I should rephrase my question. Why is a rack and pinion efficient while a high ratio spur gear is not, when in fact they are the same thing?

 

[CWatters]

I think they are assuming that a high ratio spur gearbox needs multiple stages. If you have space to implement a high ratio in one stage I don't see why it should be inefficient.

http://www.machinedesign.com/archive/second-look-gearbox-efficiencies

Most manufacturers will specify an intended gearbox operating point. Gearbox efficiencies in a spur gearbox at a 16-mm diameter vary from about 87% at a gear ratio of 6.3:1 to about 40% at a ratio of 10,683:1. A basic rule that designers use for spur gears is a 10% loss per engagement. One gear wheel in contact with another is defined as an engagement and the loss in that engagement is approximately 10%.

 

[CWatters]

Just to add that... If you want to maximise efficiency overall you need to load the motor to it's most efficient operating point. That may mean changing the gear ratio. For example a high ratio gearbox reduces the motor torque required but on the other hand a motor with no load is 0% efficient.

 

[litup]

Besides, the physical sizes would be hard on the smaller gear. If you had a 22 to 1 ratio, and the large gear was say 100 mm in diameter with 220 teeth, the small gear would need to have only 10 teeth and besides efficiency problems it would create friction, heating, and reliability problems.

 

[Dr.D]

A lot of the friction in gears is related to gear surface quality and lubrication. The values cited by CWatters from Machine Design magazine look rather low to me, but my experience is mostly with very high precision gears. This could all be better addressed if we know what speeds and torques are being transferred and in what sort of application.

 

[Baluncore]

Where large diameter drums must be driven slowly, a flat belt or chain drive is usually considered.
Do you drive the axial shaft or the outside surface of a drum?
What is the output RPM of your system?
How big is the machine?
Is the environment clean?
How will you lubricate the rack?
How many will be built?