What a Grease Interceptor Is Actually Separating

Why this matters

An interceptor is not a filter and it does not remove grease. It is a slow-motion settling tank that lets a density difference do the work, and the only currency it spends is time. That single fact explains almost every interceptor complaint a shop gets: a correctly sized unit passing grease straight through, a unit that tests clean at the outlet while the run downstream of it fills with hard white deposit, and an additive that makes an inspection sample look excellent while quietly making the problem worse. If you know what the tank is separating and what stops it separating, you can look at an effluent sample and say why.

Before a lid comes off

• An in-ground interceptor is a permit-required confined space, and a large one is a sewer with a lid. Nobody enters it. Servicing, sampling and depth measurement are all done from the surface. If entry is genuinely unavoidable it runs as a full permit entry under 29 CFR 1910 Part 1910.146 in general industry or 29 CFR 1926 Subpart AA on construction work, which is not a service-call activity.
• Interceptors are among the best sulfide generators in a building's drainage system: warm, organic-rich, and holding wastewater for a long time without oxygen. When the lid comes off, step upwind and let the headspace unload before working over the opening. Your nose is not the instrument here, for reasons a sibling card covers.
• Cast lids are heavy and the seat is a pinch point. Lift with a hook or pick, hands off the rim, feet clear of the arc.
• When taking an outlet sample or measuring depth, wear splash goggles and stand beside the opening rather than over it. Anything you dip goes down and comes up on a line, not on an arm.

What FOG is, and what "separating" means

FOG is fats, oils and grease, and the three behave differently enough that the word hides the problem. Animal fats are solid at room temperature and congeal somewhere in the range of roughly 90 to 115 degrees F depending on the fat. Vegetable oils stay liquid far below that. Emulsified fats behave like neither, because they are not a separate phase at all until something breaks the emulsion.

Separation is a density race. Oil is less dense than water, so a droplet rises; solids are denser, so they sink. Between the two, in the middle of the tank, is a clear zone, and the outlet tee is positioned to draw only from that zone. The rise is described by Stokes' law, which is derived for a single spherical particle moving slowly through quiescent fluid, not for a turbulent tank full of interacting droplets. Two things it says are still true in a real tank and both are load-bearing:

• Rise velocity scales with the density difference between the droplet and the water. Hot water is less dense, which narrows the difference.
• Rise velocity scales with the square of droplet diameter. Halve the droplet and the rise slows by a factor of four.

Interceptor in section. Flow enters left, leaves right.

   inlet tee outlet tee
      |                               |
   ___|_______________________________|___
  |   |     floating FOG cap          |   |
  |   |===============================|   |
  |   v                               ^   |
  |        clear zone: the only       |   |
  |        layer the outlet draws     |   |
  |   .............................       |
  |         settled solids blanket        |
  |_______________________________________|

The three things that stop it separating

Temperature. Hot influent does three unhelpful things at once: it keeps fats liquid so they arrive as fine dispersed droplets rather than lumps, it lowers the water's density and so narrows the density difference driving the rise, and it lowers viscosity in a way that helps rise but does not offset the other two. A dish machine discharging at 140 to 180 degrees F is delivering a well-mixed hot emulsion to a tank whose entire method is patience.

Emulsification. Detergent is engineered to keep fat suspended in water; that is what makes it detergent. It reduces interfacial tension so droplets that collide do not merge, and small droplets are the ones the square-of-diameter term punishes. This is the mechanism, not a side effect: a kitchen that switches to a more aggressive degreaser can push a compliant interceptor out of compliance without changing a single fixture.

Short-circuiting. The tank's rated capacity assumes the water crosses it evenly. It does not. A jet from the inlet, a missing or damaged baffle, an inlet tee that has broken off, or simply a rectangular tank's own hydraulics all mean some of the flow crosses in a fraction of the nominal time. The rated volume is therefore an upper bound on the working volume, not a design value, unless somebody has tracer-tested that specific tank.

What passes through, and where it lands

Grease that fails to separate does not disappear. It leaves the outlet as a milky or faintly turbid effluent that looks acceptable in a sample jar, travels down the lateral, and cools. Somewhere downstream the pipe wall is at ground temperature, which is well below the congealing range of animal fats, and the emulsion breaks against a cold wall. That is where the hard cream-coloured deposit comes from, and it is why the deposit is almost never in the interceptor and almost always in a length of pipe some distance past it.

This has a direct consequence for additives. Enzyme and surfactant products that "clear" an interceptor are often doing exactly the wrong thing: emulsifying the cap so it passes the outlet, which empties the tank and moves the deposit into the public lateral. Many jurisdictions prohibit them in FOG-permitted facilities for that reason. If a customer's interceptor cap keeps disappearing between services without a pump-out, ask what is being poured in.

Worked example: a unit that is the right size and still fails

An outdoor gravity interceptor, rated 750 gallons liquid capacity, serving a three-compartment sink and a dish machine. The facility is FOG-permitted, the last two effluent samples failed, and the operator's position is that the unit is the size the plans called for. That position is correct and irrelevant, and the arithmetic shows why.

Nominal retention time. Peak drainage flow through the unit during a dish cycle measures 40 gallons per minute.

• 750 gallons divided by 40 gallons per minute equals 18.8 minutes

Correction 1, the tank is not empty. The cap and the settled blanket occupy volume that no longer holds water. This unit is at its service trigger, which the sizing and service card defines as combined cap plus blanket reaching 25 percent of liquid depth; how to measure that belongs there. What matters here is the volume it costs.

• 750 gallons times 0.75 equals 562 gallons of working volume
• 562 divided by 40 equals 14.1 minutes

Correction 2, geometric volume is an upper bound. Nobody has tracer-tested this tank, so short-circuiting is unquantified and cannot be assumed away. The honest statement carries one inequality sign and no interval: actual retention time is at most about 14 minutes.

What that time has to buy. At the temperatures and droplet sizes a well-behaved kitchen produces, coalesced droplets on the order of 100 microns rise through the working depth of a tank like this in a few minutes. Call it 3 minutes, which fits inside 14 with margin to spare. That is why the unit was compliant when it was commissioned.

Correction 3, the droplets are not that size any more. The kitchen changed degreasers eight months ago. The effluent droplets are now on the order of 25 microns. Apply the square-of-diameter relationship, remembering it is derived for single spherical droplets in quiescent fluid and is being used here as an order-of-magnitude comparison rather than a design calculation:

• Diameter ratio: 100 divided by 25 equals 4
• Rise velocity ratio: 4 squared equals 16, so the small droplets rise 16 times slower
• Time required: 3 minutes times 16 equals 48 minutes

The finding. 48 minutes of separation required against at most 14 minutes available. The tank is not undersized for the flow; it is undersized for this emulsion by a factor of roughly three and a half, and no pump-out schedule fixes a factor of three and a half. Restoring the 25 percent of working volume takes the available time from 14 to about 19 minutes, still nowhere near 48.

What the answer actually is. In order of what removes the mechanism rather than what is easiest to sell: change the degreaser back to one compatible with a gravity interceptor, and confirm it in writing with the chemical supplier; get the dish machine discharge temperature down or give it a cooling leg so the fats arrive coalesced rather than dispersed; verify the inlet tee and baffle are intact, because a missing tee turns 14 minutes into far less. Adding a second tank in series is a real option and it is the expensive one, and it should be the last one considered because it treats the symptom of an emulsion by buying more time rather than by not making the emulsion.

What flips this. If the failing analyte were settled solids rather than FOG, none of the above applies: solids are a sinking problem, they are governed by the blanket depth and the pump-out interval, and the answer is service frequency rather than chemistry. If the facility runs vegetable oil almost exclusively, the downstream congealing story weakens considerably, because those oils stay liquid at ground temperature and will travel to the treatment works rather than depositing in the lateral - still a permit violation, but a different failure and a different conversation. And if the unit is a hydromechanical interceptor rather than a gravity tank, retention time is not the governing variable at all; those units are rated on a flow rate with a flow control fitting upstream, and the first question becomes whether that fitting is still installed.

How to tell which failure you are looking at

Four field observations that separate the cases, taken in this order because each one is cheaper than the next:

1. Look at the effluent sample against light. Clear means separation is working and the failure is elsewhere. Milky or uniformly cloudy that does not break on standing means an emulsion. Visible floating droplets that coalesce in the jar within a few minutes means the tank simply ran out of time and a bigger or better-baffled tank would help.
2. Feel the influent pipe. If it is hot to the touch during a dish cycle, temperature is in play regardless of what the sample looks like.
3. Measure the cap and blanket from the surface with a corer or a sludge probe, never by entering. If the combined depth is well under the trigger while the effluent is failing, service frequency is not your problem and you can stop selling pump-outs.
4. Ask for the chemical inventory. A degreaser change, an added enzyme dose, or a new pre-rinse product will not appear on any drawing and is the single most common thing that changes between a passing sample and a failing one.

References

• 29 CFR 1910 Part 1910.146, permit-required confined spaces, general industry; 29 CFR 1926 Subpart AA for construction work
• The plumbing code and the FOG or pretreatment ordinance as adopted and amended by the local jurisdiction, which set interceptor requirements, effluent limits and any prohibition on emulsifying additives
• Manufacturer documentation for the interceptor and, for hydromechanical units, the required flow control fitting and rated flow
• See related: How to Size and Service a Grease Interceptor Without Guessing; Why a Food Waste Disposer Changes the Drainage Problem; What Hydrogen Sulfide Does to a Sewer Above the Waterline