Air Cooler with Warm Air Recirculation Control

[shvet]

Good day, forum

We struggle with air cooler control loop design. We have a new air cooler:
- air cooler with warm air recirculation
- ambient air min/max -35/+35°C
- process fluid freezing temperature +43°C
- louvers control - automated with pneumoactuators from DCS
- fans control - first fan on/off from DCS, second fan VFD controlled from DCS
- process fluid control - thermocouple in air cooler outlet, signal to DCS
- cooler internal air temperature control - 2 internal thermocouples, signal to DCS

Vendor has provided us with manual where it described that there 6 cases exist for air cooler autocontrol:
- 3 for overheating protection (modes "cold summer", "moderate summer", "hot summer")
- 3 for antifreezing protection (mode "cold winter", "moderate winter", "warm winter")
And 1 more mode for manual start-up during winter using steam coils.

All seems OK but we can't realize how overall control loop works and finish process and instrument diagram. We can't imagine and explain to a programmer how algorithm works. We have 3 input signals and 5 output signals in DCS:
1. How to choose air cooler internal temperature that is controlled during each mode?
2. When do we have to switch one mode to another?

If anybody have detailed P&ID or algorithm how air cooler with warm air recirculation works please share it.

 

[moltenmetal]

Just my opinion, but an air cooler is the WRONG tool for the job of cooling a material with a freeze point of 43 C. You're trying to drive screws with a hammer- there is a better tool for the job.

 

[LittleInch]

I think you're confusing "mode" with "case"

Without seeing the detail, the vendor seems to have provided you with a number of different "cases", however unless there is some sort of step change between them in terms of control output, the control function remains the same.

The only thing that will change is the speed of the fan and the amount of air being re-circulated.

Without reading the process and control description presumably supplied by the vendor I can't be certain, but I would imagine that the system tries to maintain a fixed air cooler internal temperature by use fot he louvres and a fixed fluid oultelt temperature by use of the fan speed.

There may be times when if the fans are max speed then the control point fot he internal air temp is reduced.

Might be a bit of a process map that needs to be followed for all the combinations, but that's what control systems are for.

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

 

[moltenmetal]

Remember though: if you can't start it up, you can't run it.

 

[shvet]

air cooler is the WRONG tool for the job of cooling a material with a freeze point of 43 C

It is my fault (simplification). Air cooler cools column bottom product and 43°C is a viscosity theshold to pump it to storage.

You're trying to drive screws with a hammer- there is a better tool for the job.

We are just a detailed design Contractor and do what Principal and Licensor had prescribed. If we shall drive screws with hammer - ok, we do our best as better as possible.

I think you're confusing "mode" with "case"

Thank you. I thought I should use "program" language as my question is concerned only DCS.

Without seeing the detail, the vendor seems to have provided you with a number of different "cases", however unless there is some sort of step change between them in terms of control output, the control function remains the same.

The only thing that will change is the speed of the fan and the amount of air being re-circulated.

Without reading the process and control description presumably supplied by the vendor I can't be certain, but I would imagine that the system tries to maintain a fixed air cooler internal temperature by use fot he louvres and a fixed fluid oultelt temperature by use of the fan speed.

No. See vendor's manual.

https://drive.google.com/open?id=0By-iev9g_oHZZU8yM3VELWtSSzg

Remember though: if you can't start it up, you can't run it.

I did not understand what you are talking about.

 

[LittleInch]

Follow the manual.

Looks pretty straightforward to me.

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

 

[lilliput1]

You should therefore seek clarification from the Principal, Licensor and the manufacturer of the air cooler, not from us!

 

[georgeverghese]

It is a pity this air cooler vendor has only provided a mode operating narrative, but not a seamless control scheme that will cover all these modes. A senior process engineer familiar with setting up process controls should be able to do this for you after a few reads of this document. Once done, you may have to get the buy in from the air coooler vendor that the proposed scheme will cover all modes described.
As with all such complex schemes, process control loop configurations evolve with time before you settle down with something that is perceived to be as simple as it can be, stable and cover all operating cases/ modes ( and you have carry an increasingly irate project manager to boot during this evolution). So take your time on this loop / scheme before you declare it to the project team. Be prepared for field changes when / if things dont turn out the way you intended.

 

[moltenmetal]

OK, let's go at this again:

Imagine a shutdown situation. The air cooler is filled with a material which solidifies at a temperature above ambient temperature, and you lose flow. What's your plan?

Yeah, I know it's not water- we're not going to rupture the tubes- but we still have a problem, don't we?

What do you do when power goes down? What do you do when the material has solidified in the tubes as a result? You're going to try to "cook" it out so you can get it flowing?

You're presumably going to try to close louvers on the fan box and then attempt to HEAT the material in the tubes- the top of which are still probably exposed to the environment. OK- let's say that the whole thing is boxed in, so you can avoid the problem I've just described. Seems to me that you've just impaired the NORMAL functioning of the unit, just to handle a start-up case!

Personally I consider running a material which solidifies above ambient temperature through an air cooler to be bad design practice. It might be a necessary compromise of bother for capital savings if the unit is huge, though, especially if the unit rarely shuts down.

The licensor likely has experience doing this and thinks what has been recommended is just fine. Frankly though I'd speak with a couple USERS of this design first and see if they have complimentary things to say about this air cooler arrangement.

 

[LittleInch]

I don't think this stuff goes "solid", just very thick and viscous. - "column bottom product and 43°C is a viscosity threshold to pump it to storage" - but that's why the steam coil is there for start up mode in winter..

The operating modes are a set of options ranging from convection cooling controlled by the inlet and outlet louvres, fan driven with re-circulation control to start up ( total re-circ with a steam coil).

Not the simplest of cooling systems but equally not too hard either once you read the operating manual and get it clear in your head. Lots of combinations of temperatures and changing of control functions but quite doable.

LI

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

 

[shvet]

It is a pity this air cooler vendor has only provided a mode operating narrative, but not a seamless control scheme that will cover all these modes. A senior process engineer familiar with setting up process controls should be able to do this for you after a few reads of this document. Once done, you may have to get the buy in from the air cooler vendor that the proposed scheme will cover all modes described.
As with all such complex schemes, process control loop configurations evolve with time before you settle down with something that is perceived to be as simple as it can be, stable and cover all operating cases/ modes ( and you have carry an increasingly irate project manager to boot during this evolution). So take your time on this loop / scheme before you declare it to the project team. Be prepared for field changes when / if things dont turn out the way you intended.

Thank you for reply. I am that senior process engineer. I am an "ultimate/final" person in company is responsible for safe and stable operation and financial viability of designed facility.

but we still have a problem, don't we?

Yes, we have.

What do you do when power goes down?

Do you mean electricity or all kind of power simultaneously? If only electricity was shut down:
Winter - switch control loop to mode "winter 2" (see vendor's manual). If it can't help then lower process fluid flowrate to control fluid outlet temperature no less 43°C. If it can't help then switch control loop to manual mode, close intake louvers, close exhaust louvers, open recirculation louvers, open valves on steam coil inlet and outlet.
Summer - switch control loop to mode "summer 1" (see vendor's manual). If it can't help then lower process fluid flowrate to control desired fluid outlet temperature.

What do you do when the material has solidified in the tubes as a result?

Switch control loop to mode "winter 1" (see vendor's manual) and manually open valves on steam coil inlet and outlet.

You're going to try to "cook" it out so you can get it flowing?

Call to storage control room and warn them about emergency, open cooler bypass, switch control loop to mode "winter 1" (see vendor's manual).

Seems to me that you've just impaired the NORMAL functioning of the unit, just to handle a start-up case!

It seems to me you misunderstood. Sorry for my obscure posts. We don't intend to automate start-up or shut down. Emergency, start-up and shut down are to be done in manual mode. My question is concerned only normal operation - how air cooler has to be controlled using 4 input signals (ambient air temperature, first internal temperature, second internal temperature, process outlet temperature) and 5 output signals (intake louvers, exhaust louvers, recirculation louvers, first fan start/stop, second fan VFD)?

It might be a necessary compromise of bother for capital savings if the unit is huge, though, especially if the unit rarely shuts down.

It is a straight run vacuum gasoil hydrocracking facility (~180 barg) revamp, capacity is to be increased from 1.00 to 1.25 Mt per year. Process fluid is main fractionator bottom product 360+°C. Overhaul period is 3 years.

The licensor likely has experience doing this and thinks what has been recommended is just fine

Yes, it has a long reference list.

Frankly though I'd speak with a couple USERS of this design first and see if they have complimentary things to say about this air cooler arrangement.

We have no these USERS. We have nobody to ask. It is a first air cooler with air recirculation that is fully automated.

not too hard either once you read the operating manual and get it clear in your head.

I have done many times and this manual starts to be dreamed. I can't catch how:
1. to choose set point of internal temperature. Yes, I understand that vendor has provided equations in manual, but they give us a wide range of possible set points.
2. to define boundary conditions when DCS has to switch modes from "summer 1" to "winter 2" and back.
3. to control all 3 louvers (intake, exhaust and recirculation) simultaneously. Should it be linear and equal or stepwise or something else?
4. ambient temperature should be included in algorithm.

 

[lilliput1]

In HVAC outdoor air is only required for ventilation or economizer operation. I don't think the product need ventilation so I would think that if the product need to be cooled or heated below or above ambient respectively then only the recirculation damper be open and the temperature control is by modulating the cooling or heating medium. The recirculation damper would be closed and both outdoor air and exhaust damper would be open only if the ambient temperature matches the product setpoint temperature and the heating and cooling mediums would be shut off or diverted.

 

[shvet]

if the product need to be cooled or heated below or above ambient respectively then only the recirculation damper be open and the temperature control is by modulating the cooling or heating medium. The recirculation damper would be closed and both outdoor air and exhaust damper would be open only if the ambient temperature matches the product setpoint temperature and the heating and cooling mediums would be shut off or diverted.

It contradicts vendor's manual. According to vendor's manual in "winter 3" and "winter 4" modes 5 (five!) control devices are used simultaneous:
- intake louvers
- exhaust louvers
- recirculation louvers
- first fan start/stop
- second fan VFD
See vendor's manual, page 2, modes 3 and 4 (link above).

 

[lilliput1]

Draw a schematic diagram of the setup showing fan, product, air flow paths, heating equipment and heating medium control, cooling equipment and cooling equipment control, location of sensors and types. Indicate also desired product temperature and if fixed or range if variable/adjustable.