We get this question a lot - I think we need to add this to our FAQ list.
First - remember that what folks call "waste water" really would be better thought of as "flush water" in that this water serves the important purpose of internally flushing the surface of the semipermeable membrane to keep the membrane from fouling/scaling.
When you configure a system with two membranes in series (the waste from the first membrane going to the "in" port on the second membrane), for this discussion let's say it's two 75 gpd membranes, the system behaves like you have a single long (75 gpd x 2) 150 gpd membrane.
Now - if you use a proper flow restrictor, that is, one for a 150 gpd membrane, you'll have about a 4:1 waste to product ratio. Sounds familiar, right?
If however you don't change the flow restrictor - meaning you keep using the same restrictor you were using when you just had one 75 gpd membrane, then you'll see a waste to product ratio much lower than 4:1. But remember that the recommendation for a ~4:1 ratio comes from the membrane manufacturer. They are telling you that you need about a 4:1 ratio to keep the membrane flushed and keep the membrane from fouling or building up scale. Run the system with a lower ratio and you will foul/scale the membrane(s) quicker than would have otherwise been the case.
Instead of adding a second membrane to lower that ratio, you could have just changed out your flow restrictor ($4) instead. A much less expensive approach to get you to the same endpoint in terms of saving on waste water.
Now, to confuse things just a bit. Filmtec specs call for the 4 to 1 ratio on the basis of assumptions about the water that will be supplied to the membrane. If you have very soft water you MAY be able to get a decent service life from the membrane running at a ratio lower than 4 to 1 (e.g., 3 to 1). Remember that the waste water from the first membrane is about 25% harder than your tap water.
Bottom line: If what you are after is reduced waste water, experiment with a different flow restrictor for $4 instead of messing around with a second membrane plumbed in series.
As a side note, you can also lower the ratio by increasing the pressure delivered to the membrane (with a booster pump), because flow restrictors are sized assuming you are providing factory spec conditions (50 psi and 77 degrees for Filmtec membranes). Increase the pressure and you'll drive more water through the membrane and viola - less waste water. But as I mentioned above, if you do this (just like over-restricting a membrane) - the lower the waste to product ratio, the shorter the lifespan on the membrane.
Makes sense?
Russ