Why worm drive crawlers are slow...

[Olle P]

... compared to regular cog wheel drives.

Introduction
This article is intended to provide a true technical explanation to why crawlers with worm drives are slower than other crawlers with same motor. In the process it will defy one or two misconceptions too often expressed in this forum.

Background
I’ve got a MSc in engineering and was very curious to find out why my LCC (which I bought half a year ago) didn’t run as well as I thought it should. The all too common explanation, “Worm drives provide a huge gear-down, therefore they’re slow!”, just isn’t true (as is obvious to anybody with mechanical knowledge). The also common comment, “Worm drives eat brushed motors!” seemed to have more truth behind it, since my motors got very hot. (*)
Obviously the worm drives impose drag and power loss, but how much?

The answers
First let’s kill the gear-down misconception:
- Worm drives do have a bigger drive shaft to wheel axle ratio. That much is correct, but...
- Worm drive crawlers compensate that, in full, by having a lesser motor to drive shaft ratio.
- For a given motor speed the wheels will therefore turn about as fast no matter if there are worm drives or not!

So why is it in reality so that worm drive crawlers are slower?
- To move the crawler takes mechanical power. The power is equal to torque times rotation speed (“angular velocity”, is the correct term).
- The higher the speed of the car, the more power must be provided to the wheels. (Same or more torque times more speed.)
- While providing power to the wheels, the motor also has to overcome some power loss within the drive train. Here’s the key issue: The power loss is far greater with worm drives than with cog wheels, so for a given motor power more of it is consumed by the drive train and less reaches the wheels, hence less vehicle speed can be reached! (All of that consumed motor power is transformed into heat, causing the worm drive to get warm and eventually hot with risk of overheating!)

Why are worm drives so inefficient?
I found this very informative home page that explains it in detail.
The main issue is friction.
- Power transfer between cog wheels is a rolling motion with very little friction. Most of the resistance is between each cog wheel and its axle.
- Power transfer in a worm drive is a sliding motion with the pinion (“screw”) sliding sideways against the gear cogs. This cause considerable friction between pinion and gear.

I applied some of the equations to the worm drives in my LCC, using approximate numbers for angles and friction.
The result was very sobering, to say the least.

Given the relatively slow sliding speed (<1 m/s at 1,500 rpm of the drive shaft) and less than optimal lubrication (grease instead of an oil bath) the friction coefficient stay above 0.1.
The screw angle is only about 10 degrees with the standard drive, and somewhat more with the HD drive. With a friction coefficient of 0.15 this result in an efficiency as low as about 50% for the standard drive! (Slightly better with the HD, and also better (but not good) at higher speeds.)

Neglecting all other losses in the drive train, which is a fairly good approximation by the magnitudes, that means an LCC will need a motor roughly twice as powerful as that of an AX10 to reach the same speed!

It’s also worth noticing that at very low speeds the power is low, so the absolute loss in power also stays low. Here the emphasis is on raw (motor) torque. Therefore worm drives do well for low speed use.

Combining these findings it becomes clear that building a rock racer with worm drives is a non-starter:
- High speed requires lots of power, so a high power motor is required.
- The worm drives will consume a fair bit of the delivered power, so even more power is needed from the motor to keep up with the competition.
- All that power will make both the motor and worm drives to run much hotter than the non-worm equivalents, and heat is THE limiting factor.

Late addon:
So why use worm drives at all?
There are a couple of advantages to using worm drives in crawlers:

• Slim pumpkin.
  - Increases the ground clearance under the axle.
• Pinion axle above the wheel axle.
  - More clearance under the drive shaft.
  - The drive shaft is straighter, causing less friction in the joints.
• Built in drag brake. (Not adjustable though.)
• Large amount of end gearing, in a compact format.
  - Reduced gearbox size/weight.
  - Less torque on the drive shaft and therefore very little torque twist.

__________________________________________
(*) The worm drives wasn’t my main problem, initially. I made a couple of mistakes mounting the motor and wheel axles that made up for most of the difficulties.

[JeremyH]

Good information!!

Originally Posted by Olle P

The main issue is friction.

You hit the nail on the head. I have never had any other crawler that would heat up diff pinions like the LCC.

Originally Posted by Olle P

The screw angle is only about 10 degrees with the standard drive, and somewhat more with the HD drive. With a friction coefficient of 0.15 this result in an efficiency as low as about 50% for the standard drive! (Slightly better with the HD.)

Thanks! I've always wondered what the angle of contact was between the two gears.

Originally Posted by Olle P

Neglecting all other losses in the drive train, which is a fairly good approximation by the magnitudes, that means an LCC will need a motor roughly twice as powerful as that of an AX10 to reach the same speed!

Which is why a brushless motor works so well in these trucks....they are much more efficient than a brushed motor yet you dont see any of the negative side effects (reduced drag brake) as you will in another style of crawler.

[Olle P]

Originally Posted by JeremyH

[10 degrees] Thanks! I've always wondered what the angle of contact was between the two gears.

That's just an estimate by a quick look at it. It doesn't matter much to the calculations if it's 5 or 15 degrees since it makes about the same difference as varying the friction coefficient between 0.10 and 0.20.

[JeremyH]

Ahh...ok. I thought you measured the two gears...

[Olle P]

Too many unknown factors to make proper measurements an issue:

• Normal pressure angle: 20 degrees is a standard, but I don't know if Losi followed it in their design. I simply assume they did.
• Worm lead angle. This is the estimated 10 degrees. The correct number is in the 5-15 ballpark.
• Friction coefficient.
  - What materials are used for the pinion and spool? I don't know, but it doesn't seem like any of the combinations provided with the table.
  - What's the friction coefficient between those materials, with the lubrication in actual use? Impossible to answer, since each user have different lubrication!
  - What's the sliding speed? Between 0.0001 and 10 m/s there's roughly a factor 10 difference in friction coefficient!
  Actual coefficient at 2.6 m/s (5,000 rpm of the drive shaft) is most probably within the 0.03-0.15 ballpark. Definitely not below 0.02 and probably closer to 0.10.

5 degrees lead angle and 0.15 friction ends up at 35% efficiency,
whereas 15 degrees lead and 0.02 friction ends up at 92% efficiency.
The truth is somewhere in between.

Since these facts are generic and there are other crawlers than the LCC that also use worm drives the specific LCC numbers are less important.
What's important is that worm drives typically can eat up about 50% of the provided power!

[JeremyH]

Originally Posted by Olle P

What's important is that worm drives typically can eat up about 50% of the provided power!

Oh, definitely. You can tell that by the amount of heat generated. Just like the CFL vs. incandescent light bulb efficiency comparison....

[Charlie-III]

Originally Posted by Olle P

... compared to regular cog wheel drives.

Neglecting all other losses in the drive train, which is a fairly good approximation by the magnitudes, that means an LCC will need a motor roughly twice as powerful as that of an AX10 to reach the same speed!

OK, did you mean the AX10 or the XR-10? I would hazard a guess that the XR-10 would be even more efficient since there is no 90* power direction change.
The AX10 has a ring & pinion that has it's own power losses.

Great write-up by the way!

[PS, I used to deal with various worm drive set-ups in manufacturing, I know of the issues there but packaging can be better for high torque applications like plastic extruder drives.....]

[Olle P]

Originally Posted by Charlie-III

OK, did you mean the AX10 or the XR-10? ...
The AX10 has a ring & pinion that has it's own power losses.

I did mean AX10. The ring & pinion, as well as the rest of the drive train of course have its losses, but these losses are on par with the LCC drive train power loss, excluding the worm drives...

[zodiaci]

Originally Posted by Olle P

With a friction coefficient of 0.15 this result in an efficiency as low as about 50% for the standard drive! (Slightly better with the HD.)

I remember being on the fence about getting an LCC once, but I thought about how hot the worm-drive gearboxes on the machines we build at work get vs. the regular gearboxes. After researching it a bit, I came to the same conclusion you did; The friction and lack of efficiency weren't worth it to me so I went MOA!

Great write-up BTW!

[EeePee]

Originally Posted by JeremyH

any other crawler that would heat up diff pinions like the LCC.

Kyosho Rock Force was my first experience with that.

[syp]

Thanks for the info this was very good. I have a kyosho rock force. Now I understand the wormdrive and the heat and power isue and the usage beyter.

[djjiz]

Wish you were around before the LCC.

I would have kept the ax10 axles I'm built.

[Olle P]

Added a paragraph about the worm drive advantages in the first post.

[GraniteBandit]

I put the LCC axles on my SCX10. They work great!

[JokersWild]

Not exactly the same but slightly in the book

My worm drive skill saw has huge tq and can be loaded down with no ill effects....... But if I'm looking for a beautiful cut...... Totally pulling out my Makita with hypoid gears. Spins a lot faster and cleaner cuts with fine saw dust.... Just won't hold together cutting concrete or running my 10 inch blade with my Bigfoot kit or my 12 inch bar on my parazi attachment.

Good info to boot tho. Not an engineer by any means, just an old tool who stands in the rain running the saws.....

[syp]

Is it possible to compensate for the lack om power the wormdrive gennerate?

[Olle P]

Originally Posted by syp

Is it possible to compensate for the lack om power the wormdrive gennerate?

Yes, just add more motor power!

That's why we mostly use brushless motors for the LCC and LNC. Brushed motors pushing out the power we want will get very hot. (I've toasted two of them myself.) My current 17.5T motor on 4S doesn't even get warm, and the wheelspeed on full power is sufficient.

And as pointed out in the first post: Wormdrive heat is the limiting factor!
It's fine to have a high top speed available. It's not fine to use that top speed more than in short intervals, with longer cooling time in-between.

[xkill]

Very interesting read!
Thanks!

[syp]

What kind of grease do you use in the wormdrive truck?

[Olle P]

I'd say any grease for high pressure applications should do.
Personally I use the grease that came with the HD drives I bought, as well as cheap MoS2 grease.
Marine grease and propeller grease should do just as well.