Air ejector troubleshooting

[Jordan-111]

We are experiencing "sudden loss of vacuum" or "pressure spike" on our air ejector system (2-3 times a day). The vacuum unit is a two-stage ejector with a direct contact inter-condenser. The unit is fairly new.
Design data: Operational data (when broken sonic wave): When running good:
1st stage suction: 1.3 inHgA 1.95inHgA
No 1st stage design discharge pressure 11 inHgA 9.8inHgA
2nd stage suction: 6.7 inHgA 7.88 inHgA 5.9inHgA
2nd stage discharge: 29.9 inHgA Constant backpressure(discharge to atmosphere) Constant backpressure
Cooling water flow: 80 USGPM 80 USGPM 80 USGPM
Inlet temperature: 67F 62 F 61F
Outlet temperature: 98F 94.5F 93.8F
Motive steam temperature: 500F 450F 450F
Motive steam pressure: 165PSIG 166PSIG 166 PSIG

We tried shutoff pressure test 3 times, they all generated different 1st stage suction pressure. (it could be that our instrument is broken)
I think the 2nd stage was broken due to excessive non-condensable loading, either from cooling water or there might be a leak somewhere in the system that causes the 2nd stage break operation.
Thanks in advance for the responses.
Regards.

 

[pierreick]

Hi,
Air ejector system? should be steam ejector system right.
What are the materials passing through the ejectors?
How do you recover the vacuum after loss? What actions are taken by operators?
Could it be Ice built up due to deep vacuum?
Is your ejector jacketed?
Trouble shooting should be at site, having all sensor in good condition.
Good luck
Pierre

 

[georgeverghese]

Direct contact inter condensor ? Is this a barometric type direct contact device ? So how is the hot condensate effluent cooled before recontact ?

 

[pierreick]

Hi,
Is your condenser similar to the one describe in the document attached?
Pierre

 

[Jordan-111]

Hi Pierre,

Yes, it is exactly the same as what is described in the document.

Jordan

 

[Jordan-111]

Hi Georgeverghese,

Thank you for the response. I am not sure I understand your question. But in Pierre's document, the inter-condenser is exactly what we use here.

Jordan

 

[georgeverghese]

So what you call "cooling water" is "condensate" in @pierre's link document. What do you use to cool this hot "cooling water" ? Presumably, the vapor lines to and from this inter condensor are not pocketed (ie they are free draining) as requested in the link document - Fig IV on last page.

 

[pierreick]

Hi,
So far, I did not get any answers to my first reply to your post.
Can we get a set of pictures or sketch of the system?
Can we get information about the material passing through the system?
Can we get the data sheet (specification of the equipment)?
Can we get a graph of the Pressure vs time?
What about the quality of the cooling water? Is the spray system in good condition? How is the flow rate of cooling managed?
Is the problem occurring at a particular time?
Pierre

 

[pierreick]

Hi,
Additional document.
Pierre

 

[Jordan-111]

Hi Pierre,
The pictures are attached.
The material passing through the system is a mixture of air (O2 and N2) and some carbon dioxide. Based on a theoretical calculation, the non-condensable loading is approximately 52 kg/hr, the ejector is designed for 90.75 kg/hr of non-condensable loading.
The datasheet is attached as well.
Graph of the pressure vs time is attached.
The quality of the cooling water is not so good, the water is coming from a nearby saline lake. I am not so sure on how to determine the spray system is in good condition or not (could you elaborate more on this?).
The cooling water flow rate is at 80 USGPM and it is controlled by a manual throttling valve.
The “pressure spike” does not have a specific pattern, but usually once in the morning and once around afternoon.
When the pressure spike happens, we usually backwash the ejector by closing the valve on the tailpipe from the inter-condenser for 30 seconds to flush the inter-condenser and 2nd stage. Then the vacuum will come back to normal.
Let me know if you cannot access those the attached documents.
Thank you for your time.
Regards,
Jordan

 

[Jordan-111]

Sorry, I attached the documents here.

 

[Jordan-111]

Sorry, I attached the documents here.

 

[Jordan-111]

Sorry, I attached the documents here.

 

[Jordan-111]

Hi georgeverghese,

We pumped 80 USGPM of "cooling water" directly from a saline lake nearby through the inter-condenser to cool the hot effluent from the 1st stage ejector. And it is free-drained pipe. The temperature of cooling water entering the inter-condenser is 61F and exit at 96.7F.
Hope this answers your question.
Regards,
Jordan

 

[georgeverghese]

So this is open loop cooling water from this saline lake you've got to run the intercondensor. So steam used to run the primary ejector is all lost to the saline cooling water going back to the pond.
With all the microbial load in this source cooling water, it is to be expected that there will be microbial fouling of the intercondensor and the stage 2 ejector, resulting in loss of vacuum generation.
Injecting hypochlorite into this cooling water may help to arrest microbial growth in the intercondensor / stage 2 ejector, but at the expense of increased corrosion.
The scheme shown in Perry's Chem Engg Handbook for this 2 stage ejector - intercondensor is with a closed loop operation for the coolant, which would not have this fouling problem. Modifications required to convert yours into a closed loop coolant may be a bridge too far and too expensive.

 

[pierreick]

Hi Jordan,
Thanks for sharing your documents. I suspect also the quality of the water to be responsible for the trouble. By chance do you have another source of clean water to perform a test? If not, can you perform preventive cleaning of the unit as you did, once or twice a shift to see if this is improving the performance.
To be frank, this is not a big trouble right now. As George pointed out, corrosion could be an issue in the future.
Could you share a typical analysis of your cooling water (Ph, Ca2+, Mg2+, turbidity or suspended solid, Cl-)?
Consider calculating or ask your chemist to calculate the Ryznar index to evaluate the risk of scaling of your water.
Pierre

 

[georgeverghese]

Saline cooling water from this lake exiting the intercondensor at 95degF will no doubt also precipitate out dissolved calcium and magnesium salts, so antiscale chemical injection may help some what. Glutaraldehyde is a less corrosive antimicrobial injection chemical that may be more useful here as a an alternate to hypochlorite. Talk to water treatment folks who may have other suitable chemicals also to arrest Ca/Mg scale deposition and biological fouling.

 

[Jordan-111]

Guys,

We are doing monthly lake water samples and I have attached results below.
From the results, there are magnesium, calcium and carbonate. But none of them are in a large amount.
Do you see anything that is out of spec?

Thanks,
Jordan

 

[pierreick]

Hi,
Thanks for sharing your data.
Based on the calculators attached (Ryznar and Langelier) the quality of the water is not too bad; you need to confirm with the supplier of chemicals for water treatment at your site.

https://www.lenntech.com/calculators/ryznar/index/ryznar.htm

https://www.lenntech.com/calculators/langelier/index/langelier.htm

The indicators show an ability of the water for scaling (calcium carbonate).
A deeper analysis with expert in the field (chemist) is required to assess the impact of the other cations.

Good luck.

Pierre

 

[georgeverghese]

The weeds you find growing in the lake may probably the same as that on the cascade trays in the intercondensor.