Glycol Dilution

Every dilution equation is a conservation statement wearing a disguise. The glycol you pour into a loop does not disappear and does not multiply — only the water around it changes. So the quantity of glycol before mixing, $C_1 V_1$, has to equal the quantity after, $C_2 V_2$, and the equation is nothing more than that sentence written with symbols. Because concentration appears on both sides, its units cancel: percent, litres per litre, or parts per hundred all work, provided you use the same one twice. In glycol work the concentration is almost always volume percent, and it is worth confirming that on the drum label, because a few products are sold by mass.

Take a hydronic loop I need to protect to −18 °C. The system holds 2000 L, the target is 35% propylene glycol, and what is on the truck is 95% inhibited concentrate. Solving for the concentrate volume, $V_1 = C_2 V_2 / C_1 = (35 \times 2000)/95 = 737\ \text{L}$ — a little over three and a half 205 L drums. The remaining 1263 L is water. That second number is the one that matters on site, and it is not what the equation hands you directly; you have to subtract.

This is the same relation as the chemist's $C_1V_1 = C_2V_2$ for molarity, and the same idea again as a mass balance on a blending tank. What changes between those settings is only the unit that concentration is measured in — moles per litre in a titration, volume percent in a glycol fill, milligrams per litre when I am dosing a corrosion inhibitor into the same loop a week later. The arithmetic never changes because the underlying claim never changes: the solute is conserved and the solvent is the variable.

The error I see most often is treating $V_2$ as the water added rather than the finished volume. Pouring 737 L of concentrate into a loop and then adding 2000 L of water gives 2737 L at 26% — well short of the target, and in a system that will not hold it anyway. $V_2$ is the total in the loop when you are done. Water added is $V_2 - V_1$, always. The companion mistake is forgetting that the loop is not empty when you start. A system already full of water has no room for 737 L of anything; you drain that volume first, and if what you drain is already glycol at some unknown strength, this two-term equation is the wrong tool entirely — you need a balance across both sources, and the honest move is usually to dump and refill rather than to guess.

Three more traps, in the order they cost money. Freeze protection is not linear in concentration, and freeze point and burst point are two different numbers: propylene glycol at 30% freezes around −13 °C, at 50% around −34 °C, so the second twenty points buy far more than the first thirty. Read the manufacturer's table for the product you actually have and do not interpolate from memory. Second, verify the finished mix — and verify it with an instrument that matches the fluid. A refractometer scaled for ethylene glycol read against a propylene glycol charge will report a strength that is wrong by several points in the unsafe direction, and a hydrometer reading specific gravity is only meaningful if you correct it for the sample temperature, since the density of the mix moves noticeably between a cold garage and a hot mechanical room. Third, the calculated system volume is nearly always optimistic. Pipe volume from a takeoff, plus coils, plus the boiler, plus the expansion tank, and I have yet to meet a building where that total matched what the loop actually swallowed. Meter what you pump in, then test the result and trim, rather than trusting the number you started with.

One last thing the equation cannot tell you. Percent glycol is a measure of freeze protection and nothing else. The inhibitor package depletes on its own schedule, and a loop can sit at a perfect 35% while its pH has fallen and its reserve alkalinity is gone — which is the state in which glycol turns acidic and begins eating the system it was installed to protect. Concentration is one line on the test report, not the report.