Heat exchanger leak 1

[koonsdman]

Hi guys.
Just started my work as a process engr in a chem plant.
One of my 1st projects is on a HE which uses steam (shell-side) to heat up oil (tube-side). After the intended amount of oil is heated up & transferred to a mixing tank, the pump is turned off and system is closed, leaving oil in the HE . Steam continuously flows as it will be used in the next batch for heating oil again as each batch is considerably short. During this time, some oil leaks from both sides of the HE. We figured that the expansion of the oil in the HE is the cause for this (gaskets are good & there's no HE cap expansion). We already taught of adding a pressure accumulator on the HE, but could anybody suggest some other ideas?

 

[ash9144]

Leaks to where? Out of HEX or from tubes into steam side. How much temperature rise do you see between batches? Where physically is the leak (on the heads, valves, etc)?

 

[epoisses]

A pressure accumulator.. hmm... I'd rather put a steam bypass around the exchanger in case it is out of service

 

[gedomin]

Sounds like a typical "Safety in Design" issue.
A way for releiving liquid expansion must be insured. A PRV or procedural. This last at low or no cost.
Procedure must instruct to ALWAYS leave a pathway (be specific) via one of the block valves, left open, for expansion relief. Install a visible sign (banner) in the place, as a reminder.

 

[koonsdman]

Thanks for your replies guys.

ash9144 (Chemical)
29 Aug 07 9:53
Leaks to where? Out of HEX or from tubes into steam side. How much temperature rise do you see between batches? Where physically is the leak (on the heads, valves, etc)?

The oil leaks out of the HEx at its end caps (heads).
oil enters at 20C and reaches over 120C. Steam is at 140C.
Oil is pumped at 4 kg/cm2. When enough oil is pumped into a mixing tank, pump is turned off, valves closed and oil is left inside for about 4 hours.

As I've said, we might install an accummulator with just air inside at 0 gage. Not sure though if it'll work, I'm thinking that while oil is pumped thru, it'll just go inside the accummulator and same problem will arise.

Your comments, please.

 

[ssn596]

Vent one end of HX to atmosphere with enough vertical pipe whose volume = max expansion of oil volume + 10%.

Dan Bentler

 

[epoisses]

I guess accumulators and vent pipes would all be fine, but wouldn't it be better to take the root cause away? IMO it's just not good practice to heat up an exchanger while the other side has been blocked in. I'd simply bypass it and be done with it. To avoid operator error you could consider an interlocking system that allows the oil side to be blocked in only after the steam to the exchanger has been closed.

 

[ProcessWaterChemist]

oil enters at 20C and reaches over 120C. Steam is at 140C.

water boils at 100C and all oil has some water in it! If the wt % leaking is < or = to 1% of oil flow than this is your problem.

check % of H2O in oil before and after HE series.

Many ways to skin this cat...

Like,

don't take the oil from 20C to 120C. Try 20C to 90C in HE then send to mixing vessel with vent to atmosphere. Whatever process calls for oil at 120C could be met by running oil from mixing vessel back into HE on the way to the system.

 

[ProcessWaterChemist]

FYI The abbreviation for Heat Exchangers is HX not HE.

 

[Fudge30]

Well I thought it was HEx
(Maybe it's just me)

 

[djack77494]

Like others, I don't agree with the implication that the oil side is blocked in while the steam side is kept in operation. I would suggest that the steam side be isolated when the oil side is. If that is not possible, then I suggest that only the inlet side of the exchanger be blocked. Leave the outlet side open to its destination, providing a path that will not allow for pressure building up. Though I would be hard pressed to come up with a scenario where neither of these solutions was acceptable, PSV's or accumulators could also work, and actually a PSV is probably needed.

 

[jistre]

Others have already given some good suggestions about approaching a solution. I just wanted to ask a related question here.

How did the exchanger get pressed into this kind of service?

It just seems like you've got an exchanger here that is not well matched to the operation of your process. From what I'm reading, it sounds like what you have is an exchanger designed to work in a continuous process that is being used in a batch process. If I understood right, you're holding oil in the exchanger for four hours to heat it up in preparation to send the next batch through the process. This isn't really what these exchangers are designed to do; they are designed to heat continuous streams and just don't have the proper pressure relief built in to them for this. By blocking it in with the external valves with the heat source live, you get the leaks you see. Also, if this exchanger is leaking because of the pressure developed through expansion when blocked in, you're almost certainly exceeding the MAWP of the device. This seems horribly unwise and unsafe to me. If anyone gets hurt because of this mode of operation, the vessel manufacturer's attorneys will have a very easy time proving they weren't negligent during the lawsuit and your company will suddenly find itself very poor.

If I've correctly understood what you were saying, wouldn't a vented holding tank with some kind of heater and a downstream pump be more suited to your application?

Of course, if I've misunderstood your mode of operation, I still hold that heating a blocked in exchanger full of fluid is still unwise and potentially dangerous.