Seal plan for Vacuum Residue (Asphalt) pump

I would definately not recommend API plan 23 for this service. Besides affecting the quality of the asphalt, this is a very expensive use of gas oil. We desulfurize the gas oil and send it to our FCC to crack it into gasoline. All of our vac bottoms pumps use a pressurized double seal. We use the same gas oil supply, but instead of injecting it into the product, it is directed between two seals. The gas oil passes between the two seal and is taken out of another connection on the seal. The flow is regulated with an orifice on the outlet connection. The stream of gas oil is dumped back into a line that goes back to the vacuum tower, so this product is fully recovered. If you can maintain the pressure between the seals at least 30 psi above normal seal chamber pressure, both seals run on the gas oil. The seal manufacturer will tell you how much flow is needed for the particular seal to carry away the heat generated at the faces.

There are two reasons the Plan 32 is used here: to lower the viscosity of the product in the seal chamber and to lower the risk of product flashing in the seal chamber. There should be a throat bushing in the seal chamber to help decrease the usage of the gas oil by keeping the seal chamber pressure higher than the volute. There are devices that can be incorporated into the flush piping to monitor and control the pressure and the flow of the gas oil. But from the sound of things, you’re interested in getting rid of the gas oil altogether.

Unfortunately the objectives of keeping the product in liquid state and decreasing the viscosity are at odds. Cool the bottoms to a temperature far enough below the flash point and you increase it’s viscosity.

JJPellin’s suggestion of a barrier fluid is a good one. A double metal bellows seal with a barrier fluid at 20 to 30 psig above the seal chamber pressure (API Plan 53) is often used here. Since we’re talking about 650 F or so, you might choose a synthetic lube oil for your barrier fluid as they have boiling points at 700 F or so and have good heat transfer capabilities. You might also look into a heat transfer fluid such as Dowtherm or Therminol.

I’m sure that you warm the pump before start-up, but without the Plan 32 you’ll want to be sure you do a complete job of it. Otherwise you could end up with product still set-up in the seal chamber that will make start-up a serious hazard for the seal.

Another option would be to use a metal bellows seal with hard faces along with a Plan 11, recirculation from discharge, and a steam quench, Plan 62. The Plan 11 will create flow to cool the seal faces and the steam quench will also help cool the seal faces and control coking on the atmospheric side of the seal. If you want the added security of a double seal because of concerns for the product temperature or auto-ignition (you probly aren’t, as it appears that you are using a single seal now) a dry running metal bellows seal can be used to act like a throttle bushing for the quench and as a secondary containment seal if the inboard seal begins to leak.

Remember that with either of these seals you should check for the throat bushing in the seal chamber. If one is installed it should be removed from the pump. The restriction used to keep the seal chamber pressure so high will work against the flow required for the Plan 11 and can make it more likely for a vapor space to occur in the seal chamber as the pump is inventoried.

Give these ideas some thought and then contact your seal vendor and talk about the process and the issues that operations (yeah- you’re with operations- admit it) and maintenance have with the current seal. Talk about the process stream and corrosion too. You want to be sure that the seal hardware is of the proper metallurgy. And lastly talk about the implications that any of these flush plans has for operations. Any of the flush plans talked about here, 11, 32, 53, and 62 may have valves, pressure gauges, or in the case of the 53, sight glasses to monitor. Talk about the relative complexity of the operator’s duties with these flush plans as the responsiveness or playfulness of operators can have an effect on seal life.