Yes I understand. I'll post some pictures soon. But you didn't answer my question: does this process completely dewater, or do you have to dewater in some other way?

I think that there is no way to say that ANY centrifuge will get ALL the water out. Its not like a mechanical filter. The completeness of separation is determined by force (Gs) and time (flowrate). That said, I feel comfortable that my oil is sufficiently de-watered. If I had to deal with a poor quality feedstock, I would reduce flow rate and purge the bowl more often.

-There is a post early in this thread that goes over the calculation of the time that the oil is actually in the centrifuge. This guy ran his at a very conservative 2gph. I run at about 10gph but wouldn't go faster than 15gph.

- I am very uncomfortable with the dieselcraft flowrates (55gph OC-20)(108gph OC-50). I can't see that there is anywhere enough time to get any significant separation. And I feel even less comfortable with their answer (run it through three times). Up until recently, these have been the only thing available but they are certainly not designed for the task in question.

Leon

All I can see from your images is what looks like a ball valve at the opening at the bottom of the top cube.

How do you regulate the flow (10GPH) from the top cube into the CF?

There is a restriction in the centrifuge that reduces the flow to 10-20gph. I am working on a method of adjusting this to suit your specific needs.

The restriction is in the machined rotor piece?

No, the ports on the lid are machined as a 3/4"NPT coupling (threads both ways) so on the inside there will be a reducer then a tube fitting. the size of that tube is the restriction. I will supply 1/4" but that could be changed a size larger or smaller depending on your needs.