Interference coupling hub removal - Heating device 1

[tundelash]

Could anyone recommend a heating device (maybe a jacket, belt, etc, and its manufacturer) for the removal of pump coupling that was installed by interference.

The first challnge is that the removal will be in the vicinity of highly flamable fluids/gas. The second challenge is that removal should be fast enough for pump rotor changeout (i.e a fraction of an hour).

I do appreciate responses.

 

[JJPellin]

We install most of our pump coupling hubs with an interference fit. Most of them can be pulled off with little or not heat. In the past it was must common to use a puller plate with a jack screw in the center. Now the mechanics prefer to uses a puller with a hydraulic ram (porta-power). They might hit the hub with a steam hose for a few seconds before they start jacking on the ram. It would be rare to need to use a torch. Do you know how much interference was used at installation? I assume you are talking about a straight fit rather than a tapered fit. More details would be helpful.

 

[electricpete]

Here's my thoughts although I'm not an expert on this subject.

Typically inteference increases with shaft size and the force to pull it off will increase dramatically with shaft size. There are formula's that can tell you how much you have to heat to give a given expected removal force (depends on roughness, areas in contact, interference, and size of the pieces). Recently remvoing coupling with inteference fit on 10" shaft it was very difficult to remove even in the shop. We had to heat the coupling very rapidly (so shaft doesn't have time to warm). I can't imagine trying to remove that hub in the field near explosive gases.

You also need to consider if you pull on the hub what is providing the counterforce (don't pull against the machine)

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

I found a formula to estimate the Force for removal
p = Int_diam * E [1- (Rshaft/Rhub)^2] / (4 Rshaft)
where p is pressure, Int_diam is diametrical interference, E is Young's modulus, Rshaft is shaft O.D., Rhub is hub O.D.

Force would be
F = A*P*mu = (2*pi*Rshaft*L) P* mu
where mu is around 0.03 - 0.3 (use the max value to be conservative)

As a rough estimate of the force which would be required under uniform heating, you can assume the inner diameter of the hub expands at coefficient for steel ~ 6.5 micro-Inch per inch per degree F. Although you will probably not be lucky enough to heat the hub uniformly and not heat the shaft at all so make an adjustment...


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[tundelash]

The pumps being discussed are centrifugal pumps handling hydrocarbon liquid (and of course entained gases. The coupling are interference fitted and the manufacturer said can be removed by heat. The environment requires that no flamable/explosive situation is allowed, and as such a flame heating, hot air heating are being excluded.

Consideraion is being given to induction heaters, (akin to those used for installing bearings) , to rapidly heat up the hubs to about 120 deg C, and thereafter pull them out from the shaft.

The challenge is to get an induction heater with a configuration that can heat the hub quickly while on the shaft for removal without heating up the shaft.

Your suggestions are appreciated.

 

[macmil]

a small steam generator ?

 

[JJPellin]

I don't believe that an induction heater will work. There is no good method to put the bar down the center that carries the current. If this was a manufacturer installed coupling hub, the interference is probably very light. I would pull the hub off cold or heat it a little with a steam hos. We put off interference fit hubs cold all the time. If the shaft and coupling hub are carbon steel (no Stainless), you should not have a problem with galling. Hopefully the manufacturer provided pulled holes drilled into the face of the hub. Otherwise, you have to use a two or three jaw puller to pull from behind the hub. This makes it much harder to get a got pull.

 

[bingopin]

Some general comments:

Removing the hub without heat is, of course, a function of fit...i.e. how much interference. As previously mentioned, it is also a funtion of size. A large shaft, say >6", could be extremely difficult or impossible to remove without heat.

Use of heat requires heat input as quickly as possible. Typically a couple of large oxy/acetylene "rosebuds". The idea is to remove the hub before the heat reaches the shaft. This means you have to be set up properly and work quickly. When the hub starts moving during the pulling, don't stop the movement until the hub is removed. If you stop, it is likely that the hub will "freeze" and require allowing the parts to completely cool and before another attempt can be made.

As mentioned, stainless on stainless can cause big problems. Galling is likely.

While it doesn't help you now, my suggestion would be to convert to a tapered fit with a nut on the shaft end or go with a hydraulically fitted hub. Either of these setups should not reauire any heat for hub removal. We typically use the tapered shaft arrangement for between bearing pumps which require the coupling hub to be removed for seal or bearing work on the coupling end. We use hydraulically fitted couplings on larger turbo machinery.

 

[sprintcar]

As JJPellin pointed out, there are hydraulic pullers available that can be fitted to the coupling OD and press against the center of the shaft. Depending on the diameter of the unit, you may want to avoid the universal 3-jaw type in favor of a more rigid unit.

In this case you could fab a thick plate to match the bolt circle on your coupling, install it with a number of high strength studs or long bolts such that there is a large enough gap between this plate and the shaft to insert a short hydraulic ram or pancake cylinder ram. This will require some spacers behind the ram to advance it as the coupling is pulled off. Like any advice in here, use at your own risk.

"If A equals success, then the formula is: A = X + Y + Z, X is work. Y is play. Z is keep your mouth shut."
-- by Albert Einstein

 

[MikeHalloran]

If you're at the stage of fabricating a big tool, and contemplating a different coupling design, you might as well make a giant nutcracker, and remove the coupling destructively.




Mike Halloran
Pembroke Pines, FL, USA

 

[electricpete]

On the off-chance someone might try to use my formula, I'm posting a correction:

p = Int_diam * E [1- (Rshaft/Rhub)^2] / (2 Rshaft)

Note the change from 4 in the denominator to 2 in the denominator.

Lindenburg's EIT Reference manual 8th ed had 4 in the denominator.
Shigley's Mechanical design had 2 in the denominator.
Fieldings Programs for Rotating Equipment design had 2 in the denominator.

I double checked all were using radius, not diameter.

2 out of 3 wins. EIT Reference manual must have been wrong.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

All were using radiuses (radii) for the hub and shaft dimensions, but the EIT Ref manual was using diameterical interference while the other two were using radial interference, which explains the factor of 2 difference.

Also you can have the calculation done for you here:

http://www.tribology-abc.com/calculators/default.htm

(Scroll down to "C5.13 Design of an interference fit (shrink-fit / press-fit / expansion-fit)")

They also have an adjustment for heating/cooling required during mounting.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Ralph2]

Adding my 2 cents after the job is likely already finished.. But.. You are replacing the pump rotor... So, leave the coupling hub on untill you move it to a "safe" area.. then exchange the hub..
We heat to remove couplings all the time.. saves scoring the shaft.. But.. the secret is lots of heat fast..The idea is to heat the coupling and not the shaft..Once the shaft heats up you loose the advantage.
Have a puller arangement that will pull the full length of the fit (you do not have time to re-position your pulling equipment.. Apply a "moderate" force and heat rapidly.. Keep heating for a few seconds after the coupling starts to move.. "then" apply more jacking power and you should have a clean removal..

Most folk just don't appreciate the size of the torche(s) to heat the coupling.. a cutting torch will not do it..

 

[Enkidu]

Just curious, what is the pump shaft diameter, pump speed and HP? Quite a few folks are going to sliding fits on lower HP applications. Typical fits would be limited to 0.001" loose. The hub should still have puller holes. That may be a long term solution.