Vibration table for tests (7-200Hz) from extruded aluminium profiles bolted together - good idea (?)

[filiply]

I need to design a vibration table (approx. 1000x1000m) to test products between 7-200Hz. I was wondering if it is recommended to make the table structure from standarised, extruded aluminium profiles bolted together to reduced the cost (material + time) or does one have to go with a welded structure?

Yes, I am aware of the results of the Junker vibration test with regards to bolted connections and it seems like a bad idea, because all the fasteners would loosen. However, maybe there is a workaround to use these aluminium profiles?

I sincerely appreciate any advice and/or previous experiences with these.

 

[BUGGAR]

The latest Aluminum Association Codes (and hence the UBC) considers friction connections in aluminum structures with fully tensioned bolts. Have you considered this?
ps: I'm trying this in bolting a differential carrier in a vehicle. It better work.

 

[SnTMan]

One thousand meter? Better weld it

Regards,

Mike


The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[JohnRBaker]

I would opt for welded steel, if for no other reason than to add additional mass to the foundation of your vibration table. If there's insufficient mass in the platform, this could have a significant impact on the efficiency with which the vibration source transmits its motion to the item(s) being tested.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[Tmoose]

"the Junker vibration test with regards to bolted connections"

I dunno, forced lateral displacement/slippage in a rollerized bolted joint doesn't seem like a fair test or frankly even a realistic test to me.
At least, I sure hope and expect no properly designed and executed bolted structural joint in service would ever come close to resembling what goes on in that Junkers test rig.

That said, and also knowing nothing of your test or your DUTs, I suspect a successful "vibration table" would require some sharp engineering indeed if it were to be made of bolted extrusions.

And Engineering often goes beyond whipping up a table or anything else based mostly on rules of thumb like "rmmm . cast iron and welded steel good, bolted aluminum bad."

https://www.youtube.com/watch?v=xKdtuwTr-iM

For instance What would most folks' reaction upon hearing a rugged, accurate machine tool base described as using a bunch "plastic?"

http://gearsolutions.com/features/polymer-composite-bases-for-machine-foundations/

 

[racookpe1978]

I too cannot see how a lightly-built "table" made cheaply of bolted Al extrusions could stand under a vibration test stand, and let you assume all of the energy of the test stand motor and drive are not "driving the table" under the test stand just as much as they are "driving the test stand fixture + part" on top of the table.

 

[EdStainless]

The table needs to be stiff enough to transmit the force. With super low modulus Al seems like a poor choice, and bolting does not sound very good either.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[IRstuff]

Don't understand why you "need" to design a vibe table, since that presumably implies the "need" to design the shaker as well. It's going to take you years to learn the lessons the existing manufacturers already learned.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Strong]

"design a vibration table (approx. 1000x1000m)"
Is your table 1000 meters or 1000 millimeters square?

Do you want to spend more time and money rectifying a poor table design or focus on the test object?

Walt

 

[SwinnyGG]

I dunno, forced lateral displacement/slippage in a rollerized bolted joint doesn't seem like a fair test or frankly even a realistic test to me.
At least, I sure hope and expect no properly designed and executed bolted structural joint in service would ever come close to resembling what goes on in that Junkers test rig.

The dynamics of loosening of bolted joints is relatively well understood at this point- and basically every single non-tension-critical structural steel joint in the entire world is directly replicated in the Junkers test. Billions of joints probably.

 

[EngineerTex]

If you're going to go with bolting, use Nord-Lock washers. Verify that the extrusions that you plan on using are an alloy suitable for welding, so that if bolting doesn't work, you can go back and weld it. But welding is very cheap. You will probably be able to find someone to weld it for you for cheaper than it would be to machine it and bolt it.

If you're only going to be able to use the hardware that is designed for the 80/20-style clamping, you may have a difficult time having the bolted joints hold.

Engineering is not the science behind building. It is the science behind not building.

 

[IRstuff]

I think this is one of the reasons many products ship with Loctite on their screw threads

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[SwinnyGG]

The difference between welding and bolting this proposed rig together is, in my opinion, likely to be very minimal. Personally my first concern would be making sure this vibration rig actually functions well before I started worrying about durability.

In order to generate data that is actually useful, vibration test systems have to do two things very well:

1) they have to be strong enough to endure forces that can potentially be much, much higher than intuition would lead you to believe

and

2) they have to have foundations, structures, and drive trains that are stiff enough such that the vibration profile you're trying to get into the parts to be tested can be delivered with accuracy and consistency. In other words, the transfer function between the input waveform and what the parts under test actually see must be as consistent (and ideally, as linear) as possible.

Aluminum extruded profiles are neither massive nor stiff. In my opinion, they should not be evaluated for use as a component of a vibration fixture unless the parts under test are very very small and thus the forces are very very low.

It's entirely possible that OP needs to test some individual surface mount electrical components or something like that, which are extremely small- in which case this might work out fine. Until we get more information about what will be tested and at what loads, it's impossible to tell.

 

[MintJulep]

It depends on the mass of the thing you are shaking.

It depends on the accelerations you are imposing.

 

[SnTMan]

Hello, OP?

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[filiply]

Hi all,
thank you very much for the replies, they have been very helpful.
I have decided to go for the welded one, I do not want to experiment with unproven solutions.

In order to address some of the replies:
1. obviously [mm]
2.

The latest Aluminum Association Codes (and hence the UBC) considers friction connections in aluminum structures with fully tensioned bolts. Have you considered this?

No, I haven't considered it.
3. I obviously want to design a good table.
4.

"It's entirely possible that OP needs to test some individual surface mount electrical components or something like that, which are extremely small- in which case this might work out fine. Until we get more information about what will be tested and at what loads, it's impossible to tell.

I wish - these will be battery packs, up to 15kg.


At the moment I am trying to figure the forces needed given:
max frequency = 200Hz
mass of the vibrating mass = 50kg
amplitude of 0.8mm

Should I calculate the amplitude acceleration a_max and thus the force? It gives me crazy values, something like 62500N, but I am afraid it is correct.

I have attached the image with the approximate calculations:

 

[btrueblood]

Any real world vibration levels of 120 gravities at 200 Hz imply very energetic things like rocket launches or explosions. Is your device going to do those things?

 

[SwinnyGG]

Your calculation is correct.

But for designing your structure, it's potentially even worse than that.

You need to be thinking not just about peak forces, but also transmissibility and power.

You vibration OUTPUT needs to reach 0.8mm @ 200 Hz sinusoidal- at least that's what you've stated.

Your input to the machine needs to be designed such that the input value, when conducted through all the various moving parts that link the actuator to the mass under test, results in your desired signature actually being transmitted to your test parts.

There will be a transfer function relating the input to the actual signature applied to the parts; the less stiff and robust your machine is, the more complicated this transfer function will be.

In short, the design of fixtures and machines capable of generating reliable, precise vibration test data is very specialized and highly non-trivial. There is a very good reason that machines capable of performing these tests are expensive.

It sounds to me like the best thing you could do for yourself and your employer is to engage someone who actually manufactures this type of equipment, and figure out how to procure the equipment you need.

 

[BrianE22]

For a table that small you could mill it out of a block of aluminum and avoid any welding and bolting together of parts. You'd still need to attach to your specimen of course.

 

[IRstuff]

The engineering cost of designing your own vibe table without a serious level of experience will be an expensive lesson learned. Bear in mind that a commercial table's development and validation costs are amortized over the life of the product, whereas your development and validation costs are all on a single unit.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm