HELP!!! Centrifuge type assembly 1

[Rodneyella]

Good evening all.

I have been tasked with the job of building a rotating assembly with the following criteria:-

Servo drive with the capability to rotate an assembly at a maximum speed of 3000 rpm.
The drive shaft is 50mm diameter, steel, vertical, supported in a spindle design and 250 mm long from drive pulley to top of shaft.
On the top of the shaft I have a 7075 T6 aluminium bowl with an outside diameter of 650mm, 300mm high and a 15mm thick wall and base.
Within the bowl there is a horizontal 450mm 7075 T6 aluminium plate bolted to the aluminium bowl base and shaft, on each end of the plate there are two upright 7075 T6 aluminium plates which are 250mm high, 60mm wide and 200mm deep. Basically a U shaped machined 7075 T6 aluminium bracket assembly. The uprights are bolted to the horizontal plate using three M8 bolts class 12.9 Alloy Steel.
These brackets are 450mm apart and centred in the bowl.
The brackets carry a cylinder which weights 35kg. The cylinders axis is perpendicular to the vertical drive access.
The cylinder is secured to the uprights using c clamps with bearings which allows the cylinder to rotate but prevents lateral movement and ensures the cylinder stays securely on centre.
This fully loaded assembly will weigh almost 90kg (shaft, brackets, bowl, tooling, cylinder).
The moment of inertia is roughly 4kg/Msq and I would like to have a maximum of 30 second ramp up and down.

Please note that the assembly will be balanced as well as possible.

My questions are:-

Is this going to tear itself apart at 3000rpm?
Have I over spec'd the brackets?
Are the bolts suitable to hold the uprights?
I estimate one upright will be subjected to 1000Kn of Centrifugal Force even without the 35kg cylinder.
I estimate the yield strength of each bolt to be 30Kn and the upright is estimated at 4000Kn.
I could remove the need for bolts and have the U shaped piece machined in one piece.

Please let me know your thoughts on my approach and calculations.
Remember the centre of the 35kg cylinder is supported 265mm above the top of the shaft.

How thin can I go with the aluminium bowl? This bowl is only there to minimise the amount of air the assembly has to move during operation.

I would be grateful for any advice or help you can provide.

I need to have a solid design for Tuesday morning........HELP!!!

Have a lovely evening.

 

[gruntguru]

Who are you trying to kill?

Seriously, you have provided nowhere near enough detail for anyone to validate your design. Hint: perhaps a diagram, better still: a set of engineering drawings.

je suis charlie

 

[Rodneyella]

Grunt I hope I am not responsible for anyone's death....

Please see the attached PDF which gives a general overview of the assembly.
The shaft is supported with a ABGCB7010-S-250 Misumi type housing assembly which is spec'd to take the load..

Please excuse the basic sketch but I hope it contains enough information along with the description above.

The "Drum" will be motorised via a small motor which will rotate it slowly inside the assembly. (This is the least of my worries, if I can get the bowl assembly to spin at 3000rpm then the rest with be for a different day)
I want to know if the shaft, bowl and bracket assembly can withstand the centrifugal forces involved. I have a big fat ZERO amount of experience with centrifugal forces at high speed.

I appreciate any help.

Rod.

 

[Tmoose]

For 3000 rpm service the eccentricity of the center of gravity from the center of rotation should be less than 0.001". Labeled Page 3 here -

http://www.irdbalancing.com/assets/balance_quality_requirements_of_rigid_rotors.pdf

I don't picture An assembled mechanism loaded with slurry to achieve anywhere near that.

Centrifuges and washing machines usually rely on supercritical flexible mounting to allow the rotating assembly to self center, and exert minimal force on the support structure.

https://www.youtube.com/watch?v=eka0I3kyD20

 

[BUGGAR]

If you like the centrifugal forces, you'll love the precession forces added by the rotating drum within the drum.

In view of these additional forces, perhaps I can interest you in supporting the bowl on a hub unit bearing. The fixed flange of the unit bearing would be supported on a tube capable of handling all drum forces except rotation. Rotation torque would be provided by a shaft within the tube driving the splines of the rotating part of the bearing supporting the bowl. This reduces forces on the drive shaft. I would recommend a Cadillac CTS hub unit bearing for this application. And they have built in speed sensors and it's easy to add brakes.

ps. Jimmie Johnson will give me a ride for pushing other GM products as well.

 

[Jboggs]

This one statement tells me all I need to know about this structure: "the assembly will be balanced as well as possible". I will remain far away.
It is possible to both specify the actual maximum imbalance, and take steps to insure it. I would recommend you do it.
But then you load it with an uncontrolled movable mass of slurry? Ever hear a washing machine that gets out of balance?

 

[Rodneyella]

Hi guy's,

Please excuse my late response.

Allow me to provide more clarity.
The Drum is a solid weight of 35kg;- it is a machined shaft. There is no slurry involved.

Thanks Tmoose for the ISO balancing standards, that is a great help.
Buggar, I like your idea of the hub, that will give greater support to the assembly and handle the drive torque separately.

With regards to balancing of the assembly; I plan to bring a balancing company in to balance the assembly:- I don't think perfect 1000000% guaranteed balancing is possible on this type of assembly.
Jboggs, thanks you for your advice and I do plan to have a specified imbalance to provide a tolerance for the balancing guys.


What do you think of the bowl being able to withstand the centrifugal forces, how thin can I make the walls and how thin/light can I go with the support brackets? Is 7075 T6 a suitable material for the job? I need to loose weight

Thank you all for your comments, thoughts and opinions. I greatly appreciate it.

 

[moon161]

A similar application maybe, for cleaning up waste fryer oil:

http://www.wvodesigns.com/raw-power-beast-centrifuge-6000rpm-230v.html

 

[BUGGAR]

I suggest 6061 T6. It's common, comes in shapes (see Aluminum A$$ociation Design Manual) and can be welded (at a loss in strength). You are likely in fatigue stress range which will affect whatever aluminum you use.
It's Tuesday!

 

[Rodneyella]

Hi guy's,

Thanks, I really like the centrifuge link, it seems very robust with a direct drive setup.
If 6061 T6 will withstand the forces involved then it would also reduce cost, happy days

Thanks guy's.

 

[Tmoose]

"If 6061 T6 will withstand the forces involved .... "

Crummy Geometry trumps gorgeous materials almost every time.
And vice versa.