Epicyclic bevel gear train ......... Cast without rework, modeling with homothety / scale factor?

[spectrum]

Hello

I am in the phase of looking for cost reduction on a compact gearbox.

By contacting and after many exchanges the draft for a gear is 1°30.

My gearing to respect the torque will be 12 mm wide (in RDM by hand ...........).

The unmolding will not be a problem, indoors it is to be determined.

My difficulty is to have an opinion before making the gear parameterized to the real profile and therefore to transmit to the foundry a first step for feasibility study and tonnage pressure compatibility of his press, my volume is important unique structural part and the internal toothed gear will be obtained directly without the use of insert added to the toothed part.

To allow a large print run of coins I am on a zinc alloy.

My current research base, the gear will have an apparent modulus and a real (conical) modulus, if I decide to vary the modulus between two values and by dividing with a zero point into two symmetrical intervals by adding spring washers to plate the gears, will the modeling solution of using the scale factor on the tooth profile be viable?

A feedback on the different attempts already validated on a CAD/CAM model, otherwise I will opt for another modeling approach, a sweep between the two extreme sketches related to the own support plan.

Thank you for your advice.

Spectrum.

[Zozo_mp]

Hello

two or three things that will not fail to intrigue my colleagues

Since the gear seems to be internal to the crankcase and will be made without re-machining, it means that you want to have a good gear the first time!

You say apparent modules and real modules.
It would have to be specified because I understand that the gap between the two would be the foundry tail. However, gears have precise dimensions and very tight clearances. Will in your case be any play between the fixed teeth of the crankcase and that of the epicy gear?

I don't really feel the scale factor because it will reduce the internal diameter while it seems to me that in your case the entax between two teeth must at least be as good as possible.

In addition, you know this better than me, the foundry shrink is not homogeneous. It all depends on the type of foundry (sand, lost wax, zamack under pressure, etc...) I have the impression that it's die casting but which one.

Can you give us some information about the type of foundry?

Kind regards

[spectrum]

Hello Zozo,

Let's say that in France no answer..... The foundry contacted is in Italy and we were directed to a die mold, you were right.

Here is the site https://fr.fonderiabenini.org/prodotti

From 500 parts, it is an acceptable cost compared to machining rework.

We need different games to achieve a launch of 30,000 pieces over 3 three production campaigns, it's part of our budget.

We are betting on containing the foundry stops (your evocation narrows from foundry).

This company can hold an angle of 1°30', i.e. in our projection on a simple sketch, a variation of 31/100  on the available thickness of 12 mm.

One solution considered is to start from the anomaly imposed by the foundry and make it a technological solution, with the addition of a pressing force of 5 N to 15 N to bring the teeth into contact and obtain a blank break-in if necessary, we use  spring washers adapted for ball bearings (prestressed).

The basic assembly is simple, one planetary and five satellites and the crown for the output of the reduction retrieve on the satellite holder and fixed crown.

Calculation manual for the epicyclic train:

Fixed element: satellite door

Leading: Planetary

Number of Planetary Crown Teeth: 90

Number of Planetary Sprocket Teeth: 30

Module: 1

Driving Element Speed: 3500 RPM

Input torque of the driving element: 50 N.m

Results and characteristics of the epicyclic train

Planetary Sprocket to Planetary Ring Ratio – 0.33

That's a gear ratio of -3.00 to 1.

For 1 revolution of the planetary pinion, the planetary ring is -0.33 revolutions.

Primitive diameters

Planetary corona: 90.00 mm

Satellites: 30.00 mm

Planetary gear: 30.00 mm

Satellite gears

Number of teeth of satellite gears: 30

Optimal number of satellite gears: 5

Angular velocities

Planetary Corona: -1,166.67 rpm

Satellite Carrier: 0.00 rpm

Planetary gear: 3,500.00 rpm

Resulting torque

The resulting torque on the planetary corona is -150.00 N.m, without taking into account mechanical losses, which are certainly small but not zero.

Fixed element: planetary corona

Leading: Planetary

Number of Planetary Crown Teeth: 90

Number of Planetary Sprocket Teeth: 30

Module: 1

Driving Element Speed: 3500 RPM

Input torque of the driving element: 50 N.m

 

Results and characteristics of the epicyclic train

Multiplication ratio between planetary pinion and satellite carrier: 0.25.

That's a gear ratio of 4.00 to 1.

For 1 revolution of the planetary pinion, the satellite holder makes 0.25 revolutions.

Primitive diameters

Planetary corona: 90.00 mm

Satellites: 30.00 mm

Planetary gear: 30.00 mm

Satellite gears

Number of teeth of satellite gears: 30

Optimal number of satellite gears: 5

Angular velocities

Planetary Corona: 0.00 rpm

Satellite Carrier: 875.00 rpm

Planetary gear: 3,500.00 rpm

Resulting torque

The resulting torque on the satellite carrier is 200.00 N.m, without taking into account mechanical losses, which are certainly low but not zero.

Lubrication becomes an important parameter, the games will be cancelled.

We are at an impasse on our Cycloide reducer the reduction will not correspond in a reduced footprint, this principle is great but on larger sizes.thank you for the help given on this functional but too imposing reducer.

Kind regards.

Spectrum.

 

[Zozo_mp]

Good evening

I guess the casing is made of aluminum or zamak maybe you could ask people who doelectro-erosion by CNC sinking CHARMILLES (I've never done on aluminum but everything that conducts electricity is eligible): on steel molds it's nickel especially with twisted shapes in addition you have a surface condition close to polished mirror with a  Ra = 0.4 to 1.6 μm

In your case it's interesting because by making the foundry 0.2mm larger than the dimension you would have the dimension to the hundredth and as there is little material to remove it must go quite quickly.
Of course, it's a rework, but you can make all the teeth in one go (or in quarter circumferences depending on the size) in a very short time (pay attention to the size limit of the electrodes).

Given the quantities announced, it's worth asking if you haven't already explored this path.

Kind regards

PS: how big is your gear?

[spectrum]

Good evening

Thank you for the alternative, we will get closer to them and discuss our project.

The internal gear of the crown casing must be 90 mm in diameter, inside a wheel that will ensure the mobility of the object.

The number of wheels is multiplied to have different mobility programming or configuration.

Work on a modular model / reduction of the final cost and clamping by control electronics.

The Ø 30 mm planetary gear  constrained to date to allow the torque to pass with the material envisaged.

The connectors are basically on a collector (power and control, search for carbons in a material (ok solved) which will pass over a low surface the high amperage, the electric motor per wheel is reconstituted in the structure with elements bought in spare parts directly from a motor assembler.

torque of 50 N.m per wheel, mass of the object (not definitive) but payload of 80 to 120 Kg.

Total torque available 200 N.m. with the epicyclic landing gear.

Inverted pendulum base and possibility of an impulse by a kinematics that allows to exceed 1G in acceleration (vertical or oriented jump to pass an obstacle (hole in the surface of the displacement), search for the recovery of stability afterwards.

Our important demand is to be able to recover in a very short period of time  and over a very short distance.

Have the most standard components minimize specific parts, maintenance after 2500 hours of operation;

Kind regards.

Spectrum.