Vacuum Pump to Create 3.0 Bars Vacuum

Hi Noel,

Total Vacuum is -1 barg (0 bar absolute pressure). I don't believe 3 bar vacuum is possible.

But hey, I'd love to be proved wrong.

Cheers,

Anthony

Hi Anthony,

Thank you. If I need to create a suction pad to pull a installed floor marble of 300mm x 300mm in size. Assuming it needs a pressure of 3 bars to debond the marble from the substrate concrete, what portable vaccum pump would you recommend ? Do advise any other approach you have encourtered before.

thank you

There is something wrong with your requirement. Maximum vacuum which can be created is 1.01325 bar, which will take several hours with a precision vacuum pump. You need that marble sould be removed in few minute. A very rough vacuum of 0.3 to 0.9 bar is only possibility to creat.

is 1bar is=atrmospheric pressure? how to transfere it to pisg?

Hi Tornado,

thank you for the correction and much appreciated your sharing of know-how

Cheers.

Noel

i am not realy understand why he cant use vacuum more than 1 bar, if he want to get this marble out of an adhesive floor, it is not a loose marble it is allready adhered to the floor

Because maximum atmospheric pressure is 1.01325 bar on Earth. Driving out the entire air from the space where vacuum is to be created is too difficult. If we create vacuum of 1 bar, the absolute pressure remaining inside the vacuumed space will be 0.01325 bar or 13.25 millibar only. For this application (removing marble), getting vacuum of even 1 bar is too difficult because of sealing problems.

Elaborating on this, one can imagine a perfectly flat and polished tile, and a perfectly flat and polished tool to do the lifting. Place them together , seal around the edge, and you have a perfect vacuum inside. This would be called a 1 bar vacuum, vacuum being relative to the surrounding air pressure. It could also be called 0 bar absolute pressure. Mona Sayed can see, I think, that having removed all the air, there is no way to further reduce the pressure.

This would not be impractical if the bond being tested was a couple globs of chewing gum (anything that will part with a force of about 2000 lb.)

If you'd increase the surrounding air pressure to 3 bar, your vacuum would be 3 bar. And you would be able to lift a tile 3x heavier.

brgds

Snel

Correct. This is what Randall was showing.

Yes agreed, Randall's model is theoretically possible but difficult to implement. It requires perfect sealing at two concentric circular surfaces which may be near impossible, because once a tile is removed, edge of outer circle becomes rough. More ever, it looks like using a machine gun to kill a bird.

More ever, it looks like using a machine gun to kill a bird.
Of course. Randall posted it in jest.
But it shows the fundamental issue, namely that something extraordinary is required to use a suction lifter if a differential of 3 bar is required. The "atmosphere" must be of much higher pressure than found above sea level on earth.

K Fry, I do not agree with your following statement:

"Place them together , seal around the edge, and you have a perfect vacuum inside. This would be called a 1 bar vacuum, vacuum being relative to the surrounding air pressure. It could also be called 0 bar absolute pressure."

As I have already commented at #33, if you create 1 bar vacuum, absolute pressure in the vacuumed space will be around 13.25 millibar, not 0 bar. Also 1 bar vacuum is not a perfect vacuum. In fact creating a perfect vacuum (0 bar absolute pressure) is almost impossible like achieving absolute 0 degree K temperature. Different levels of vacuum in terms of reducing absolute pressure are as follows:

Rough or low vacuum : 1013.25 mbar (atm. pressure) to 1 mbar

Medium vacuum : 1 mbar to 1 micro bar

High vacuum : 1 micro bar to 1 nano bar

Ultra high vacuum : below 1 nano bar (extremely difficult to achieve)

You are intermixing Pascal units with bar units.

Full vacuum = 0 bar_a = -1 bar_g = 101.325 kPa.

This is close enough to 100 kPa that it can confuse people (and is only a coincidence).

Dear Tarnado, I am not intermixing Pascal units with bar units. In fact the bar unit is derived from pascal unit only for convenience because it is almost equal to kg/cm2. Your conversion statement is not correct. The correct statement shall be as follows:

Full vacuum = 0 bar_a = -1.01325 bar_g = -101.325 kPa_g = -1.0332 kg/cm2_g

That makes 1 bar have two unequal meanings, differing by a factor of 1.01325.

(A piece of double-think introduced in 1985, a year too late.)

No, bar has only one meaning. 1 atomosphere pressure is equal to 1.01325 bar and not 1 bar as you have considered.

Please read some history. Before the 1985 change, 1 bar was defined as atmospheric pressure (average at sea level.) Convention may agree (mostly) now on one definition, but historically there have been at least two meanings.

Perhaps you missed my notation. I didn't write 1.00 bar. 1 bar is not the same as 1.0000 bar.

The point I was making was for mona sayed's benefit, who thought that perhaps 3 bars (as proposed by the OP) would work. The (roughly 1%) difference between 1.01325 and 1 is of no consequence and only confuses the issue for someone who has no idea why a 3 Bar vacuum would be effectively impossible to achieve on earth.

Your explanation to mona implied that the problem was the 13.25 millbar remaining after lowering pressure to 1.0000 bar. But even if it were possible to achieve a vacuum of one nanobar, the tile would still not come even remotely close to moving. The OP is not off by 1%, he's off by 300%.

So your answer to mona is wrong in these respects:

Because maximum atmospheric pressure is 1.01325 bar on Earth. Utterly and totally wrong. At altitudes below sea level, and at sea level during a "high", pressures are higher than 1.01325 bar. Driving out the entire air from the space where vacuum is to be created is too difficult. Wrong. The issue it that a very high "atmospheric" pressure (i.e pressure in the region of the tile) of 3-4 bar pressure is needed. Drawing out enough air to lift things is done all the time in robotic systems. It's easy. You just can't attain, in a typical plant setting, much more than .9 bar differential... nor is there a need to do so... you just correctly size the tooling. If we create vacuum of 1 bar, the absolute pressure remaining inside the vacuumed space will be 0.01325 bar or 13.25 millibar only. Wrong. 1 bar has the incorrect implied accuracy to say this. Also, the bar is only roughly equal to the unit "atmosphere" (atm). atm is defined to be exactly 1.01325 bar. You have your units confused. For this application (removing marble), getting vacuum of even 1 bar is too difficult because of sealing problems. This is a non sequitur. The issue is not sealing. The issue is basic physics, not 1% differences but 300% or 400% differences.

In summary: Lifting things by using vacuum (as the term is used in factories) is routine and simple. There is no need to achieve anything close to perfect vacuums. However, lifting things by vacuum, when the differential pressure required to do the lifting is 3 bar, is effectively impossible. That is what the OP and mona had failed to understand. For them, the difference between bar and atm makes no difference at all.

Hello Fry,

Thank you for your sharing of know-how with the layman like me and Mona.

What I am interested is to seek experts like you all to brainstorm for the benefit of those who know less to catch up.

The test requires to achieve at least 0.3N/mm2 adhesion bond strength. For simplicity, I translated this value to approx. 3 bar suction pressure in view of uisng some kind of suction method. This method, if work, would be absolutely perfect in term of no need to deal with troublesome in cleaning of epoxy from the steel plate and also the test could be conducted in short period of time. Currently, it takes one day to set up and the next day to test, and another day to clean the steel plate.

Do you have any tangible idea to share?

Thank you
Noel

Thanks. It was like concentrating on pennies, leaving pounds.

Hi Anthony,

Good sharing of your know-how. Thank you.

My colleague was wondering would it be possible to create an open end box to fit the 300mm x 300mm marble size and then use a powerful vacuum machine to suction the marble out from its concrete or rather provide a pulling / suction force to a predetermined/specified value, say 2.5N/mm2. Of course, the box structure needs to withstand 3 bars pressure. A reversal of pressured vessel? a crazy idea.

Currently he uses very strong epoxy adhesive to glue the installed marble and a same size steel plate together. When the epoxy adhesive harden after 24 hours, he then pull the steel plate using hydraulic jack to cause debonding of the marble. This gives an indicative of the bonding strength of marble to the substrate concrete. The typical tested bond strength is in the range of 0.3 to 0.5 N/mm2.

Guess there is no other way to test the bond strength without using the messy epoxy glue.

thanks

Verify this math:

300mm * 300mm = 90000 mm2

0.5N/mm2 * 90000mm2 = 45000 N force required by vacuum lifter.

I do not think it is possible for a vacuum lifter to generate that much force over a 300mm by 300mm square.

You may want to contact a vendor like Ingersoll Rand and see what they think. See these links:

http://www.ingersollrandproducts.com/IS/Category.aspx-am_en-3772

http://datasheets.globalspec.com/ds/3274/IngersollRandIndustrialTechnologies

OK understand a little more.

So you've measured 0.3 - 0.5 N/mm2 via the steel plate epoxy method that you are currently using.

0.3 N/mm2 = 0.3 MPa = 300 kPa ~= 3 bar so this is where that number came from.

So there are two parts to this. The first is to de-bon the marble from the concrete base, and the other is to lift the weight of the marble.

I don't believe the box idea will work. Quite simply because you are already trying to exceed the possible lifting force of a vacuum.

Max vacuum is 101 kPa.

1 Pa = N/m2

101 kPa = 101,000 N/m2

300mm x 300mm = 0.09 m2

101,000 N/m2 x 0.09 m2 ~= 9000 N

This is the maximum possible force if a perfect vacuum was applied over the entire surface of the marble slab.

As already calculated you need up to 45,000 N to de-bond the marble.

Therefore no matter what the mechanism, box or vacuum lifter suction caps - this won't work using vacuum.

Back to the drawing board - you need a new idea. What is it you are trying to achieve. I understand you already have a working method. What is so bad about the way you are doing it currently?

Your current method is using an epoxy glue. This works because the glue is able to transmit the tension force from the lifting mechanism via the steel plate to the surface of the marble. The tension force transmitted by the glue exceeds the possible vacuum force.

I guess the good thing about a vacuum is you can easily remove the lifting device, and you have the marble back (basically good as new). I'm assuming this is what you want to achieve.

I've no real knowledge with this, but there are glues that are temperature sensitive. Apply the glue, possibly some ice-packs to the steel plate. remove the marble, then apply some heat bags (or use a hot air gun) to weaken the glue and remove the marble.

Just an idea.

Hope this helps.

Anthony

I've seen marble floors that had to be repaired & I have never seen a slab that has been properly installed come out without damaging the one next to it,To suck it out, unless it wasn't put down properly is nigh impossible, for a start marble is porous so you wouldn't make a seal, the only way to remove it is to drill holes in the middle then gradually chiesel the slab towards the edges being careful to keep the chiesel facing in.

Bazzer

Unless it is a heritage piece or otherwise valuable piece, the cheapest and fastest option is to break the marble and remove the pieces.

3 bars of vacuum? WALOOB.

Change the planet and when you are on the new one let me know the gravity it has and I would then give you the supplier's name.

Would this work

NO

Well, it's a way of creating a pressure differential of 3 bars but it still won't work because:

Vacuum never pulls anything!

. Sounds a bit confusing to you? Well consider what vacuum is: Vacuum is the absence if the surrounding pressure. Pressure is the force applied per area. So it is always the pressure that applies the force.

what you believe to be the force from vacuum is indeed just the fact that on the opposite face of your object you did not take away the air pressure and this pressure pushes the object.

while at first this might sound like semantics it is not: If you consider a marble slab that adhers to something, wherever that adhesives is (if it is perfectly glued on the entire backside of your plate) there will be no air there which can through its pressure push the plate towards the vacuum!

as a side note: once you have grasped the sence of this you will only chuckle at how lame the semantics of an airplane "hanging" in the air rather than being pushed up by the air really is.

I was relying on the +2.1 relative pressure getting under the tile to be lifted the same way as you rely on atmospheric pressure getting under it when using an ordinary vacuum tile lifter.

But it was really intended to be tongue in cheek.

Would it work?

Well in principle I think it's OK. However there are several practical reasons why it would fail miserably.

1. The acrow would probably go through the ceiling above or at least cause considerable damage.

2. Depending on the structure of the floor under the tiles you would probably blow loads of plaster off the ceiling below.

3. Assuming that the average force needed to remove one of the tiles is equivalent to about 2 atmospheres (and the three figure is to ensure that the best fixed come off), then some of the tiles surrounding the outer hemisphere would start popping out of the floor as soon as the high pressure pump got up to speed.

My learned colleagues keep insisting that -1 bar is the limit and they are correct in a system that uses air as the medium, since once all the air is evacuated they cannot "pull" any more vacuum.

However, what if the space being subjected to suction was "hydraulic" in nature? Finding a fluid that does not de-gas at such negative pressures is simple.

Using a fluid also gives significant benefit, since once the volume is primed, it will only require removal of a very small amount of that material to create significant negative pressure.

Your "box" only needs to be deep enough to accomodate the surface irregularities of the tile, but remember it is effectively a pressure vessel and will need to be of substantial material.

Nope you still only get one bar of negative pressure.

Even less if you are above sea level to any degree.

disagree. the max is 1 bar negative. This is from the surrounding air pressure (0 kPag = 101 kPaa) pushing in against the vacuum. Regardless of how you create the vacuum, the limit is still the same.

Using denser, more viscous, incompressible (and those that don't evaporate) will give you a better seal, and more effective vacuum. but the limit is the same.

Contact Schmalz, they are a company I have done much work with. They have a dedicated Engineering staff to help people with projects just like this, and as long as you buy their equipment, the Engineering is usually free.

I think this project needs to be re-thought, and started from the basics.

It's very easy to over complicate things

Thanks for the confirmation of my insanity.

I am torn between whether what I put up is utterly stupid (the pressure reduction below 1 bar will just create cavitation) or otherwise.

I'm thinking of scenarios with adhesives that are still flexible that are able to support pressures (in tension) more than 1 bar and wondering why they don't cavitiate etc. It's got me screwed up today so I'll drop out and defer to others who are thinking straight at this time.

I am torn between whether what I put up is utterly stupid (the pressure reduction below 1 bar will just create cavitation) or otherwise.

Yes. It is utterly stupid*.

Replace the OP's box with a cylinder, and your fluid with a steel piston which starts with a perfect fit against the marble. Lift the piston. Still 1 bar.

* Oh alright, we'll chock it up to insufficient coffee. Sometimes adhesive tapes are described as acting like little suction cups. But if that were the only effect, then the bond strength in tension across the adhesive film would be limited to 14.7 psi. Van der Waal forces have to come into play: there are removable tapes with higher bond tension strengths than 14.7 psi.

If the marble is extremely smooth, then a little water could replace the whole vacuum pump system. The lifter could be a very thick piece of marble with a polished surface. 144 sq inches x 14.7 = 2117 lb. Unfortunately, the epoxy bond strength being tested is likely to be higher than this for a single square inch.

An ordinary (but very large) suction cup could work too. (You'd need to use a lot of force to "burp" it.) But still his 12" x 12" tiles would be pulled up with very low force relative to the bond strength being tested.

Unfortunately, the epoxy bond strength being tested is likely to be higher than this for a single square inch.
Upon rereading, I see that the tile adhesive is probably not epoxy.

Get a small chisel and a hammer. Start in the middle of the tile and work slowly and carefully toward the edges. Faster than your method...

A perfect vacuum, better than interplanetary space, is zero bar, zero Torr (mm of Hg).

So I regret 3 bar vacuum makes no sense. You might like to reconsider your question.

If you make a 600mm sq box and connect it to a vacuum pump capable of creating -0.85 barg (0.15 bara), then you will be able to lift your 300 mm sq slab assuming you need 3 bars to lift it. See below.

Sorry, but that won't work.

Isnt 3 bar equal to 43.5 psi ?

How is this a vacuum ??

Do you mean -3 bar ??

If so, I have never seen any vacuum pump capable of achieving that deep of a vacuum

A recommendation to many of good hearts giving an answer : revise you BASIC physics and answer to a question when you are aware of what is required. In case of doubt look somewhere to verify what you want to give as answer. It would be better for the overall result and quality.

By the way nobody noticed a flaw in what I wrote: a change of gravity will affect at same time the gas density (thus its pressure at sea level) and the piece of stone to be lifted !

Hi folks,

Randall has shown the way. If you create an ambient pressure of >3 atmospheres, you will be entitled to create a "-3" vacuum somewhere in this closed volume. Problem is, in order to raise your marmor slab the additional pressure must be transferred to all the voids in the mortar or glue under the slab. Will this take place within a usable time frame?

brgds

Snel

If you can get between the tiles: drill holes through the grout all the way round the one you want to remove; insert tubes with holes facing "inward" under the tile then force water under the tile as explosively as possible (I'm thinking of a cylinder a plunger and a sledge hammer). The cylinder would need to be about 1 inch (25mm) in diameter and you would need equal lengths of hydraulic hose to the tops of each of the inserted tubes.

This is unlogical because the maximum theoretical vaccum is 1 bar

Dear Noel, something wrong in your statment. Vacuum (absolute) means zero bar pressure. I believe you need a gauge pressure relative to your external pressure over your cylinder or metal box. See at sensorsone.co.uk/pressure-units-conversion, there are good tabels of what you neeed.

At 3,0 bar pressure you will have 2.96 atmosphere pressure, not vacuum.

Er, which planet is this -3bar on? It's important to remove any confusion.

Dear all,

Thank you all for sharing your know-how in many ways. Pictures below are the test pieces on bond strength. The steel plate is recycled after labouriously removed the debris from it. It would be so much productive if suction approach can be used to suck the tile up to the specified load, say 0.3N/mm2 to pass the test without destroying the installed tile. May be there is some RF method to check the bond strength?

Is there any more productive method to remove the debris from the steel plate. Perhaps, you may have a super idea.

thank you

Noel

Depending on the adhesive used, some solvent like MEK or ethyl acetate is probably necessary. Soaking the plate in the solvent for 30 minutes or so to let it dissolve the adhesive.

Unfortunately these solvents have significant health and fire safety issues that require a good deal of caution during their use.

Some type of enclosed parts washer would be ideal, but somewhat costly, to use. How many times a day (or week) do you want to perform this test? Is the test in a fixed location, or in various field locations?

Hi Ried,

Is MEK the brand name or the solvent name, so it is a ethyl acetate solution ? What kind of enclosed parts washer would you recommend. Perhaps can be a glass hood with air suction function to expel the volatile and toxic fume.

The tests are carried out in various field locations (dwelling), all tested plated are brought back to the laboratory everyday. We have subtantial nos of steel plates and easily need to use 20-30 plates per days.

To remove the debris from the steel plate, the guy just drop the steel plate at waist height as this would remove substantial hardened /brittle epoxy adhesive and the tile. Sledge hammer is used to knock off the remaining subborn epoxy debris. Sometime it is necessary to chisel it off. The last 20% debris would take 80% effort to remove. This is the reason why I am looking for a more sensible method to improve productivity.

Thanks

MEK is the abbreviation for methyl ethyl ketone.

Since you're using epoxy, these solvnts might not be suitable. Check with the epoxy manufacturer what they recommend for clean up.

Maybe a small sand blast booth is better for cleaning?

I've also seen a demonstration of CO2 pellets used for cleaning paint off metals. Baking soda is also used for blast cleaning operations.

Thanks for the note on epoxy and solvents.

A question for Noel, why are you performing the field test?

Is the test really necessary?

You earlier mentioned using some non-destructive test or scan. Is this for customer acceptance, new product development, or governmental requirement?

Hi Ried,

Yes the test is necessary to satisfy the quality control as specified by housing contract.

The test is conducted on newly installed tiles for new dwelling / housing as per contract specification. This has to be on site /field at different localities.

The non-destructive test or scan is new testing approach development in the hope to replace this messy epoxy pull out debonding test.

Thanks

Noel

http://www.coldjet.com/en/information/what-is-dry-ice-blasting.php

http://www.google.com/search?sourceid=navclient&ie=UTF-8&rlz=1T4ADFA_enUS388US389&q=cardox+co2+pellet+cleaners

http://www.co2dryiceblaster.com/

I do not know what your economic situation is, but rental or subcontracting a CO2 dry ice blaster might be a better cleaning alternative for reusing the plates.

http://www.google.com/search?sourceid=navclient&ie=UTF-8&rlz=1T4ADFA_enUS388US389&q=baking+soda+blast+cleaners

Baking soda blasting is an alternative to sand blasting that can be less damaging to the metal plates than sandblasting.

Ken Fry gave a good list of alternative adhesives that might offer an easier clean up than the epoxy currently used.

Good luck with your sticky problem!

Yes!

I cannot imagine what a vertical pullout test is really accomplishing -- it is simply not a failure mode that is likely to occur.

I'd guess there may be something non-obvious that this test is to serve: maybe the tiles are on a space shuttle ); But the ordinary floor tile adhesive is is more likely to fail in a mixed mode in which, for example, one side is forced down into a void, levering the other side up, a fracture propagates as a result of an impact, etc. etc.

Perhaps a restaurant standard of some sort? Pharmaceuticals? Someone determines that tiles must not be loose, because filthy crud can collect underneath. So they come up with a test standard: "Subject to an upward force of x the tile must remain in place." So strange that these bonds are tested to failure though. Puzzling.

Hi Fry,

Please to read your good sharing of knowledge.

"Subject to an upward force of x the tile must remain in place."

This is absolutely correct, the x is 0.25N/mm2 as per test specification/requirement on Quality. It is not necessary to test it to failure but see below.

"So strange that these bonds are tested to failure though" . Puzzling.

The typical debonding failure load is between 0.3 N/mm2 to 0.5N/mm2. This was conducted so that the steel plate can be removed with the debonded tile and allow for installation of new tile.

Thanks

Hi Tompa,

Appreciate your sharing of good engineering view points.

The whole surface of the tile and the steel plate are plastered fully with 3-4mm thick industrial epoxy; a movable hook which allows self adjustment in the hole of of the backing fixture of the steel plate is used to pull the steel plate. This arrangment minimizes the potential deformation due to misalignment and allow the pulling force / pressure to be distributed evenly across the test surface preventing premature brittle failure.

Thanks

noel

What about getting some sacrificial plates made. The mild steel would cost less than $3 per tile. I imagine that some one would weld the required fixings onto the plates for a dollar or two.

That way someone goes round epoxying the sacrificial plates to tiles one day. The next day someone goes round with the rest of the equipment.

You throw away the tile and sacrificial plates, or, ask someone to tender to recover them for less than the cost.

Stingy Singaporean boss lah.

Hi Wal,

Noted your interesting point, perhaps you can elaborate further to share your thought to why you came to such opinion. I believe you have something concrete to share in term of economic numbers.

We will adopt your idea if it is valid and economically viable.

thanks

Noel

That was a typo error.

It was a question. The question mark was omitted.

How much does the plate you want to save cost? How much would it cost to save it?

How much it would cost to scrap it using a chisel and sacrificing the marble, clean the surface and prepare for a new plate?

The whole is an economical problem. Compare the costs and if the solution with a sacrificed steel plate which you bond and you never know if you can get the marble free from it unharmed is more economical then use it.

Hi Randall,

Currently, the 300mm x 300mm x 4mm thick steel plate with welded backing stiffeners fixtures cost about USD30 each. Refer to #46 on plate shape. I am not sure how did you cost it out.

thanks

Noel

I figured you could get away with 2mm sheets 250 X 250. That would leave a 1" (25mm) border around the edge.

http://www.themetalstore.co.uk/products/mild-steel-sheet

You could get 50 sheets out of the 2500x1250 sheet at £94. So that's less than £2 ($3) per sacrificial sheet. Now the difficult bit you've obviously got to change all your existing steel fixtures for something which you can mate easily to the modified sacrificial plates.

Maybe an aluminium extrusion would be a better idea for the sacrificial plates. They would just slide horizontally onto the new steel fixtures.

Hi Randall,

I do apologize that my imagination is terrible as I am unable to visualise how your suggestion using sacrificial plate function.

What is purpose to provide 25mm edges? If you use 250mm x 250mm size plate, this is lesser that the required 300mm x 300mm area to be tested. Perhaps, you can enlighten with some sketches.

If you glue the 250mm x 250mm onto the 300mm x 300mm tile, then how is the new steel fixture to be in order to pull the plate out?

tks

rgs

Noel