Enveloping worm gear quieter? 2

[AHartman]

Hi everyone,

I have a pretty basic question and haven't really ever done a gear design before, so excuse me if the question is ignorant. I've been asked to reduce noise in a hand held pump system that uses a worm gear on a small DC motor to turn a gear to get the pumping action. The worm and gear are both non-enveloping (not throated). I've read in many places that a single or double enveloped (worm or worm+gear) is more efficient, but does that mean it's quieter too, or is that completely unrelated?

The worm is reducing speed 100:1, and a ballpark torque for the worm is 0.01 N-m, with a brass worm and nylon gear. The output gear turns just about 60 rpm. Given this "flavor" and scale of gear train, is it even worth considering an enveloping design?

Thanks in advance!

-Adam

 

[hydtools]

Brass worm and nylon gear should be fairly quiet. Are you sure the gear set is the source of the primary noise? How much noise reduction do you need to achieve? Have you done an analysis to get an idea of frequency content?
The 6000 rpm motor may be the primary source.

Ted

 

[MikeHalloran]

Pretty much _anything_ turning 6000 rpm is going to make some noise, and more so in proportion to the degree to which it is imperfect.

The first thing that comes to mind is ball bearings. If any are present on that shaft, use better ones, or if life and load limitations allow it, use Oilite(r) bearings.

Also, if neither gear is enveloping, only one gear wheel tooth at a time is carrying the load, and some noise will be produced when the load transfers from one tooth to the next. You may have to retool both gears in order to add throat to even one; it may be easier to isolate/damp the noise than to prevent it.




Mike Halloran
Pembroke Pines, FL, USA

 

[plasgears]

If the stresses allow, consider a nylon worm against an acetal gear. There is much damping and resilience in neat nylon. If the loads are light, nylon against nylon is very quiet and resilient against stall. Enveloping is not the solution in small sizes.

 

[AHartman]

hydtools: The motor runs at 32 dB (tested in an anechoic chamber) and gets louder when installed in the case, but without any gear train. We know that this means the case is amplifying motor noise, and we're working a parallel effort to address the problem from that angle. Installing the gear train adds more noise on top of that amplification.

MikeHalloran: We're using bushings throughout the drive train. To my knowledge, the motor manufacturer indicated that their bushing motor was quieter than their bearing motor.

plasgears: I just learned about acetal as a possible gear material while investigating this problem yesterday. Is the loss of accuracy in a nylon worm outweighed by it's damping properties? Also, what constitutes a light load?

Thanks to all of you for your input!

-Adam

 

[Tmoose]

I'd feed the output from a microphone (or the output from the sound level meter) into a computer sound card, capture it, then run the file thru a freeware FFT program to identify the frequency content. The mesh frequency of worm gears is generally considered nearly continuous and smoother than spur, bevel, etc types, but if the shafts deflect or there is a lot of clearance or the profiles are crude I'd wonder if a collision could not occur.
Here's a way to calculate the gear mesh frequency.

http://www.mhm.assetweb.com/DRKNOW/APLPAPR.NSF/apweb/41739AE6017A735F852565A2005FF450?OpenDocument

I don't know, nor could find much about worm gear pumps tp speculate if there is are pressure pulses at the gear mesh frequency, but other pumps can make a lot of hydraulic noise at their various meshing frequencies

 

[AHartman]

Tmoose: Thanks for the info and link, I'll certainly check it out. The worm isn't actually providing the pumping action. It's turning a roller for a peristaltic pump.

-Adam

 

[plasgears]

AHartman:
Case study. We had a design of sub fractional gearmotor that went into stall. Option of slip clutch was not avail. We changed the last stage pinion and gear to neat nylon and added a thermistor in the circuit. Results were very satisfactory. We got this advice from a plastic mfr. They see all kinds of designs and solutions.