Hot Glycol Expansion Tank Sizing Basis

No, your calculation is not correct. You have given the steady state operating conditions of your hot oil loop. In steady state operation the volume of oil in the loop is not changing at all. An individual liter of oil may change volume, up and down, as it goes around the loop, but the loop volume is not changing.

The expansion tank has to accommodate the volume change of all of the oil in the loop, from its coldest condition when shut down to its warmest condition when in operation.

Hi Compositepro,

Compositepro said:

The expansion tank has to accommodate the volume change of all of the oil in the loop, from its coldest condition when shut down to its warmest condition when in operation.

Click to expand...

If I understood you correctly the volume of the Glycol in the loop should be equal to volume of the piping during the coldest months ( temp is -18 Deg C). I then calculate the Cubical Coefficient of expansion between this coldest temperature ( -18 Deg C ) to the warmest when it is in operation at 60 Deg C. I then use this value to calculate the change in volume to size the expansion tank is that correct?.

Also a paper by Ken Devore ( attached ) says that the Expansion should be 25% full in the coldest condition and 75% full in the hottest condition.

Thanks and Regards,
Pavan Kumar

You've got it.

Pls note that the expansion tank also acts as a "thermal buffer" to heat load changes in the system. Inadequate drum volume can mean large surges in the feed temperature of heating medium when there are multiple users, and the shock load changes originate from one large user which suddenly sheds thermal load. This is because of the inertial heat that is already present in the fired heater bricks and insulation, made worse by the fact that typical RTDs' used in TIC loops have a slow response to load changes. Failure to recognise this can lead to temperature control instabilities. Measures to address this are :
a)Beef up on holdup volume in the expansion drum
b)Use faster responding thermocouples as TT for TIC control at fired heater
c)Use thin walled thermowells to further cutdown on thermal inertia for sensing TT

This dos not apply when there is only one user.

Hi georgeverghese,

Thank you for your valuable inputs. I will post my calcs once I am done.

Thanks and Regards,
Pavan Kumar

Hi Compositepro,

I used the guideline in Ken Devore's paper attached, which says that the Expansion Tank should be sized such that in Cold condition the liquid should fill 25 % of the Expansion tank while in the Hot ( Operating Condition) it should fill 75% full volume. And this means the volume increase should due to thermal expansion should be accommodated in 50% of the tank volume. This is also reflected in the calculation shown in this link copied below.


However, Ken Devore's paper also says that following his guideline the tank would be sized for 30-50% of the total system volume which does not make sense to me. This is because if I use this guideline the Expansion Tank volume becomes huge. I want to know if your experience if you feel the same.

Thanks and Regards,
Pavan Kumar

Define "Huge"

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Hi LI,

I will illustrate this with the calculations I did for my system.

Heat Transfer fluid used = 40 % Ethylene Glycol
Vsystem= 114.5 ft3 (Piping plus Heat exchangers)
t1= -18 Deg C = 0.4 Deg F
t2 = 60 Deg C = 140 Deg F
Rho1 at t1 = 65.363 lb/ft3
Rho2 at t2 = 63.741 lb/ft3

V1 = ( 0.25 V + 114.5 ) ft3

where V is the Volume of the expansion tank being sized. Per Ken Devore's paper, the % fill in the Expansion tank in cold condition should be 25 % and 75 % in the hot condition.

m-cold = V1 * Rho1

V2 = m-cold / Rho2 = (0.25V + 114.5)* Rho1 / Rho2

Delta V = V2-V1 = (0.25V+114.5)* [Rho1/Rho2)-1]= 0.50V

Solving for V I get V = 5.904 ft3 = 45 US Gallons

V2 = (0.25*5.904+114.5)*[65.363/63.741) = 118.92 ft3

Volume increase = 2.95 ft3 = 22 US gallons.

Expansion Tank as % of Total System Volume = (5.904 / 118.92)*100 = 4.96%

This as per Ken Devore should be 30-60%. If the expansion Tank has to be 30 to 60 % of the total system volume then V = 0.30*118.92= 35.7 ft3=266 US gallons.

This is huge for a an expansion of only 2.95 ft3!.

Thanks and Regards,
Pavan Kumar

During system turndown to zero heat consumption, the entire inventory will be at the hot oil heater's exit temp. Is 60degC the hot oil heater's max normal exit temp?
Also, see point (a) in my previous reply.

OP,
The paper says, will usually result not should. Consider the paper and the link are describing hot oil systems. Hot oil has different expansion coefficients, larger temperature ranges, heat capacities and conductivities than glycol. If the maths are correct, consider the other posters recommendations as well. I will note, the paper describes the maximum level at being 75%, which would indicate that you should consider a hot volume level of less than 75%.

Hi,
Consider this link, different application but similar fluid. At least you will get some figures to compare with yours.
Part 2
Pierre

I'm really not sure what you've done wrong here, but your change of volume is, as you correctly say, 2.95 ft3 for in temp to max temp, i.e. the range of your expansion tank.

This represents 50% of the volume of your expansion tank if you go from 25% full to 75% full min temp to max temp. So your expansion tank needs to be 6ft3. Lets say 10 ft3 for good measure.

So that's 75 US gallons max.

Keep it simple in your calcs.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Hi georgeverghese,

georgeverghese said:

During system turndown to zero heat consumption, the entire inventory will be at the hot oil heater's exit temp. Is 60degC the hot oil heater's max normal exit temp?Also, see point (a) in my previous reply.

Click to expand...

Yes the Glycol heater's maximum outlet temperature is at 60 Deg C. It is a Biogas fired Heater. Yes I will take your suggestion for note (a) and LI's suggestion and add some extra volume for the expansion tank.

Thanks and Regards,
Pavan Kumar

Thanks Mr. Pierre and Heaviside1925 for your valuable inputs and reference material.

The expansion tank does not have to be kept hot. It can be teed into the loop so that only expansion and contraction cause any flow to the tank. The advantage is that the gas oil/interface in the tank can be cool. Also, the oil in the tank will not contribute much to the volume changes. However, low-boiling contaminants can accumulate in the tank.

OP,
I appreciate the gratitude you expressed. I will note LittleInch's advice. I only note this because I think he is getting to the root of your struggles, which appear to be you are looking to find a way to determine the exact volumetric capacity required for the expansion tank, instead of asking "what does the design basis dictate the minimum capacity of the expansion tank to be?" and then using good judgement select a size greater than the minimum.

I would like to reconfirm something georgeverghese has asked about which is max temperature and turndown ability of the system.
- Does the biogas fired heater truly have the ability of zero turndown?
- If not, then consider no and low flow scenarios not just the BMS controlling the outlet temperature.
- Does the heater have a minimum flow requirement in order for the BMS to control the outlet temperature?

A few more questions/comments for consideration, some or all may not apply but I don't think they have been commented on yet.
- Will the tank be atmospheric, or will it be gas blanketed?
- Will the tank be at the systems high point or will it be located near the same elevation as the pump?
- Will there be a need to hold the system volume, in case the system needs to be pumped down for maintenance?
- If a level glass is installed, ensure the height of the level glass allows greatest visibility of the entire tank
volume and not just the calculated minimums and maximums.

Pls note that many oil/gas operating companies are wary of the hazards involved in operating fired heaters, this being rooted in the poor reliability of burner flame detection despite 2oo3 voting detectors per burner. I've seen and heard of many incidents involving fired heaters myself, at least one resulting in fatalities due to hydro desulfurisation (HDS) furnace fired heater explosion. As a result, typical siting rule is to locate fired heaters at least 30m away from other process equipment and locations where operators frequent. This is, if I am not mistaken, applicable for both induced draft(not safe) and forced draft (preferred) heaters. Check your operating company guidelines and also talk to technical / process safety / piping engineers familiar with equipment siting. And account for increased pipe lengths in system heating medium inventory supply and return piping.

On the topic of turndown from @Heaviside, I vaguely recall you saying (in a previous thread a few weeks ago on this system) that this heating operation is cyclical and not continuous. So you would be running this fired heater flat out until desired temp TCH (temp control high) is reached in digestor, and then auto stop this fired heater, with pilot burner still on and main burner off ? This then implies field startup every time digestor contents hits TCL (which could well be in the middle of the night in a blinding Canadian blizzard) - it is NOT safe to remote - auto start fired heaters, since operators need to MANUALLY confirm main burner flame is ON - check with operators and project team design intent.

This previous thread may be of interest on the topic of inhibitor selection, mode of operation for the expansion drum:

Hi All,

I apologize for the long delay in my reply. I Thank you all for your invaluable inputs and notes. The expansion tank sizing at this phase only for cost estimation purpose only. I will keep all that you suggested in mind during detailed design.

Thanks a lot again for the invaluable help.

Thanks and Regards,
Pavan Kumar