15 replies to this topic

[stefanb]

Hi,

 

Even if I read for long time the topics this is my first question here.

 

I would like to get an idea of what should be the design pressure drop on the steam control valve going to a Kettle reboiler for proper control.

 

I mentioned that is no de-superheating after steam control valve going to reboiler. Steam de-superheating is done upstream after the pressure reduction valve.

 

Also what is the appropriate type of valve.

 

Thanks!

[Bobby Strain]

You don't necessarily need a steam control valve. So, in this case the pressure drop is nothing.

 

Bobby

[stefanb]

I will be more than happy to eliminate the valve but I am afraid this is not possible. We control with it the reboiler duty needed to achieve a certain temperature in a tray column doing super-fractionation.

 

Using condensate level for control is not an option since system was not design like that and the reboiler are huge as they are and still not enough for our needs.

 

I have to specify the process conditions for new control valve and I don't know how to calculate the best pressure drop across valve for optimum control.

 

I appreciate if somebody has more info regarding this subject.

 

I found and read this article but it's using for calculation purpose Umax and Umin that I don't have.

 

http://www.docfoc.co...tling-reboilers

[breizh]

Hi ,

Consider these resource to support .

 

Probably good to check with vendors too .

 

 

http://www.cheresour...ol-in-reboiler/

 

Good luck,

 

Breizh

Edited by breizh, 09 May 2016 - 12:51 AM.

[stefanb]

Thanks for your replay. I read already the discussing pointed by your link, before posting the question and is not helping me.

 

The article posted is a good general one but is not addressing my issue.

 

Nevertheless I manage to calculate the dp using Hysys. In the end I found that I don't need Umax and Umin but reboiler heat duty for two cases, checking that no cross temperature take place.

 

Thanks!

Edited by stefanb, 10 May 2016 - 06:22 AM.

[Bobby Strain]

So, why not share your results with all of us? And, you can control condensate from the tube side of the kettle. Level control is not necessary and usually won't work. You simply connect your tower temperature controller output to the condensate valve. The cost is just a fraction of the much larger steam valve.

 

Bobby

[Pilesar]

stefanb,

You should plan for at least 10 psi pressure drop across the control valve at maximum flow to provide control. The appropriate type of valve is a steam control valve designed for the conditions. The technical sales engineer working for your valve vendor should be able to guide you. A globe valve is not an unusual choice.

[stefanb]

Here is my issue. We have right now vee-ball valves not globe. The original pressure drop specified in the vendor DS was just 5 psig for all three cases low flow/normal and max.

Off course I don't believe this so we manage to measure the drop with field data and found that we have more, around 20 psig, for the low case.

My problem is that from control point of view we need some DP on the valve but if we accept to much then reboiler will suffer since superheated steam is used instead of saturated.

If you use globe valve for small dp then you end up with a bigger valve size.

 

Bobby sorry but I don't understand your control mode. You need to change a variable in the equation Q=AU(Tc-Tb) in order to control so as the article posted by me describes that you can either change area, by the condensate level, or you change Tb by the steam pressure. You can also change pressure column, in theory, but this is not used normally.

 

To control level inside reboiler, the condensate pot has to have a right alignment with the tube bundle. In my case the pot level is much lower than tube level so we can't use this method. Even if we can't is not advisable for our case since we do super-fractionations and the response time will be very low. Also this method of control is harder to tune.

 

From my hysys calculation i found 3 psig drop with valve at 90%open and 7 psig drop with valve at 35% vor the max and min case.

Edited by stefanb, 11 May 2016 - 12:10 AM.

[Pilesar]

Reading through this discussion shows that your original questions were answered. You added additional information but I do not find where you are asking additional questions. It sounds like you are concerned that if you reduce the pressure of the steam into the kettle reboiler with an adequate control valve, then your exchanger heat transfer will be reduced so that it limits your distillation capacity. You stated that the reboilers are not large enough for your needs. If you are trying to relax the reboiler constraint, then you have several options:

1) Increase steam chest pressure

2) Decrease process boiling temperature

3) Add additional exchanger surface area

4) Increase exchanger heat transfer coefficient 

Moving the duty controller to the condensate side would maximize the steam chest pressure for your system. Lowering the column pressure will decrease the boiling temperature. Consider whether modifying your exchanger is worth the expense. Enhanced tubes such as UOP High Flux may be cost effective in your situation. If the column is at a constraint also, then you will not get any additional capacity by just improving the reboiler performance.

[Bobby Strain]

stefanb,

     I am replying in the hopes that others can learn from this discussion. With the condensate control I suggested, the heat transfer area varies by changing the condensate flooding the tubes. You don't need to know or control the level, so a condensate pot is unnecessary. And, yes, it really works, and quite effectively. For some new distillation startup I have seen designers use the level pot and try to control the column by controlling condensate level in the pot. I fixed it easily by ignoring the level and controlling the condensate directly from the column temperature controller. If you understand that condensate is flowing from the lower tubes, it's not difficult to visualize that condensate will occupy a greater volume of tubes when throttling the condensate, thus changing the area available for steam condensation.

 

    And I don't understand your concern about control valve pressure drop. It should control effectively even with 2 psi pressure drop. After all, the heat transfer control is effected by changing the steam pressure, and hence the condensing temperature. Also, even if you have as much as 50 F superheat the tubes are still wet and there are no dry tube surfaces." If you talk, you only speak of the knowledge you already have. If you listen, you gain new knowledge." And, you misspeak when you say that condensate flow is too slow, when that is usually exactly the kind of response that you want. Again, you are simply postulating with no basis.

 

Bobby

Edited by Bobby Strain, 11 May 2016 - 09:15 AM.

[breizh]

Hi ,

You may find some interest reading this paper .

 

Hope this helps

 

Breizh

[fallah]

 

  With the condensate control I suggested, the heat transfer area varies by changing the condensate flooding the tubes. You don't need to know or control the level, so a condensate pot is unnecessary. And, yes, it really works, and quite effectively. For some new distillation startup I have seen designers use the level pot and try to control the column by controlling condensate level in the pot. I fixed it easily by ignoring the level and controlling the condensate directly from the column temperature controller. If you understand that condensate is flowing from the lower tubes, it's not difficult to visualize that condensate will occupy a greater volume of tubes when throttling the condensate, thus changing the area available for steam condensation.

 

 

Bobby,

 

Yes, it's not needed the condensate level in tube side to be controlled and the flow of condensate is normally controlled by cascade control via column temperature controller; but in transient conditions and process upsets it's needed, let say, the condensate FIC to be overriden by LIC (relevant to the condensate level in tube side) when low liquid level is detected, or the condensate control valve to be full closed via an interlock trigerred by low low condensate level all these protective operations just can be performed having the condensate level in a place other than tube bundle itself i.e. a condensate pot which shows the real condensate level inside the tubes...

[Bobby Strain]

There is a recent thread about condensate pot level problem. If you have a look at it, you will see one of the problems that arise when trying to control a pot level in a horizontal tubeside steam/condensate system. There are good explanation for this erratic behavior. And this problem is not uncommon, as I stated earlier. The solution to good control is to go direct from the temperature controller to the condensate valve. The pot can be equipped with level control to override and prevent steam blow-by, acting as a steam trap. Some folks like to base load the reboiler. UOP is one, and they suggest a scheme such as Fallah describes. With reboiler fixed load, there is no cascade from the column temperature.

 

Another point. You will notice all the articles that discuss condensate level control are based on a vertical tubeside thermosyphon with steam in the chest. It works well in this configuration. But almost never works with horizontal tubes where steam is in the tubes. And most of the authors have no experience with any other configuration.

 

Bobby

[stefanb]

Thanks you all for your replays.

 

I was just trying to understand what type of valves and what dp this valve has normally for best performance.

 

I found a discrepancy between the design data sheet and field measurement. My surprise was even higher after discussing with vendors and seeing that a slight change of 10 psig in dp across valve can change the valve size from 16" to 18". This was the reason of inquiring on this forum in the hope that I can find an answer.

 

 

I don't intend to modify the control strategy. I understand very well how both system works and what are the advantages and disadvantages.

I just want to know how this valve are normally design since I am not a designer engineer.

 

 

Thanks.

[Bobby Strain]

I guess we pretty much beat this subject to death. But, I am curious as to why you use a kettle reboiler in what you term a super-factionator? Kettles are generally used where vaporization is low. Super-fractionators  may vaporize up to 90% or more of reboiler inlet liquid with a kettle. Horizontal thermosyphons are typically used.  Is this a new unit yet to be started? And, maybe you can share the service. Then I'll follow up with you to see how it works. You can send a private communication if you prefer.

 

Bobby

Edited by Bobby Strain, 13 May 2016 - 12:47 PM.

[Zubair Exclaim]

Stefan,

 

First of all I still cannot understand why you are so much concerned about DP of control valve, There are lots of types and variables you can play with in control valve and can make DP of valve dance around.

 

Steam system usually would have enough margins in pressure to loose some at control valve. plus  if you trying to size the control valve just for replacing existing control valves just find out what your kettle reboiler pressure is intended to work on and what is down stream usual pressure of desuperheater. The difference you have is your freedom to loose on a control valve

 

As for control scheme i would stay with what bobby said, condenste removal is a better way of control, infact  the way you say your reboilers are not enough mayy mean you are flooding tubes on your reboiler and have not a effective real control.