Natural Gas Immersion Heaters

What did the original equipment manufacturer have to say on the topic?

OEM is long gone...and the systems originally (1980) had different process solutions and have been 'repaired' and 'modified' several times.

RSM

It sounds as though the process equipment is not far off becoming recycled.

• Consider materials compatibility data for selection of alternative materials.
• Consider alternative heating methods: microwaves, induction, etc.
• Consider lab work to identify materials not listed that might be used as an alternative: google "Green Death Test" for ideas.
• Consider modifying the process such that the material is neutralised elsewhere before being heated.
• Consider substitute materials - can a surfactant do the job instead of an acid (rhetorical question - NNTR)?
• Consider process intensification: make the equipment smaller. "What you don't have, cannot leak." _{- with apologies to the late Dr. Trevor Kletz, acknowledged as a worldwide process safety guru.}

Thank you, I do appreciate input... The compatibility database is fairly sparse; we have considered a major change of heating, container, and size (all required to be feasible), but cost has put that off; the green death test does look interesting and viable for hard data; the process and the material are dictated by customer...kind of stuck with that mess. Thanks again.

One last detail...the dipping vats being heated are mild steel structure with a 316L liners...the heat exchange tubes are welded to both.

One last question...what specification(s)/standards should I investigate for welding?

Can you describe the pitting or provide some pictures? Is the pitting fairly uniform or is it heavier on the top/bottom vs the sides?

Are you sure you have 12" diameter tubes and not pipes?

I will dive right in, since the angels have left the building.

1) If your pickling rinse (temperature way above room ambient) is chloride based, and I suspect it is...I am shocked it has lasted this long. You are on the verge of or already doing something that is really really bad for the environment: fugitive release of hydrochloric acid gas from co-mingling leakage with the exhaust gases from combustion. You need to protect the heat exchange tube with an overweld of a higher alloy (at least I seem to recall that as one less expensive option). You need an entirely different grade of stainless steel that is not as susceptible to chloride pitting, SCC, etc. I would look at 316Ti, or one of the Moly steels, but be prepared for sticker shock.

You could even try teflon coating the outside of the HX tube as a thin layer, or look into something known as CeraKote. There might be a coating that will not interfere with heat transfer enough to matter in the long run, something that might be renewed periodically during outage season (if there is one).

2) Thick internal sleeves will not protect the 316 steel, and will retard heat transfer in a really bad way.

Thanks much...I have recently found that the chlorides are likely carry over from a (temporary work around) initial rinse...tap water. All of our process solutions start with grade b water.

I have started looking at chrome-moly steels but will expand to 316Ti...

CeraKote....ceramic coating...inside fire tube or out?

No sleeve...thanks very kindly.

CeraKote may be available in different classes. Some (I have a sample for my own evil designs) are room temperature air dry cure. Others require bake on.

These coatings are no thicker than paint. They can impart chemical corrosion resistance to various substrates. Part cleanliness prior to application is however, extremely critical.

I think it might stand up well to Grade B process water. If you get samples and get a successful test with a cheap substrate, you may have a eureka moment.

Okay, some clarification here would be helpful.

This is a for pickling operation, right? That means acidic pickling liquor. Often it means use of HCl. H2SO4, HNO3, H3PO4, and or HF are sometimes used. Assorted chlorides are also sometimes added. It won't matter from the perspective of chlorides if it came from the makeup water, or what was added.

The resistance to the pickling liquor will depend heavily on the chemistry of the pickling liquor. It would be very helpful to know what is being added, even if exact ratios are not disclosed.

.

Additionally, the temperature of the surface in contact with the pickling liquor can have a strong effect. Combinations of increased surface area and intermediate heat transfer fluid can help reduce the peak temperature in contact with the pickling liquor.

Here is a third try at a photo of the pitting on the stainless. The immersion heater pipes to be replaced are in (1) straight grade b and (2) 2% phosphoric with grade b.

Thanks. Knowing a little about the chemistry and the pic help.

What is the scale of the pic roughly? How many inches across do you think area is that is captured in the photo?

To me that doesn't look like pitting corrosion. It looks more like generalized corrosion. How long had the pipes been in operation?

As far as replacement materials. Aerated phosphoric acid with heat will preclude things like monel, titanium, 400 series stainless, plain carbon steel, bronzes and cast iron. Better stick with 300 series stainless, perhaps consider some higher grades.

Sorry 'bout that....the crevice is about 3 mm across the long way. There are a BUNCH of these, and localized is the way I put it. The whole picture is about 2cm x 2cm...The heater pipe is below...the far right is the flame injection point, with the worst damage in the first 45 feet and at turns in the pipe. Most of the rust are iron salts that carry over from the sulfuric pickling and settle on the pipe...not rust coming from the pipe.

Maybe it is pitting. I am uncertain now.

Since the corrosion is mainly at the flame injection point, one thing to do would be the make the first section of pipe thicker. Also, anything you can do to stave off boiling on the tube will help. If the fluid is being circulated, having flow directed at that first section of pipe may allow sufficient heat transfer to stave of boiling.

I suspect the buildup on the pipe and localized boiling is leading to build up of impurities and very acidic conditions, leading to excessive localized corrosions.

You might also consider tightening the requirement for chlorides. If the main loss is to evaporation, chloride levels will build up and could be pretty high after a while.

I think you about have it covered, as 2% phosphoric acid in grade b water is fairly aggressive. I think that excessive skin temperature is leading to iron phosphate type deposits, most likely. He could increase velocity of the fluid at the fire inlet end of the heat tube, with the caveat that flow-induced corrosion may be the result.

From the photo, it appears to be under-deposit corrosion due to concentration effects that assist in removing the passive chromium oxide layer.

I still think AP (OP one) could also attempt some means of flame temperature moderation?? They could also try CeraKote of various types on that heater tube to see if it reduces deposition/corrosion.

Yeah, you are right about the flame temp moderation....especially in the initial portion of the pipe.

I thought the idea of a coating on the inside was silly at first, but if it insulateds that first section a little and moves the heat transfer down the pipe some, they might be able to achieve sufficient heating without local boiling.

My first consideration was to put a refractory sleeve...say 1" thick... for the first 10 to 12 feet...where the flame lives...would that work to keep the initial section surface heat down? What about a ceramic pipe sleeve...similar to a radiant heater protection tube? I'm also considering reducing the btu output but weighing that against longer burns to maintain temp in the load.

If you can get away with longer lower heat burns, that works in your favor in a large way.

The very best would be putting the minimum in so that heat could be added around the clock (or as long as is reasonable per day given staffing, etc).

There are two other things that can be contributing significantly that you can have control over without huge changes.

.

First, buildup on the heat pipe should be minimized. The buildup is creating confined areas with limited flow that increase the likelyhood of local boiling. All the non volatiles, like phosphoric acid and some residual chlorides concentrate there. So intermittent cleaning and changes to flow that limit/minimize/correct buildup needs to be addressed.

.

Second, how much makeup water is being added to make up for losses (to evaporation etc) over the life of a pickling liquor batch? What I'm asking is how much water is add per day/week, for how many weeks typically, compared to the volume used in the tank/initially mixed?

If the regular makeup water is significant and/or a batch is used for an extended time, chlorides might be a lot higher than 1 ppm. They are bound to be much higher than the tank average in the localized boiling spots under the buildup anyway.

.

It might be cost effective to insulate the tank and improve cover/reduce surface area. Limiting heat losses will save on fuel, and reduce makeup water and reduce necessary heat input to further distance the pipes frpm localized boiling.

If you did that with a refractory material that has minimal/similar heat expansion coefficient to the heat tubes, it might work, but it needs to not be highly efficient as heat transfer retardant, almost effective at 1 mm thick if you could get that to stay in place.

Take care you don't insulate the lead section of pipe too much and end up with a hot spot down in the middle, or make the pipe inefficient to the point where most of the heat is going out the exit end.

Agreed. I am almost shocked you agreed with me on anything.

I suppose that is what makes debate interesting, and a departure from a featureless plain of boredom.

Of course there are things about which I agree with you....I can't help it, sometimes you are right.

Rarely, but sometimes yes. So are you, at least up to a point, but I would not take a poke in the eye with that sharp stick.

Right now my most likely candidate is INCOLOY 825...chrome-moly alloy with high stable high temp properties....cost yet unknown :O\

I suspect compared to 300 series stainless, the incoloy 825 will be very expensive, especially with the bends, and getting it welded (incoloy 825 and a lot of the 'inco-' alloys are more finicky than 300 series when it comes to cleanliness and proceedure).

If you keep the heat transfer to low rates and don't let buildup get excessive, you will be fine with a good 300 series stainess....if you are itching to spend more to make it more durable, use much thicker tubing especially close to the burner.

You might look at some sort of ceramic or even just stainless concentric liner with small holes drilled intermitently in the bottom for some of the initial section.....as long as you can access it after new install without having to cut then reweld. It going to be tricky to get the heat transfer just right to limit it meaningfully yet not so much that you just push the problem further down.

Probably spending the money on thicker 316L is the way to go.

You might consider shielding the areas around the weld from some internal heat so that boiling doesn't occur in the HAZ or in the weld.

Also, if you can reduce heat flux sufficiently to avoid boiling, the material will typically last much longer. This can be accomplished in a number of ways....slower heating from cooled states....insulation of the tank....more surface area of the heat pipes, i.e. 4 6" tubes instead of 1 12" tube.

Of course thicker material will also typically increase longevity for a couple reasons.

Hey, I forgot to ask an important question:

What is the operation of the heater like?

Does it run continuously? Cycle frequently? Off more than on, vise versa? Adjustable or on/off?

This is an off/on constant fire Eclipse Immersopak...mostly on. I'm working on some heat load/loss estimates to see if I can reduce the size of the burner...I'm also updating the controls with digital flame control.

Yes, having some turn down on the burners would be better than cycling.

I wonder if you could make the first straight run with flanges at or near both ends for periodic replacement of that vulnerable section. I have some skepticism of this idea, but still it might be worth a look.

I've thought about that...I would need very high temp and impervious gaskets and hardware...very tempting!

If you make it out of 316L cutting out and welding in a replacement pipe for that first section could be done pretty quickly, possibly not that much different than replacing a flanged section depending on the gasket and what kind of pain in the ass the flange bolts decide to present.

316L is typically very conducive to great welds and a pleasure to work with.

What I'm getting at is you should have no problem finding a decent welder with 316L experience who can reliably replace a section of 316L pipe for you in a hurry. If on the other hand you use one of the nickel based 'inco' tubes, it becomes a little more tricky, the pool of experienced tallent narrows greatly...but the chance of having to replace that section in a given time is also much smaller.

How will the 316L hold up to the chloride residual with left-over pickling juice? It does not seem to like chloride with organic acid, or am I falsely blaming the failure of another alloy on this one?

No, your concerns are legitimate and if no action is taken to reduce local concentrating of corrosive species via porous fouling and high heat flux (high local OD pipe temp) then 316L will probably provide a similar life/wall thickness it did up to this.

If however, some relatively inexpensive changes are made to minimize/prevent localized concentrations of chlorides/acid, 316L should prove far less expensive jot just initially, but over the long haul.

There are several ways to minimize peak temperatures on the pipe OD:

Lower heat input requirements via improving tank insulation and minimizing uncovered surface area (this will also reduce chloride buildup as less makeup water will be required)...

....which reduces demand for heat input. Extending heating times to the longest practical will also reduce the required heat input.

Using a thicker schedule pipe for the first section will reduce heat transfer (316 is not a great conductor of heat. At least compared to many other metals) and therefor reduce peak heat flux by smoothing the output by pushing more further down the pipe.

Steps need to be taken to minimize material cleaned off the work material does not build up on the heating pipe or is removed regularly before localized boiling occurs.

Nucleate boiling has some characteristic that could make it ideal for an indicator of build up that needs attention.

First is that the formation and aubsequent implosion of tiny steam bubbles creates a sound most people who have cooked pasta are familiar with making it an easily noticeable indicator inany situations without much if any additonal equipment....or something as cheap as a stethoscope.

Second, the surface doesn't dry out with nucleate boiling, so not much concentration is occurring.

Third, nucleate boiling is incredibly efficient in terms of heat transfer, so there is a little safety zone of comfort knowing that heat transfer rates would havr to increase a lot of flow would have to decrease a lot to transition from initial nucleate boiling all the way to departure from nucleate boiling to dryout....

Anyway, regular audible checks for nucleate boiling would be a sign to remove build up, improve flow, and or reduce heat input.

Keep dryout from occurring and things will last. But if that can't be relied upon, go with the expensive option.

I agree that one does not want to hear from localized hot zone explosions of steam in the bath. More of a crackly sound? Not a bang, or thump.

As you are not sure about the outcomes you should consider taking professional help, my brother took furnace installation NJ services, they also provide maintenance services. One should always take suggestion from an expert before trying anything on your costly HVAC units.

The entire problem is they may have previously listened to some "professional" expert, and ended up with this problem where 316L is totally inappropriate for the services they are asking of it, stress crack corrosion appears to be the primary failure mode, where 316L is in contact with acidic (or even scale-forming) waters with high chloride concentrations (brackish to seawater chloride levels).

Have a nice day.

This acquires the need of preheating the flammable gas previously the weight decrease. Warming the gas previously transmission can conquer this Joule Thompson impact (more - here ). The preheating of flammable gas makes the temperature of the gas rise enough so that after weight throttling and temperature diminishment.