The key to the problem is that it's not just about what the surface tension of the liquid is - it's also about how the solid surface interacts with that liquid. Metal surfaces are generally what we would call hydrophilic surfaces - surfaces that "like" water. It is also classed as a "high energy" surface. You want the surface to behave like it is "hydrophobic" - "water hating". Most of the time when you want to change the interaction you change the surface not the liquid.
Raising the surface tension of your solution will probably not have the effect you want because the metal surface would still interact well with the water based solution. Oils are the opposites of water. Water is a polar liquid where oils tend to be nonpolar liquids. The metals do not interact strongly with oils and so oils bead on metal surfaces. And the surface tensions of most oils are in the 30-40 dynes/cm range - much lower than water (which is about 72 dynes/cm) or water/glycol (which is in the 50-70 dynes/cm range, depending on the amount of glycol) . Despite having a lower surface tension than water the oils bead better on the metal surface because they chemically interact less with the surface. What counts is the interaction between the surface and the liquid and although surface tension can have a moderating effect the biggest determiner of the contact angle (the curvature of the bead on the surface) is the interaction between the two substances. There's a decent discussion of this in the beginning of the Wikipedia article on "wetting" Wetting - Wikipedia
That being said, the only common mention I'm seeing of a material that can increase surface tension is adding a lot of salt. I've seen citations that say a 6M solution of sodium chloride in water raises the surface tension to about 82 dynes/cm. A 6M (M = "molar" or "moles/liter") solution of salt is about a 35% solution of salt! (This is essentially a saturated solution of salt in water; the glycol would probably decrease the amount of salt you could dissolve so you probably could not achieve this high a concentration.) I doubt that you want all that salt in contact with a metal surface. Salts are some of the few substances that increase the interactions of the water molecules and so strengthen the surface tension. The glycol would be working against that so I doubt even saturating the glycol/water mixture with salt would raise the surface tension appreciably. And adding a salt to the mix would probably only make the metal surface interact even more strongly with the liquid and, despite any increase in the surface tension, would probably still not bead on the metal surface.
The only way to make the liquid bead on the surface would be to make it more "oily" - less polar, less chemically interactive with the surface. To do that you might add something nonpolar (oily) to your solution but it would not mix with the water without adding a surfactant (detergent). And then the addition of the detergent - to form an emulsion of the oil in the water/glycol - would increase the solutions ability to wet the surface (since the detergents tend to increase interaction with surfaces and decrease surface tension). I don't know what glycol you are using but using a less soluble (longer carbon chain) glycol might be more help. Trying to make water less "watery" is a very difficult proposition. Which is why most people change the surface instead of the liquid.
I hope this is clear. I tried to avoid as much jargon as I could. If you're into the math you can look at the bottom of the Wikipedia article for its discussion of the Young equation which is a much more rigorous explanation.
Thanks very much for your thorough and insightful response. It is very useful and helpful. You are certainly correct that adding salt would not be desirable due to our customers very low tolerance for chlorides. The end result that they are looking for is a high flash point, environmentally friendly antifreeze that also provides an easy way to dry the piping once it is emptied. Our customers often use a methanol water solution because the methanol dries easily but the flash point is too low and it is not that environmentally friendly.