Afters Andys request, It has made me think about how to make a zero backlash worm drive.

As I have found in the past, you think you have a good idea only to find that it has already been done.
B4 I spend time drawing up this idea, does any one know of a zero backlash worm drive? Do they already exist?

The beauty of my idea is that it does not rely on fine tolerance machining.

Well lets hear about it, I have some dought that this is possible.
There are two kinds of frictions in a power transmission rolling and
sliding friction where rolling is clearly the lower and better.
A wormdrive is the worst of all with the sliding friction been the
dominant part of transmission.
Reduceing clearance or even preloading would significantly increase
the friction produced.
One way could see this happening is with the large wheel having
roller bearings with radiused running surface mounted perpendicular
to the shaft instead of the teeth.
Each Tooth would be replaced by two bearings slightly offset one making
contact with front the other back side.
The bearings would need to be very small in order to be able to have
tree teeth of the worm engaded at all times.
The precision involved in making this would be outside of the abilitys
of most hobby shops and likely be expensive as a industial product.
Unless its mass produced of course.

Benny:

I remember reading an article somewhere about a type of zero backlash worm drive that uses the same principal as a ball screw. It's not that both gears have grooves. The two gears have teeth that mesh with a ton of backlash, then several sets of captive balls are run between the contact points of the teeth.

I know I read about this (The article had pictures) long, long ago; even before the World Wide Web. To be honest I think it was in the same article where I first heard of ball screws. It was in a "Popular Mechanics" or "Popular Science" type of magazine. I know when I read it I thought "This will never take off", but then again, I also said "This will never take off" the first time I saw a color computer!

I will do a drawing of what I am thinking about, but in the mean time I will try to verbally desribe it.

Image you the worm wheel is split in half. The split is vertical, so esentally you have two thin wheels when they are taken apart. when the wheels are put together they form a working worm wheel. There is a spring inserted in the assembled wheel that forces the two halves to be rotated a few degrees out of sync from each other.

When the whole assembly is put together, it will put a load on both sides of the screw. This will illiminate any backlash. I guess the comprimise is that the pressure put on the screw would be increased a lot. The other thing is if you have and exsessive load to drive the spring load must be greater than the load the assembly is driving.

One advantage is when either the wheel or screw wears down, the slack is taken up by the springs.

I dont know if this description is clear. I know it will be hard to draw, but I will give it a go.

Benny, there is a manufacturer who has stolen your idea and put up a website about it already ;)

http://www.logan-mmk.com/zero_backlash.htm

Somewhere, sometime, I also read about another design that incorporated sort of a cam-roller stud design for the worm wheel (like a starfish) and a special worm thread that accepted these roller tips.

Thats a bummer.
One of these days I will invent something that is completely new.

These days I have stopped getting excited about my ideas, 4 point something billion people in this world, someone surely must have had the same thoughts.

Years (actually decades) ago, I saw zero backlash worm gears used in radio tuners. They consisted of two thin gears mounted on the same shaft with a spring causing them to be forced to rotate in opposite directions relative to each other (just a few degrees). They meshed with a single worm.

One of the gears was screwed to the shaft. The other just applied pressure against the worm.

Ken

Years (actually decades) ago, I saw zero backlash worm gears used in radio tuners. They consisted of two thin gears mounted on the same shaft with a spring causing them to be forced to rotate in opposite directions relative to each other (just a few degrees). They meshed with a single worm.

One of the gears was screwed to the shaft. The other just applied pressure against the worm.

Ken


http://www.antibacklashgear.com/ :cool:

I will do a drawing of what I am thinking about, but in the mean time I will try to verbally desribe it.

Image you the worm wheel is split in half. The split is vertical, so esentally you have two thin wheels when they are taken apart. when the wheels are put together they form a working worm wheel. There is a spring inserted in the assembled wheel that forces the two halves to be rotated a few degrees out of sync from each other.

When the whole assembly is put together, it will put a load on both sides of the screw. This will illiminate any backlash. I guess the comprimise is that the pressure put on the screw would be increased a lot. The other thing is if you have and exsessive load to drive the spring load must be greater than the load the assembly is driving.

One advantage is when either the wheel or screw wears down, the slack is taken up by the springs.

I dont know if this description is clear. I know it will be hard to draw, but I will give it a go.

Yes that is how it is done on regular gears too.
Another option would be to split the driving screw.
Instead of been one long thread it could be split with
a slight gap just over the center portion of the wheel.
If this gap could be adjusted you could take out the lash.
Because it is further away in the gearing transmission the
spring load could be less then what would be required when
splitting the worm wheel.
Ether way you will reduce the active surface area used for
the driving pressure by half.
This should be considered in the design of the wheel and its
contact areas.
Just simply modefying a existing wormdrive would lead to much
reduced loadratings.
I would love to hear from somebody that has actualy done it and
can tell about the results obtained / is it worth it?
Good Luck

Hello Torsten,

It seems this product exist for decade already. it's made in germany. I know rotary table manufacturer that are using it to get the highest precision possible.

You may find information and drawbar on Allytech's website (it's a ".eu")

http://www.allytech.eu/index_fichiers/Backlashfreewormandwormwheel.htm

to have better accuracy and power transfer they do make high teeth. there for the line of action seems to be very good.

Hope this helps

Hi, time to think outside the box......why use a worm drive?........to get the low ratio or as for a dividing head or rotary table the division ability.

You can also use a compounded set of gears meshing together that provided they are meshed tightly will give practically backlash free drive.

I once toyed with a design for a friction free worm drive that had a series of ball races place around the periphery of a worm wheel in place of the teeth, and these engaged with the worm shaft between the flanks of the worm.

The faces of the bearings were to be curved to allow the meshing with the worm shaft, much like the involute curve of spur gears.

That is as far as it went, paper engineering....LOL.
Ian.

Hi handlewanker,

I think the advantage of the wormgear is also to drive a high torque with précision.

However, in some case you may need high speed rotation then precise move or position. This is supposed to be torque motors specificity, how ever on the same system the system I talk in my latest post it's possible to make speed variation with a worm gear (actually it's dynamic backlash variation to allow important shaft rotation change.

you can visite http://www.allytech.eu/index_fichiers/OTTMATIC.htm

Hi Clem, high torque comes at high wear, and that means contact face area of the worm wheel.

The system you showed in your post #12 shows two worm wheels which I suspect sit in the worm shaft and are made to be offset to each other giving the maximum back and front flank contact on the worm, a condition that on normal worm drives cannot exist.

I have a small motor gearbox set that has a set of spur gears inside and is arranged to give ten revs per minute with the motor being one of those shaded pole type.

It is very compact, being about 60mm diam and only 30mm long.

The advantage of this arrangement is that the drive input is in line with the output as opposed to the right angle output of the worm drive.

There is more friction in a worm drive gearbox than an equivalent reduction type using spur gears, due to the sliding contact as opposed to the rolling contact of spur gearing.

The other thing is the cost.

I could not imagine how much it would cost to buy the Backlash free worm drive you show, but I expect it would come into the realm of meggabucks, which if you were contemplating using one for a CNC rotary table drive would probably cost more than the machine which it's used on, correct me if I'm wrong.

It has been said before that the simplest way to get backlash free worm drive is to have a small motor applying a reverse motion to the wormwheel while the worm shaft is driving the wormwheel forward, not very ellegant but it would work, provided the reverse bias motor was applying sufficient load to resist the back force of the cutter etc.
Ian.

Hello Handlewanker

you said "The system you showed in your post #12 shows two worm wheels which I suspect sit in the worm shaft and are made to be offset to each other giving the maximum back and front flank contact on the worm, a condition that on normal worm drives cannot exist"

Actually it's not a 2 worm wheel concept but a 2 shaft worm concept, the point is that you can use higher teeth and transmit better torque.

you said "It is very compact, being about 60mm diam and only 30mm long" the OTT worm and wheel system is a bit bigger, I think minimum distance between axis is 50 mm (to be verified), in your case of course the system I show in my previous post doesn't fits.


you said "I could not imagine how much it would cost to buy the Backlash free worm drive you show, but I expect it would come into the realm of meggabucks, which if you were contemplating using one for a CNC rotary table drive would probably cost more than the machine which it's used on, correct me if I'm wrong." So I do correct you ;-) I do not have the price of the system but you can found some in rotary tables that cost less than 13000 € (sorry I leave you the math). So it's a lot less than the CNC

What would be the price of a system that is having a motor to give the power + a motor to adjust backlash, knowing that the alignment might be a bit tricky and the added motor may takes some place inside the "milling zone" (sorry I miss a little vocabulary)

let me know if I am wrong.

enjoy your week-end

Clem

Hi Clem, just had another look at the link you posted, first time I didn't see the additional link to the design concept of the worms and worm wheel.

I see what you mean by the two worms driving the worm wheel, pretty clever device.

Now that the design concept is clear I think that anyone who can cut a regular worm wheel by the gashing and hobbing method would be able to make this design as well.

The only difference between this and a regular wormwheel is that the tooth shape is longer and more slender, so if the details of the tooth shape were available it wouldn't take much more to make this design.

A hob is within the realms of manufacture by most competent lathe users, and it would certainly solve the problem of a rotary table for CNC manufacture.

I would go so far as to say that a regular Chinese made rotary table could be remade using the design concept as a guideline, and quite cheaply too.

By the looks of it the included angle of the worm tooth is less than the 29 degree angle of the normal rotary table worm, to accomodate the longer tooth form of the wormwheel, so that would not be a problem to make a hob......piece of cake.
Ian.