Steam Jet Ejector Condenser Performance 1

[grant464]

I have a 3-stage steam jet air ejector system that produces vacuum for a process. The two inter-condensers are direct contact, not surface.

The condensers currently operate on 75-80 F closed loop water. A big problem in the hot summer months is that our cooling tower cannot make water cold enough for the condensers. When the water temperature climbs to 90-95 F, we have to slow down the process significantly, or turn it off.

In theory, increasing the water flowrate through the condenser should allow us to operate at a higher water temperature (i.e. 90-95 F), since this is just a heat transfer problem. Does this hold true in practice, or are there other factors that need consideration?

I can also tell you, that the water inlet valves to the condensers are only 30% open right now. This is because the drain pipe was never sized correctly and water starts to back up into the condenser if the valve is opened too much. The maintenance guys found that 30% open gives the "best performance" with the current setup.

Also, the small amount of documentation I have on these VERY old condensers is: Maximum condensing water temperature 95 F.

I have ruled out purchasing a chiller to bring the water temperature down because of excessive cost! Our flow rates are in the order of 500 gpm.

Thanks for the discussion,

- Grant

 

[CJKruger]

grant464,

- To really comment on the temperatures we also need the absolute pressures at each stage.

- Check the size of the outlet drain pipe yourself. Maybe it is just partially blocked by deposits. Where does the water drain to, what is the destination pressure and how much is the elevation change ? Does each stage drain separately or are they tied together (before getting to a drum) ?

 

[grant464]

Sorry, I don't have the absolute pressures at each stage. The system, overall, is producing a vacuum of 720 mm Hg on a typical day.

The inlet/outlet temps on the condensers are:
primary - 84/102 F
secondary - 84/96 F

The outlet drain pipe is identical in size to the inlet, hence you can understand that if the inlet is left open 100%, the discharge is not going to be able to gravity drain at a sufficient rate. The outlet drain pipe is gravity fed down 4 stories, but has several bends in it, reducing flow. Each stage drains seperately back to a collection tank.

If the condenser's performance can be increased by increasing water flow rate, I will resize the drain pipes, that is not an issue.

Thanks,

- Grant

 

[rmw]

What type of internals do the intercondensers have? There are a variety of types and styles some better in my opinion than others. Whatever type you have, you need to maximize your heat transfer within the device that you have. That may be changing the sprays, curtains, trays, or whatever is in there. If you have any room for change there you may get more heat transfered to the condensers that you have.

Now, how about the jets? Are the steam pressures per the nozzle design? The steam temperature? A jet sized for a given pressure saturated steam trying to operate on superheated steam can do different things than it is designed for.

What about wet steam? Wet steam can cut things to ribbons in jets causing them to lose performance.

What about steam leakage around your nozzles? Leakage around the nozzle ends up as load for the condensers without contributing to jet performance.

What about strainers in the steam line in front of the jets? They can rob the jets of pressure needed for proper performance.

Some fine tuning may bring the jets back to design and solve your problem.

I have encountered few pristine jet systems that were operating at the design conditions on the nameplates.

rmw

 

[grant464]

To answer all of your questions...

The condensers have a spray nozzle design. So, while increasing flow rate, I may need to replace the nozzle with a bigger one?

The jets are rated for, quote: "minimum 150 psig steam". I know that we run about 200 psi through them.

I can test for whether the steam is superheated vs. saturated by measuring temp and pressure. I will do that.

I am unsure of the steam quality, other than that we do have some steam traps installed. I am not sure how to test for quality (unless some kind of enthalpy gauge exists!).

Strainers are checked, ok.

- Grant

 

[CJKruger]

grant464, I would approach the problem in two stages:

1) Stage 1. Do a few calcs yourself to see if anything stands out. For example:
- First stage outlet pres = roughly 100 mmHg (40*(760/40)^(1/3))
- Calc the water temperature after condensing steam. You can accurately calc the ejector steam flow from the nozzle throat size and steam pressure.
- Look at steam tables to see if temperature is low enough for your pres.
- Next, do the hydraulic calc for the drain line. Properly with elevations, pressures, fittings, etc.
- If you also condense hydrocarbon, account for that in the density and vapor pres calcs.
- How is the vessel pressure controlled ? Maybe that valve leaks (spillback, etc).

2) Stage 2. Contact an ejector vendor (Korting, Graham, etc) and ask them to review it and make recommendations.

 

[grant464]

Thanks, I will use your advice to proceed.

Last question, I promise...

From what I mentioned before, we run at 200 psi steam, but the ejectors are designed for 150 psi steam.

Since the nozzle is a fixed diameter orfice, does this mean that we are getting higher capacity at the cost of reduced system efficiency? Or, is the capacity of the ejector system virtually unchanged, even though we are using more steam?

- Grant

 

[CJKruger]

grant464,

The ejector nozzles are (almost) all critical flow. So increasing the steam pressure will increase the steam flow (per nozzle equation).

Whether this extra steam improve the vacuum, is a separate issue. The problem is that the ejector body may also be of a critical flow design (you get them in both designs). That means, it chokes at the start of the diffuser. Increasing steam flow will thus increase the back pressure, and not necessarily create extra vacuum (but then again, it may, only the vendors can tell for sure).

 

[rmw]

To answer your question look in your steam tables. The higher steam pressure, the lower its specific volume. The nozzle has a fixed area so if the steam is of a lower specific volume, the nozzle will pass more lb/hr of steam. More steam in a jet doesn't necessarily mean more performance for the jet but it always means more load for the condenser.

If you can pressure reduce your steam from 200 psi to 150 you will get two benefits. One is that you will get a slight super-heating effect due to the adiabatic expansion across the control valve and the other is that you will reduce your steam flow through the nozzle by operating it at its design pressure. The jet will still perform as designed if everything else is equal (no nozzle or body wear.)

If your jets are designed for saturated steam at a given pressure and your steam is superheated the specific volume for that temperature/pressure will be higher than for saturated and less steam will pass through the fixed area nozzle.

The superheat I mentioned in the pressure reduction suggestion above will be slight but it will help guarantee that you don't have wet steam. The steam off the nozzle is going sonic speed and moisture in the steam can cut the ejector body to ribbons as you can imagine.

About your spray nozzles, are you sure they aren't clogged or worn so that the spray patterns are disturbed? Go to

http://www.sugartech.co.za

and rummge around the site. Especially this;

http://www.sugartech.co.za/condenser/index.php

Whatever you can do to maximize the contact area of your water and the steam to be condensed will be helpful (unless of course your water exit temperatue from the condenser is saturated for the condenser pressure-if that occurs, then your condenser is doing all it can do.)

rmw

 

[grant464]

I appreciate the discussion guys, I will look into it.

Thanks,

- Grant

 

[davsy]

good reply from rmw. All that excess steam pressure at the inlet to the ejectors is going to hurt you rather than help you.

This is because the drain pipe was never sized correctly and water starts to back up into the condenser if the valve is opened too much

What type of installation is this, do the "drain lines" drop from height into some form of hotwell or seal tank.

If this is a barometric installation then the discharge piping from the condensers can cause all sorts of problems if, as you suggest,the pipes where never sized/installed correctly. What do you actually mean by "not sized correctly".

Typical multi stage steam jet vacuum equipment with direct contact condensers have quite specific requirements for the discharge piping. Normally the equipment vendors suggest the barometric legs be at least the same or possibly one size bigger in diameter than the outlet nozzles on the condensers. Also the discharge legs should run as close to vertical as possible with any elbows or bends, if unavoidable, being 45 degree maximum and normally not within the first 8 or 10 feet from the condenser outlet.

If you are a bit tight on the cooling water side any errors in the barometric leg sizing or installation is only going to make things worse. The water inlet valve being resricted as much as you suggest could well be an indication of this.

 

[CJKruger]

davsy, I would size the drain pipe and outlet nozzle to be free draining or self-venting (v < 1ft/s).

 

[chief]

A 3-stage steam ejector normally incorporates a check valve in the last stage to maintain condenser vacuum. Ensure it is not jammed open or worn.

Offshore Engineering&Design

 

[grant464]

The condenser is barometric, so it relies on having a fairly long drain pipe, around 40 feet vertical into a collection tank. And there are some bends in the discharge @ 45 degrees. I am looking into the design specs vs. what is installed.

- Grant

 

[davsy]

davsy, I would size the drain pipe and outlet nozzle to be free draining or self-venting (v < 1ft/s).

CJKruger
I would tend to agree with you on this however I assumed that the condenser had been supplied by the vacuum equipment vendor and as such there would have been (probably) no customer involvement in the sizing of the outlet nozzles. It is quite common (well in all the Koerting, Graham and Applied vacuum sets that I have installed) for the installation instructions to mention the diameter of the barometric legs being "equal to or larger than the condenser outlet nozzle". It is a rule of thumb from the vacuum companies that works but caculation would give an exact answer.

Grant

If this is, as you say, an "old" barometric installation have you pressure tested the barometric legs recently. I have had experience in a couple of plants where there has been a leak around the water level in the collection tank.

 

[rmw]

Graham had (I did not go back and check this) a good technical paper on their site

http://www.graham-mfg.com

concerning condensate legs and sump tank precautions.

rmw

 

[katmar]

There have been some good comments regarding the steam and the jet nozzles, but if the problem manifests only in summer the chances are that your first bottleneck is in the condensers as you originally suspected. Fix that first and then fine tune the jets.

The water flow will have an influence on the condenser capacity. The outlet temperature is what you need to work with. A well designed condenser will achieve an approach of 5 or 6 degrees F - i.e. the difference between the outlet temperature and the saturation temperature of the steam. Increasing the water flow keeps the outlet temperature lower and helps the approach.

If there are no horizontal sections in your drain pipe (i.e. all are 45 degrees or steeper) you should be able to get away with it. I have done that successfully. But it makes the requirement for self venting sizing more critical.

If your 500 gpm flow is to each condenser then you need a 12" drain. If it is split between the condensers then each drain can be 10". I agree with CJKruger that self venting design is important, but I believe he is being unnecessarily conservative with his recommended superficial velocity. In a 12" pipe you can go up to 1.7 ft/s and still be self venting. And to 1.5 ft/s in a 10" pipe.

Some condenser designers do deliberately design the drains small to get siphon flow rather than self venting flow. The theory is that siphon flow will suck some bubbles out with the draining water and increase the condenser capacity. I don't do this for 2 reasons. Firstly the entrained bubbles make the fluid in the drain pipe less dense and it therefore has to be longer than if filled with liquid water. Secondly, the siphon regime is fairly narrow and if you don't have just the right water flow you can have all sorts of problems. Its a bit more expensive to install self venting drain lines, but more reliable IMO.

Sorry for jumping in late - I have been away for a while.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[grant464]

Thank you.

- Grant