Steam Coil Failure

Vincent Byrne

I am with a Consulting Engineering firm in NYC. We have a problem where new (one to two years old) steam coils have developed leaks.

We have indoor, sectional air handling units (AHUs) made up of return air/outdoor air plenum, return air fan, steam heating coil (two coils, stacked), cooling coils and a supply air fan. The systems are 30,000 cfm with VAV operation.
The steam coil will assist with morning warm-up and temper the mixed air when needed. The coils are robust for their service, i.e. plenty of capacity.

The coils are freeze-proof construction – tube-in-tube design, rated for 200 psi at 400 degrees, the inlet steam pressure is 2 to 3 psi maximum, there is a modulating valve per coil to control capacity, there is a check valve vacuum breaker that is piped to the condensate return, an F&T trap at the condensate outlet and the steam return is a vacuum system.
The coils are from a very reputable manufacturer (HeathCraft/AeroFin type firm) and are not from the same batch as they were installed over a two year period.

In the short time frame these units have been installed, small cracks in the outer tube (the condensate tube) have appeared. This has occurred on one or two tubes in half the coils. The cracks occur close to the header, that is, between the header and the coil frame.

There has been no notice of noise from water hammer or other forces (flashing?) but do not rule this out.

Any ideas why the coils would crack?

DanHolohan

Is that vacuum breaker

on the return side of the unit? I'm thinking that it would be better if it were between the control valve and the coil. That way it could admit air into the top of the coil so that the condensate could drain when the control valve shuts. It might be that the air isn't making it up into the coil to rid it of condensate.

Tony Conner

Are These...

...splits, or cracks? When I think of tube freezing and splitting or water hammer damage, it's almost always (if not always) axial - along the run of the tube or pipe. When I think of cracks, they're across the tube, often in a stressed area, like near a weld. The failures aren't related to thermal expansion/contraction of the coil assembly, are they? Is there something in the metal properties and thickness, or welding/brazing process? I worked in a plant with a large intake air coil, and we had several tube failures due to cracking. They were copper tubes brazed into steel headers. At high loads, there was quite a bit of vibration in the plant, and the copper tubes were work-hardening, and cracking.

Vincent Byrne

The vacuum breaker

is between the control valve and the coil inlet.
There is the train of thought that a second vacuum breaker should be installed between the coil outlet and the F&T trap. A second vacuum breaker appears on some steam coil details I have found but it is not something I have historically included on my projects.
Further, there are older existing units with the single vacuum breaker piped to the return that have not experienced tube failure.

Vincent Byrne

The coil failures

are not axial (or longitudinal) but across (traverse) the tube.
I was not aware that water hammer failures are typically axial. This information would make water hammer less likley. Thank you for that.

My simple answer to your question regarding damage from thermal forces is that this is a steam coil that had to be engineered for thermal changes. The coil manufacturer is one of the biggest in the country.

The steam header is cast iron and the brazed joint of the copper tube to the header is on the inside of the header. The failure is about 1/4" to 1/2" off the header.

Lastly, I believe our system is very lightly loaded.

Tony Conner

It Could...

...still be water hammer affecting the zone around the brazed joint, I suppose. I have seen water hammer break cast iron 90*s. Is there anything in the boiler water treatment that would go after copper? What type of steam control valve is installed? Is is sized for the actual steam load, or what the coil is actually capable of doing? It isn't a solenoid (yuck) steam valve, is it? Is there a trap ahead of the control valve (any style) so there isn't a slug of water being shot into the coil every time the valve opens? Plus, an oversized solenoid "hunts" full open to full closed times per minutes, and just slams everything. (I don't like solenoid valves on steam.)

There's usually some weird interlocking of a couple seemingly unrelated factors that results in failures like this. Good luck.

George_10

condensate

Try checking the pH of the condensate at these locations. If the pH is lower than 7.5 then you are starting to produce carbonic acids which will thin the metal of these coils. If the pH is greater than 8.5, then to much steam line treatment is being used and this will cause the metal to become brittle. This is something very easy to check and will eliminate any water chemistry issues.

Scott

DanHolohan

I don't see a need

for the second one.

DanHolohan

And

just thinking out loud. If this is a vacuum system, and there's a vacuum breaker on the coil then won't that be open all the while the control valve is closed? Vacuum maker vs. vacuum breaker. Could be a lot of air moving into the coil at that point. Oxygen corrosion perhaps?

Eric

Control

I assume you have DDC control system.
Take a look at the trend log on the DDC system control valve output. bad time of year right now.

I'd bet the steam valve is cycling like mad. Here's why.

Alot a times a single control loop will control the mixed air dampers and heating coil. Control contractor tunes control loop for dampers because they are used more often (hours of operation). This leaves a unstable heating control loop.

Steam control needs own control loop or reset PID parameters.

Do you have 1/3 2/3 split on steam valves?

Eric

Control

My topic was geared toward water hammer.

On steam side. How low is trap to bottom of coil? Also any other traps in vicinity which could be causing backpressure in condensate?

Also take a look at statification in mixed air plenum. How are dampers arranged (return air on top?). And opposed blade or parallel blade?

Vincent Byrne

The vacuum return will

tend to keep the check valve closed. The check valve will open when the vacuum in the coil is greater then the vacuum in the return piping.

Vincent Byrne

The source of steam is

Con Edison. There is a utility distribution steam service in NYC the is a by-product of Con Ed's electric generating stations.

We have a modulating steam valve and have the main traps that insure dry steam is being delivered to the coil.

I am leaning towards stress failure at the point. Problem is proving the point.

Vincent Byrne

Enen though theseI will recommend

> Try checking the pH of the condensate at these

> locations. If the pH is lower than 7.5 then you

> are starting to produce carbonic acids which will

> thin the metal of these coils. If the pH is

> greater than 8.5, then to much steam line

> treatment is being used and this will cause the

> metal to become brittle. This is something very

> easy to check and will eliminate any water

> chemistry issues.

>

> Scott



the teating of the water.

Vincent Byrne

Even though the

cracks do not appear to be from loss of metal I will recommend testing the condensate to rule this possibility out.
There are many older coils in the building that are not failing from corrosion.

DanHolohan

Just so I understand

Is the vacuum breaker equalized into the return line, or is it open to atmosphere? Can you sketch and post any of this, Vincent? Thanks.

Vincent Byrne

In reply to both

of your messages the controls are independent, mixed air and tempering.
The trap is 12" below the coil outlet and there is a vacuum return so back pressure is not likely.

The mixing plenum is up stream of the return air fan so stratification is not an issue.

Thanks for your input.

Tony Conner

Can You...

...take an example of a typical failure to a lab for analysis? The water treatment guys should either have the ability to do this, or at least be able to refer you to somebody that does this kind of work.

I'm really leaning toward a mechanical cause like something to do with the wall thickness or temper of the copper, fabrication stresses, thermal expansion & contraction, vibration, or a combo of some or all of these. This doesn't strike me as an installation or a water treatment/chemical problem.

Jacob Myron_10

Tubes cracking

Hi Vincent:

I am retire now. I did a lot of work in buildings in Manhattan whee stem is supplied by Con Ed.

In reading all the replies in regaurd to the problem includings Dan's possible cause due to corrosion because oxygen is allowed into the coil via the check valve vacuum breaker which by the way I have seen occur in areas where cloils were work hardened or thined out because the copper was bent, flared or rolled to accomodate a shape or mechanical connection.

I believe the problem is at an installation error.

Here is what I see happening (hind site) Large very cold coils being heated by steam a large volume of steam in a very short period of time ( fast opening valves ) the steam coils come up to inlet to high temprature quickly. At the return manifold the coils that heat up the quickest ( out of full air flow path ) are the coils that suffer the most from expansion and contraction. the flexing of those coils against a rigidly held manifold I belive te are the ones that are cracking.

I think the remidy for the problem is to allow the return manifold to move longitudinally. This will allow the expansion movementto be diverted away from individual tubes and transfered to the entire coil assembly.

The steam side of the coil can be held rigidlly and the entire coil will be allowed to move accross the fan unit.

The other alternate would be to cause the steam to enter the coils in step stages over a five minute period of time.

This would allow the coils to warm up slowly. By doing this the expansion of the coils near the manifold would occur very slowly and possibly less frequently.

Based on the information given I offer the above as a cause and remidy. I away on vacation now and will return around august 1. If you wish to continue this conversation further feel free to contact me.

jake

Andy N._3

hey Jacob

Is your e-mail address correct?
my e-mail bounced back