Oil Water Separator Design

[doomster]

I want to modify a design on our existing gravity oil water separator so that it can cater 20 m³/h of oily wastewater from a wash bay. The highest oil concentration that was experience was 300 ppm. We have an existing oil water separator that has a capacity of approximately 2.4 m³ but still fails to comply to the Clean Water Act effluent requirements. The concentration of oil and grease in the effluent must be below 10 ppm for it to pass.

I have attached the snip of our design of our existing gravity oil water separator.

If you could also suggest a coalescing plates supplier that can supply it here in the Philippines much better.

Thank you!

 

[georgeverghese]

In most facilities, to get to a nominal 10ppm total oil in water, an IGF (induced gas flotation) or DGF (dissolved gas flotation) unit used. In some stubborn emulsions, demulsifier chemical is injected upstream into the feed. The clear water underflow may then be routed to this existing inground API separator, while the concentrated oily water skimoff is routed to a heater treater for further separation. Talk to companies like WEMCO who may suggest other equipment to suit your plant.

In my opinion, you wont get down to 10ppm with the suggested corrugated plate interceptor- CPI / coalescer.

If there are other nasties (non petroleum based) in the feed classified as scheduled waste effluent chemicals with very low permissible disposal concentrations, other more elaborate treatment processes may be required.

 

[doomster]

Is it possible to use a sand box after the conventional gravity oil water separator, to reduce the oil concentration in the effluent?

 

[georgeverghese]

Yes, you could, but I have not heard of plants where sand filters alone get the OIW content down to 10ppm. CPI assemblies with cross flow plates or other fibrous coalescers are commonly used in non oil/gas plants for oily water treatment prior to disposal, but I have my doubts you can get down to 10ppm.
Another popular DGF variant is dissolved air flotation (DAF).

 

[doomster]

Can I ask some standards in designing such separators?
The length to depth ratio, the minimum residence time for the oil to separate with water and any other parameters to be considered in the design.
Thanks

 

[BurrowX]

I've operated an oily water separator system in California and had to get oil concentration down from 500-600 ppm range to the ppb range for a combustion boiler. The process was basically as follows
1. Gravity separation in large tanks (50,000USGals)with weir systems
2. WEMCO (Air induced floatation) This system also used chemical coagulants to coagulant oil particles together and then the WEMCO bubbled and skimmed the oil into troughs
3. The water was sent through mixed bed gravity filters. Basically, a combination of sand and activated carbon.
4. This step is optional but pressure filters would add capacity to remove more oil particles.

Getting to 10 ppm will be difficult without chemical coagulation and the use of a system like a WEMCO for air induced removal of oil. I hope this helps and good luck.

Laron B.
Results. Not Recommendations
BurrowX
Sr. Chemical Engineer

 

[doomster]

Thank you BurrowX for your response. Sadly the management decide not to buy a ready made oil water separator for high efficiency due to the cost. That is why they want to push on the conventional oil water separator (gravity type). They want me to improve the conventional oil water separator, that's why i've decided to design according to the budget which is im not confident that it will not reach 10 ppm. Please see attach for your reference. Thanks!

 

[BurrowX]

If there weren't financial restraints this wouldn't be engineering! lol

First your design - Sand and activated carbon can remove oil but these systems are designed with an efficient way of cleaning them once they get fouled. The normally have pressure indicators before and after each filter system to know when they are getting fouled. Then the systems would flow water in the reverse direction to clean out the sand and activated carbon. If you do not have a cleaning system you will have one expensive "once-through" system. At ppms your seeing it would take 2 weeks (depending on your flow rate) to foul that and have no way of cleaning it.

A very low budget oil and water separator can be achieved by extreme retention time or extreme separation between the settled oil layer and where you decide to remove water from the vessel.

1. Extreme retention time could be achieved by multiple settling tanks or just slowing down the flow rate through the vessel, this allows for more oil to be moved by a weir system and for the oil to naturally collect together. You'd want to not disturb the oil layer forming above the water so a very low flow rate can do this.

2. I had a tank that was 25ft tall, the top 4ft was a thick layer of oil and the bottom 4ft was clean water. As long as you do not disturb the oil layer it's possible to separate a lot of the bulk oil this way. The oil layer was taken away by a vacuum truck and sent to an oil refinery. This may not be the case for you.

Alternative Option - I still suggest using a drop of coagulant, even if you chemically inject coagulant in your existing OWS it will improve operations by a lot. (I'm not sure if you're doing this now). Try to convince management to do a trial of just adding coagulant. You may need to inject it upstream in the inlet pipe to ensure the coagulant has time to react and work. 1 tote of coagulant, 1 small LMI pump very low cost.

I hope this helps! Good luck

Laron B.
Results. Not Recommendations
BurrowX
Sr. Chemical Engineer

http://www.BurrowX.com

 

[LittleInch]

luffy,

What you're trying to do is design an API gravity separator.

Just google that and gems like this pop up

https://www.tad.usace.army.mil/Port...equirements - Oil-Water Separators_Jan_10.pdf

Its API 421

https://www.academia.edu/8814919/AP...IES_APPENDIX_5_SEPARATOR_DESIGN_METHODOLOGIES

But even then you're going to be lucky to make 50 ppm never mind 10.

but the design of the one you have now is truly terrible. That must dump about 1000 ppm.

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