Epicyclic gear discussions
19 views
Skip to first unread message
PerBear
Oct 19, 2008, 7:29:42 PM10/19/08
to Nohassel
There is finally a nomogram available. Please have a look at the new
nomogram graph (jpg). It is for epicylcic gear RPM calculations. It
shows quite well what variable range a 78/30 cogs epicyclic gear has.
File is called nomogram-rpm.jpg. Please let me know if this is not
understandable.
I think a torque and a power nomogram could be handy, so I will try to
make those as well.
PerBear
nomogram graph (jpg). It is for epicylcic gear RPM calculations. It
shows quite well what variable range a 78/30 cogs epicyclic gear has.
File is called nomogram-rpm.jpg. Please let me know if this is not
understandable.
I think a torque and a power nomogram could be handy, so I will try to
make those as well.
PerBear
PerBear
Nov 26, 2008, 3:38:48 PM11/26/08
to Nohassel
The epicyclic ring gear is a challenge due to high manufacturing
costs. But today I got a very good suggestion from a friend at work
which I want to share:
He mentioned that aluminium extrusion could be a viable solution - the
geometry is suitable for this type of manufacturing, resulting in low
low cost per part. By extruding a tube with gears we only need to cut
it in 10mm wide slice and press it into the LRK rotor. The problem is
that the start-up cost is high, but we will try to get a sponsor on
the dies for this part (and the sponsor will hopefully get a nice
income over the time to come by making gears). This should be
interesting for typical aluminium (aluminum in US) extruders like
Alcoa, Hydro, Elkem, Sapa, Capalex, Orkla etc.
So if someone with connections to an aluminium extrusion facility
could step forward, we could come closer to a finished unit.
Per
costs. But today I got a very good suggestion from a friend at work
which I want to share:
He mentioned that aluminium extrusion could be a viable solution - the
geometry is suitable for this type of manufacturing, resulting in low
low cost per part. By extruding a tube with gears we only need to cut
it in 10mm wide slice and press it into the LRK rotor. The problem is
that the start-up cost is high, but we will try to get a sponsor on
the dies for this part (and the sponsor will hopefully get a nice
income over the time to come by making gears). This should be
interesting for typical aluminium (aluminum in US) extruders like
Alcoa, Hydro, Elkem, Sapa, Capalex, Orkla etc.
So if someone with connections to an aluminium extrusion facility
could step forward, we could come closer to a finished unit.
Per
Tore Hassel Sørensen
Nov 27, 2008, 2:30:08 AM11/27/08
to noha...@googlegroups.com
Excellent idea..!
Well done ..
Tore
-----Opprinnelig melding-----
Fra: noha...@googlegroups.com [mailto:noha...@googlegroups.com] På vegne av PerBear
Sendt: 26. november 2008 21:39
Til: Nohassel
Emne: Re: Epicyclic gear discussions
runnjarn
Nov 27, 2008, 3:13:23 PM11/27/08
to Nohassel
Aluminium as gear material, well, as I think I saw somewhere here, if
you combine it with hard anodizing and PTFE (Tufram) [url]http://
www.bodycote.se/?OBH=261&ID=448[/url] I belive in it, otherwise, drop
it.
The way to achive best efficiency in a gear is to work with as small
dimensions and as hard and strong material as possible, just as with
a
rolling bearing. Depending on number of tooths, the optimal relation
between hardness and strength varies.
For the annular gear, I think aluminium is good as it is a large part
(weight savings) and the tooth bending stress is lowest.
So, surface hardness is in priority.
For sun and plantery gears, hardened steel will always be best. For
prototypes,
it should not be an big issue to find standard parts, for very large
volume, sintered parts is perfect.
The reason for using plastic gears is:
1. If the design require a larger axis distance than needed for a high
strength material (like the first stage in a 2 or 3 stage planetary
gear)
2. Cost savings
3. Noise level
One thing to not forget is the precision, bad precision will always
result in powerloss and high noise.
Normal tolerances on aluminium extrusions are not good enough (unless
you use a very large module)
Some kind of after machining will be needed, like broathing (very
expensive tooling....) or maybe a
plastic calibration process where the extrusion is stretched in a
controlled manner until it shrinks to the right tolerances.
Also a demanding process that for sure is used to calibrate tubes but
I haven't heard about it for this purpose.
you combine it with hard anodizing and PTFE (Tufram) [url]http://
www.bodycote.se/?OBH=261&ID=448[/url] I belive in it, otherwise, drop
it.
The way to achive best efficiency in a gear is to work with as small
dimensions and as hard and strong material as possible, just as with
a
rolling bearing. Depending on number of tooths, the optimal relation
between hardness and strength varies.
For the annular gear, I think aluminium is good as it is a large part
(weight savings) and the tooth bending stress is lowest.
So, surface hardness is in priority.
For sun and plantery gears, hardened steel will always be best. For
prototypes,
it should not be an big issue to find standard parts, for very large
volume, sintered parts is perfect.
The reason for using plastic gears is:
1. If the design require a larger axis distance than needed for a high
strength material (like the first stage in a 2 or 3 stage planetary
gear)
2. Cost savings
3. Noise level
One thing to not forget is the precision, bad precision will always
result in powerloss and high noise.
Normal tolerances on aluminium extrusions are not good enough (unless
you use a very large module)
Some kind of after machining will be needed, like broathing (very
expensive tooling....) or maybe a
plastic calibration process where the extrusion is stretched in a
controlled manner until it shrinks to the right tolerances.
Also a demanding process that for sure is used to calibrate tubes but
I haven't heard about it for this purpose.
Miles
Nov 27, 2008, 3:47:44 PM11/27/08
to Nohassel
Tolerance on the actual diameter of an extrusion the size of the ring
gear would be +/- 0.6mm according to my extrusion design guide - so,
not tight enough, as you say....
I'm keen to try the Keronite process for surface treating the ring
gear..... http://www.keronite.com/ I could make some enquiries....
Miles
On Nov 27, 8:13 pm, runnjarn <pontus.claes...@telia.com> wrote:
> Aluminium as gear material, well, as I think I saw somewhere here, if
> you combine it with hard anodizing and PTFE (Tufram) [url]http://www.bodycote.se/?OBH=261&ID=448[/url] I belive in it, otherwise, drop
gear would be +/- 0.6mm according to my extrusion design guide - so,
not tight enough, as you say....
I'm keen to try the Keronite process for surface treating the ring
gear..... http://www.keronite.com/ I could make some enquiries....
Miles
On Nov 27, 8:13 pm, runnjarn <pontus.claes...@telia.com> wrote:
> Aluminium as gear material, well, as I think I saw somewhere here, if
Per Hassel Sørensen
Nov 27, 2008, 6:02:41 PM11/27/08
to noha...@googlegroups.com
I would prefer not to use steel gears if possible due to weight and noise issues, especially the planet gear carrier would benefit from planets being POM or maybe ABS with glass-fibre filling. I think the solution is to make a QDWP - Quick & Dirty Working Prototype ASAP and then use the reuslt to do a final design of the ring gear and order a batch machined ones made from a tough polymer.
Perhaps a plastic ring gear can be found inside some existing unit, an appliance maybe? It does not need to work for a long time, only long enough to test the prototype before ordering machined parts.
I also got a slightly depressing mail yesterday basically saying that the useful torque from the HS motor would be much less than what I have used in my gear calculations. If this is correct (it probably is, I am planning to do some deep-diving into motor calculations to try to settle this) then the sun gear should have a lot less teeth than in the current design. This might require planets with two gears on same axle, making epicyclic calculations much more interesting! Anyway, it is probably to early ordering a batch of ring gears from India or China just now...
Per
2008/11/27 Miles <miles...@hotmail.com>
Miles
Nov 27, 2008, 7:22:32 PM11/27/08
to Nohassel
Per,
If you have to rethink the ratios, this might be useful..:
http://www.mitcalc.com/en/pr_epcgear.htm
Miles
On Nov 27, 11:02 pm, "Per Hassel Sørensen" <perhas...@gmail.com>
wrote:
If you have to rethink the ratios, this might be useful..:
http://www.mitcalc.com/en/pr_epcgear.htm
Miles
On Nov 27, 11:02 pm, "Per Hassel Sørensen" <perhas...@gmail.com>
wrote:
Reply all
Reply to author
Forward
0 new messages