I wrote an exceedingly long write-up at my blog and at some lists:
about some testing I've been doing - testing the 3/27 test against
other tests a bit, primarily trying to characterize the cold-temperature
appearance of the 3/27 test and trying to see what various cold-weather
behavior means. I'm trying to see if you'll get a really big range of results at specific temperatures- ie if the temperature of the pass/fail point of a 'borderline' 3/27 varies wildly depending on the type of oil or not.
I"m hoping to have some lab access in the next few
months to work the 3/27 against a GC or other test of conversion (there
are several tests of conversion, including the SafTest, GC, the periodic
acid titration, and the pHlip Test).
In the course of this testing I also stumbled across some bad fuel-
which I figured out after I'd put it into my tank. Bummer. It turned out
to be slightly off-spec (.30 total glycerides is more than just
"slightly" offspec in my opinion, but it's too terribly awful- lots of
homebrew seems to come in at .50something-ish from talking to a lab tech
who's seen a few samples, and it can get worse from there...).
ANyway, the goo in my fuel tank- and samples of it really look like goo
if chilled even a little bit- tests OK on the 3/27 at 72F but fails
dramatically at 66F. unfortunately, I put it into my tank before testing
it- and the person who made it had tested it, without chilling the 3/27
materials to 68F or lower, so he'd gotten a false pass, the situation
we're afraid of with accuracy of testing (ie the lesson is, chill your
3/27 materials before testing, to ensure a more accurate result).
>From this experience, and from other times when I've seen bad commercial
fuel that was offspec, and had seen GC test results for it, I think that
"slightly" offspec can be enough to occasionally cause filter-clogging,
cold starts, viscosity problems, and exceedingly bad cold flow behavior
(the first batch of bad commercial fuel I'd ever seen was gelling up and
clogging filters very badly one California August, by the way, and that
was a batch in the .30something range on total glycerol if I recall
I'll be posting more results of this testing as I do it- I want to test
many of the parameters of the 3/27 against samples of known conversion
level, and test things like temperature, contamination with water, and
different fatty acid profile feedstocks.
First post is at: