My third suggestion for improving the manometer is to use a solvent when testing WVO.
The Sandy Brea test unit calls for mixing a dry solvent with the WVO at a 1 part solvent to 3 parts wvo rate. In recent discussions with Sandy Brea Labratories it appears this is neccesary since the reactive reagent must be able to quickly contact all of the water present in the sample and thinning the sample aids in this process.

The fourth..and possibly final suggestion is that since the reaction may take in excess of 5 minutes to complete and is exothermic the air traped above the sample and some of the gas produced will experience expansion and skew the results slightly. As long as the temperature of the sample rises this will always result in an indication of more water in the sample than actually is present. And for most of our purposes that is OK.

But if accuracy is paramount it may be advisable to add a digital thermometer to the test vessel.

I always have a "bath" of room teperature water handy in which to set my Sandy Brea test unit. Snce it is machined aluminum I am sure that heat is transferred from my hand to the contents (even though I wear an insulative glove). By setting the aluminum test vessel in the bath of room temp water after an initial reading I find the reading a few minutes after is usually slightly lower. This may be a good dea when using CC in place of CH since the reaction may be a bit slower and it is wise to be certain that the reaction has completed before trusting accuracy of the result.

I do not think it is practical to add a digital thermometer to the SBL test unit. However it would be very nice to be able to simply adjust the results for the rise in temp inside the test vessel. And probe type digital thermoeters are relativly inexpensive.

Again...this may not be neccesary for anyone who is not obsessive about accuracy.

EDIT:

OK..four more suggestions which might make it more accurate.

1.I just added an inexpensive valve to the cap of my pressure vessel which allows me to shake the sample without getting any in tube leading from it.

2.I assembled it from PVC fittings which allow me to easily clean and dry it to be sure none of the previous sample remains. It is capable of holding over 60psi without leaking. So If I have multiple samples I can ready each sample shake and allow them to sit to complete the reaction..then hook them up to the manometer for a quick test of each by simply opening the valve.

3.I also just added a way to seal and zero out the pressure in the vessel and THEN add the CC.

4.Instead of making a U of my clear tubing I plumb the valve on the test vessel to the cap of a polycarbonate soda bottle with a bit of that tube. I then half filled the soda bottle with colored water and installed a second length of clear tube through the cap which extends all the way to the bottom of the bottle. I then run that length of tube strait secured to the wall. Behind the tube is a graduated scale which has a line where the water in the bottle must be to be "zeroed out". This avoids having to add colored water to the clear tube and allow smaller tube to be used..which allows greater accuracy due to its' smaller volume per inch.

more later...