Remove an Aluminum Tube

[Thomas Lyons]

Any ideas on how I would remove an Aluminum drop tube from a gasoline or diesel tank? The tube goes thru a 4" steel pipe that extends 3' to 4' above the tank.

The problem is that after years of being in the tank the aluminum has seized to the steel pipe. I am thinking about trying to melt the aluminum or put a rubber plug in and fill it with dry ice or liquid nitrogen and hope it breaks loose.

I've seen a caustic bucket on the rig make an aluminum hard hat disappear in a matter of minutes so I started researching a possible method of using caustic to strip the tube out but I don't know how to clean up the mess or what would happen to the gasoline if the chemical fell down into the tank. (I can't imagine it would be good) but maybe I could melt it suck it out somehow then neutralize it.

I think I could use a rubber plumbers plug down to tank top to hold the chemical and once neutralized and removed or cleaned remove the plug and the bottom of the tube would just fall into the tank. right now we have to take a metal bar and try to cut it away from the steel pipe and just work our way down to tank top and then fight it to get it out.

Any suggestions? At least can someone tell me what happens to the chemical if it hits the gasoline/methanol and how worried do I need to be about the fumes the reaction will create. Thank you for any help>

 

[anorlunda]

Does the tank still have fuel or fumes in it?

 

[Tom.G]

I like your idea of Dry Ice or Liquid Nitrogen as they are the least dangerous approaches, and no aftermath if they fail! That's assuming the steel pipe is malleable, not cast.

Of course the other approach is remove the steel pipe from the tank, then either repair or replace it with a new assembly. If the pipe is welded on the tank, and it being a fuel tank, you would have to use a manual pipe cutter (similiar to a tubing cutter) for removal. Re-attaching afterwards is a separate problem left as an exercise to the reader.

Alternatively, can you just put a slightly smaller Aluminium tube down the pipe and leave the old one in place?

What is the plan to avoid a repeat of the situation in the future?

Cheers,
Tom

 

[Thomas Lyons]

These are 10 to 20k tanks at a regular gas station. New EPA mandates that we check the overfill protection on the tanks. The tanks are buried 3 to 4 feet in the ground thus the steel riser. The drop tube is an industry standard part that has been used for 50 years or more. It is basically an aluminum sleeve that runs from grade to the bottom of the tank to prevent static build up, splashing and in some cases vapors from escaping. Those have to be removed to introduce a new product that has a flapper valve cut in which has a float that raises with the level of fuel and when the tank reaches 90% full the flow from the transport is stopped thus creating an overfill safety device. When the steel riser was new and the aluminum sleeve was new it all slid together very easily but over the years scale, dirt and of course our issue of dissimilar metals bonding has occurred. And yes the new tubes are exactly the same tubes they just have an overfill device cut in at about 10inches below tank top.

 

[Thomas Lyons]

Just to mention that removing the riser from the tank is an option but the cost associated with removing a 4'x4' chunk of concrete plus removing and shoring 3' of pea gravel backfill material not to mention permits and such is about 3k and then add the price of the new drop tube $1,500 and the top spill bucket that gets ruined and all the new brass adapters and caps and the cost of adding the overfill device just went from about $1,800 to about $5,000. I think we need to figure out a way to remove as many of these as we can without breaking ground so any ideas will be helpful - my little contract alone is for 500 stores times average 3 tanks per site so about 1500 of these new tubes need to be installed and about 50 percent of the old ones coming out are stuck. We currently hook things to the bottom to try and pull up and if that doesn't work we try a long bar and bend the tube in and pound the bar down to tank top and sometimes we get lucky but usually not and at that point we have a tank that cannot take a drop until we do construction. These are active gas stations and if the transports can't drop they can't sell fuel and everyone is angry.

 

[anorlunda]

Wow, that is a good engineering problem. I have not thought this through yet, but I have had problems with the sacrifice of aluminum in a bimetal contact because of galvanic currents. I wonder if you can harness that?

Do you have a time constraint, such that you must remove the tube in minutes once you start work, or could it wait several days?

 

[Thomas Lyons]

Depending on the solution, we could conceivably pre visit the sites with the solution as long as it did not interrupt the normal course of operation for the tank or the site.

 

[Borg]

I missed the original comments about dry ice but, that was also my first thought since the different expansion coefficients of aluminum and steel should allow the aluminum to shrink more than the steel. This also may be why they're stuck in the first place.
Just to be sure, I Googled - how to remove a rusted aluminum sleeve from a steel pipe
A few results down had nearly the identical question - https://www.practicalmachinist.com/...-post-stuck-steel-tube-best-way-remove-93604/
One of the comments in that thread pointed to an extended article on removing stuck seat posts on a bicycle.
Cure #10 involves dry ice.
Good luck. That's a lot of tanks to upgrade.

 

[CWatters]

Perhaps...

Some sort of expanding bung or balloon that can be put down the pipe and expanded. Test for leaks into the tank by filling with gasoline. Remove gas and flush with ??. Fill with Sodium Hydroxide? Think that dissolves aluminium but not steel. Keep circulating the Sodium Hydroxide. Flush out and remove bung. Would leave a short length of the aluminium pipe where the bung was but perhaps that's ok?

 

[Dullard]

How thick is the aluminum tube wall?

Is there other tank access which would allow removal of the part of the aluminum tube that extends beyond the steel entry pipe (assuming that it was cut off just inside the tank)?

 

[jedishrfu]

Perhaps you could build a pipe probe with a circulating loop inside that you then insert into the aluminum sleeve. From there you could start circulating freon or some other freezing agent through the loop freezing the aluminum pipe insert. The aluminum insert might shrink enough to attach to the probe and perhaps allow you to break it away from the outer pipe.

The idea of circulating the freon would keep it constantly cold preventing dissipation of the coldness as would happen if you used dry ice in the hole. Then again using the same pipe idea filled with dry ice might work too with no need for a circulating loop.

I'm not a mechanical engineer and so this idea maybe crap but I thought I'd mention it. If it works it could be a useful tool for other stuck sleeves.

 

[Thomas Lyons]

I am ok with leaving the bottom of the tube floating in the gas. I have rubber plumbers plugs that I can blow up to seal the tube. What I really need to know is how can I recover the Sodium Hydroxide and does the aluminum turn to gas or will I have a pipe full of melted aluminum and Sodium Hydroxide. If so when I pull the plug will the Sodium Hydroxide react in an unfavorable way with the gasoline or Diesel fuel? I think it should be ok since I believe it is used in the refining process at some point but not sure.

 

[Thomas Lyons]

If I use liquid Nitrogen would I be able to put in in the tube in large quantities? Could I suck it out after a few seconds and put it in the second and third tubes at a site. Since all tanks will be buried the same depth it would take the same amount of liquid to fill each riser. What would happen if the plug broke and the liquid nitrogen fell into the fuel?

 

[Borek]

What I really need to know is how can I recover the Sodium Hydroxide and does the aluminum turn to gas or will I have a pipe full of melted aluminum and Sodium Hydroxide.

Gas that will be produced during the reaction is hydrogen, after that you will be left with a caustic solution in a tube, aluminum will be dissolved in that solution.

If I use liquid Nitrogen would I be able to put in in the tube in large quantities?

Not more than the tube volume :)

Could I suck it out after a few seconds and put it in the second and third tubes at a site.

Highly doubtful, but after few minutes there will be no liquid nitrogen left, it will quickly evaporate.

What would happen if the plug broke and the liquid nitrogen fell into the fuel?

Nothing more dangerous than the the buildup of a pressure because of the nitrogen boiling (which, depending on the presence of other holes in the tank, can be, or can be not a problem). Plus, some of the gasoline will temporarily solidify (just like freezing water), but I don't think it will change its properties once it melts back.

 

[Thomas Lyons]

Does the Caustic solution remain useable for the next tube or will it become less potent. Can I fill the pipe with water to flush the solution up and out into the containment bucket where I would remove it with something? Is there a way to neutralize the chemical?

 

[Tom.G]

Can you explain, or better yet a drawing or photo of, the new float valve assembly and how it blocks the flow?

I'm thinking perhaps the old Alumunium sleeve doesn't necessarily need removal. For instance a smaller diameter assembly could be inserted, or maybe the existing tube could be modified in situ to accept a float valve. For working down-hole, there are video cameras available on a long lead so you can see what you are doing, Plumbers commonly have them for sewer inspections.

Another possibility for tube removal is a variation of a tubing cutter. Instead of cutting the outer circumference, you could make three (or maybe two) longitudinal (length wise) slits of the lining, collapse and remove a piece to give 'wiggle room' to easily remove the rest. You may have to make a few circumferential cuts from the inside to pull things out in sections. I have no idea if such tooling is commercially available, but the economics you mentioned would easily cover custom design and fabrication.

Edit: Instead of longitudinal cuts, you could make a long spiral cut and collapse the tube by twisting it as it is pulled out.

Cheers,
Tom


Edit 2: p.s. Please try to avoid a chemical approach. Would you want to fill the gas tank of your truck with the residual compounds?

 

[jrmichler]

Some thoughts:
1) I would try liquid CO2 before liquid nitrogen. It's readily available from any welding supplier and does not boil off in the container.
2) Pouring liquid CO2 or nitrogen won't work. Inject it through a pipe with radial holes in the end so it sprays directly on the aluminum.
3) Be prepared for noise and a big cloud of vapor. That cloud will not have oxygen in it. And you need really thick gloves to hold the pipe.
4) Do you have access to drill it out? Something like a big reamer with replaceable carbide teeth driven by a drill like this: https://www.milwaukeetool.com/Products/Power-Tools/Drilling/Right-Angle-Drills/1854-1. Be advised that a drill that powerful needs to be braced because it has enough torque to break an arm or leg if the bit catches. Correction, when the bit catches.
5) Can you get a cold chisel between the aluminum and steel to buckle the aluminum inward? You can weld an extension to a hardware store cold chisel, just make sure to preheat the cold chisel before welding. You would need to pound with at least a 3 pound hand sledge, a little tiny 1 pound hammer will just make noise. Is the steel pipe strong enough to stand up to the pounding?
6) You can neutralize the caustic with acid. Sodium hydroxide plus hydrochloric acid makes sodium chloride (table salt) plus water plus heat. Bases (caustic) react with acids to make salts.

 

[jedishrfu]

Can you explain, or better yet a drawing or photo of, the new float valve assembly and how it blocks the flow?

I'm thinking perhaps the old Alumunium sleeve doesn't necessarily need removal. For instance a smaller diameter assembly could be inserted, or maybe the existing tube could be modified in situ to accept a float valve. For working down-hole, there are video cameras available on a long lead so you can see what you are doing, Plumbers commonly have them for sewer inspections.

Another possibility for tube removal is a variation of a tubing cutter. Instead of cutting the outer circumference, you could make three (or maybe two) longitudinal (length wise) slits of the lining, collapse and remove a piece to give 'wiggle room' to easily remove the rest. You may have to make a few circumferential cuts from the inside to pull things out in sections. I have no idea if such tooling is commercially available, but the economics you mentioned would easily cover custom design and fabrication.

Edit: Instead of longitudinal cuts, you could make a long spiral cut and collapse the tube by twisting it as it is pulled out.

Cheers,
Tom


Edit 2: p.s. Please try to avoid a chemical approach. Would you want to fill the gas tank of your truck with the residual compounds?

Is there a danger of sparks using this approach?

 

[Tom.G]

Is there a danger of sparks using this approach?

Not in any part of it that I'm aware of. The cutting is done by displacing the Aluminium, not removing it with a grit wheel, so there is little heat generated even if the tubing-cutter bites into outer the steel pipe. For the removal stage, Aluminium does not spark, its softening temperature is well below incandescence.

For those not familiar with tubing/pipe cutters, see this youtube video at the 3:00min mark:
(if the ad repeats, hit the F5 key on your keyboard to refresh.)

The approach I'm suggesting is essentially to turn a tubing cutter "inside out" so the rollers and cutting wheel are facing outward to cut the Aluminium tube from the inside.

If there is still a sparking concern, fill the tank with CO₂ or Nitrogen first. CO₂ would be preferable because it is dense enough to stay in the tank, and readily available.

Cheers,
Tom

 

[Borek]

Does the Caustic solution remain useable for the next tube or will it become less potent.

Less potent, but - depending on the concentration - it could be probably used several times.

Can I fill the pipe with water to flush the solution up and out into the containment bucket where I would remove it with something?

Should work in principle, although you may need plenty of water to be sure the base was removed. Plus, when flushed solution becomes diluted and can't be reused.

Is there a way to neutralize the chemical?

Any acid will do, even sodium bicarbonate (baking soda).

That all being said (and despite being a PF's chemist on duty) I am with @Tom.G here - I would treat chemical approach as the last resort, when all mechanical solutions fail.

 

[CWatters]

I think fitting a bung and then pouring in dry ice/CO2 would be worth a try first. I think there is a risk it could make the bung fail but if CO2 gets into the gasoline it shouldn't cause any problems. I assume the tank has another air vent to prevent any expansion of the CO2 causing over pressure or pushing out the gasoline.

 

[Tom.G]

A caution on using dry ice:
That 4 in. diameter pipe has area of about 12 sq.in. and a volume of 4 cu.ft. Filling it will take 300 to 400 25 to 35 lbs. of dry ice (thanks for the correction @Borek). If there is any liquid in the tank when the plug fails, perhaps one of the resident chemists here can calculate how fast it sublimes to what volume of gas when it hits that essentially infinite liquid heatsink; and the possible flow rate out any vent openings.

I'm not saying it's a bad idea, in fact I rather like it, but I also think 'unintended consequences' should be explored before committing.

After Thought: If using an inflatable bung remember to maintain its pressure as it cools from the dry ice.

 

[Borek]

A caution on using dry ice:
That 4 in. diameter pipe has area of about 12 sq.in. and a volume of 4 cu.ft. Filling it will take 300 to 400 lbs. of dry ice.

Perhaps I am missing something, but my estimate is orders of magnitude lower. Assuming 4" diameter I get around 8 lbs per ft of the tube. Actually there will be less, as the dry ice bulk density is probably lower.

If there is any liquid in the tank when the plug fails, perhaps one of the resident chemists here can calculate how fast it sublimes to what volume of gas when it hits that essentially infinite liquid heatsink; and the possible flow rate out any vent openings.

No idea about speed (sounds rather tricky to calculate), but the total volume of the gas is reasonably easy - around 8 cu.ft. per lb of the dry ice.

Can't believe I am using these units

 

[Dullard]

TomG:
We are of one mind. I asked about the wall thickness to assess the feasibility of cutting internally, and to determine if spiral or longitudinal made more sense. In either case, a TBM-like cutter is economic if there is enough work for it.

 

[Tom.G]

Perhaps I am missing something, but my estimate is orders of magnitude lower. Assuming 4" diameter I get around 8 lbs per ft of the tube.

: No, I did. Your numbers are correct.
Thanks for the correction, that also removes the worry about large quantities of gas generation.

(must remember, Math and 1AM are not compatible)

Cheers,
Tom

 

[sysprog]

These are 10 to 20k tanks at a regular gas station. New EPA mandates that we check the overfill protection on the tanks. The tanks are buried 3 to 4 feet in the ground thus the steel riser. The drop tube is an industry standard part that has been used for 50 years or more. It is basically an aluminum sleeve that runs from grade to the bottom of the tank to prevent static build up, splashing and in some cases vapors from escaping. Those have to be removed to introduce a new product that has a flapper valve cut in which has a float that raises with the level of fuel and when the tank reaches 90% full the flow from the transport is stopped thus creating an overfill safety device. When the steel riser was new and the aluminum sleeve was new it all slid together very easily but over the years scale, dirt and of course our issue of dissimilar metals bonding has occurred. And yes the new tubes are exactly the same tubes they just have an overfill device cut in at about 10inches below tank top.

You surely know that an underground tank partly filled with a large amount of gasoline is extremely dangerous to tamper with. While the ideas presented here regarding cooling via carbon dioxide or nitrogen are apt to be less unsafe than using sodium hydroxide to dissolve the aluminum, no innovative solution should be attempted without the approval of a licensed professional engineer. The risks can be properly assessed by, e.g., a firm that does gas station underground tank decommissioning.

 

[CWatters]

+1

I don't know if anyone here has the skills (or insurance) to guarantee their suggestions are safe.

 

[Tom.G]

I don't know if anyone here has the skills (or insurance) to guarantee their suggestions are safe.

I rather doubt that anyone posting here on PF has such insurance to cover unpaid opinions of any kind, much less any other guarantee of same. You don't have to read very many threads to find things that are not all appropriate for many reasons. It is up to the person undertaking an action to choose a course and, cover themselves if they so desire.

After all, there are an enumerated and limited number of statements that don't fall under the 'Free Speech' umbrella.

 

[Borek]

I would assume someone winning a contract to reinstall such tubes in 500 gas stations is not a random Joe IhavenoideawhatIamdoing.

 

[sysprog]

I would assume someone winning a contract to reinstall such tubes in 500 gas stations is not a random Joe IhavenoideawhatIamdoing.

I agree with that assessment; however, I also think that anyone who is seeking advisory opinions on the internet about the matter, presumably doesn't regard himself as already adequately knowledgeable and experienced regarding handling the associated risks, and so would probably do better to have available, preferably on site at least for the first few sites, a licensed professional engineer who has tried-and-true experience in such matters.

 

[James Demers]

Rather than caustic, which is going to be messy and dangerous, how about salt water, with a DC voltage applied to the two pipes? You should be able to drive oxidation of the aluminum, while protecting the steel. It's how sacrificial anodes work to protect ships' hulls (and water heaters), but with a bit of a boost from an external voltage. The method is called impressed current cathodic protection; it's used on pipelines. You'd want something like 1% salt, so the aluminum doesn't passivate. DC from an automotive battery charger should be sufficient.
This won't work if there's metal-to-metal contact (short-circuiting the setup), but there may not be much contact in an old, corroded system. A quick check with an ohmmeter will tell you if there's contact. Potential downside is salt water accelerating corrosion of the tank, if it gets inside. A mild acid (vinegar) should work as well, but you're getting into R&D land there.
Once the aluminum is sufficiently corroded, you can probably rip out the remainder with a reaming tool of some sort.

Further thought: rather than the steel pipe, you can provide your own cathode in the form of a steel or iron rod inserted down the middle of the aluminum tube. Wrap it in burlap, or slide some rubber rings onto it, so it doesn't contact the aluminum.

 

[sysprog]

Rather than caustic, which is going to be messy and dangerous, how about salt water, with a DC voltage applied to the two pipes? You should be able to drive oxidation of the aluminum, while protecting the steel. It's how sacrificial anodes work to protect ships' hulls (and water heaters), but with a bit of a boost from an external voltage. The method is called impressed current cathodic protection; it's used on pipelines. You'd want something like 1% salt, so the aluminum doesn't passivate. DC from an automotive battery charger should be sufficient.
This won't work if there's metal-to-metal contact (short-circuiting the setup), but there may not be much contact in an old, corroded system. A quick check with an ohmmeter will tell you if there's contact. Potential downside is salt water accelerating corrosion of the tank, if it gets inside. A mild acid (vinegar) should work as well, but you're getting into R&D land there.
Once the aluminum is sufficiently corroded, you can probably rip out the remainder with a reaming tool of some sort.

Further thought: rather than the steel pipe, you can provide your own cathode in the form of a steel or iron rod inserted down the middle of the aluminum tube. Wrap it in burlap, or slide some rubber rings onto it, so it doesn't contact the aluminum.

The electrolytical process is too slow, the energy cost is too high, the emission of hydrogen is too much, and the risk of spark production is too great. Other than methods which begin with removing all the fuel and fuel vapor in advance, the safest alternatives proposed here so far are those involving extreme cooling, which disbonds the two metals by virtue of the fact of the aluminum contraction being sufficiently more than the steel contraction. Again, it's advisable to not be without a been-there-done-that advisor (a licensed professional engineer who has specific-to-the-task experence) on the groundbreaking frontiers of novel methods when a big underground tank containing gasoline is involved.

 

[HILSONHS]

I think this is being over thought. Here is my suggestion:

Purchase a 5 foot length of black iron pipe of an outer diamether that will easily slip into the 4 in. diameter tank pipe. The
iron pipe must have in inner diameter to easily accommodate the aluminum drop pipe.

Using a rotary tool (Dremel, hand drill, etc) grind a bevel on the _inside_ of the iron pipe so that it takes the shape of cold chisel.
You should only need to grind a quarter of the inner diameter of the iron pipe. Don't grind the outermost lip of this semi-circular chisel
edge too thin. Dress the the outside edge of this cutting end so that it is smooth (no burrs).

Tie a rope or wire around the other end of the pipe that is long enough that if you accidently drop it in the 4' pipe you and retrieve it.

Now, if you can reach the aluminum drop tube try to separate it from the outer tube for a couple of inches using a hand held
cold chisel (coated with a litte axel grease to decrease the chance of sparks). Again, tie a string around the hand held cold chisel
in case you drop it.

(This is optional depending on how long the aluminum drop tube is: tie a string around the end of the aluminum drop tube and
thread the string through the previously prepared iron pipe. Make the string long enought that if the aluminum tube drops
into the tank you can retrieve it.)

Now the fun part: grease the ouside of the first foot or so of the iron pipe (spark prevention), slide it down over the aluminum
drop tube with the sharpend inner edge between the drop tube and the outher 4 in pipe. Raise and lower the iron pipe with
sufficient force to progressively "chisel" or separate the aluminum drop tube from the outer pipe. If you get all the way down
the aluminum tube you can pull it and the iron pipe chisel out.