Hi Keith;
Most often the process comes to 95% or more complete and that is good
enough for me ( personal choice) and doesn't give me wash problems
either. Maybe 20% of the time it only gets to somewhere around 90% and
this is a problem. It tends to be most likely to happen when the
feedstock titrates with a result of higher than 5 (this is titrating
with KOH soln rather than NaOH) which is what I consider borderline for
going acid-base. Settling time is not less than 12 hours for me.
Sorry about snipping the remainder of your post, was trying to save
bandwidth.
Off for the weekend now will check messages on Monday.
Hope you are having a good weekend. And also to those about to start the
weekend.
Tirah
Joe
Keith Addison wrote:
Hi Joe
Hi Keith;
See my answers below.....
Keith Addison wrote:
Well, settling time is free.
Acid-base aside, there's the two-stage base-base process, which quite
a lot of people use and like, but otherwise why do more than one
stage? Do you mean two separate stages, with a methanol test in
between? So you process it twice? Plus extra methanol.
No,- more like aiming for a single stage and then using the methanol
test right after the settling period (after draining the glycerol of
course) it gives you a chance to hit the reactor with another small
dose of methoxide if it wasn't quite a complete reaction. The test
also tells you how much to use. There is no standardising the
process here since the feed is never the same twice.
I don't think anybody's feedstock is ever the same twice, unless
they're getting it from a food factory with a standardised operation.
No restaurant cooks exactly the same food in the same way two days
running.
Anyway, why should that mean changing anything but the amount of
catalyst needed? That's what titration is for, no? Certainly you can
standardise the other variables.
Rod decided to try a methanol test before washing, just for giggles
and it turned out to be a heluva good idea. You know right away if
you got a good reaction without having to waste all the time energy
and water washing and drying before doing a quality test and then
potenially having to reprocess.
Before you started using this extra stage, how often did you have to
reprocess? In your previous message it sounded like it happened
regularly, and it still sounds that way.
It turns out this looks just like a two step base -base type deal.
but more like 90% of it in one step and then a polishing step.
Sometimes if your process was good you don't need to do it, but if
it turns out you were not near enough to completion it's nice to
know right away. There's still methanol in the esters at that point
(before washing) so it's advantageous to push the process further at
that point if it is necessary.
Joe
This seems to raise more questions than provide answers. There were
also some other points in my reply to Tom, eg, how long do you let
the glycerin settle before draining it and embarking on the second
stage/polishing step or whatever?
I'll put the rest of my message back, below.
Please see my reply to Tom:
http://tinyurl.com/3ccqhw
[Biofuel] Glycerine Settling Time
Best
Keith
Hi Joe
Tom;
It makes sense. Glycerin is an emulsifier. Have you ever tried
dosing the batch again with a little methoxide? After you remove
the glycerin it doesn't take much to get the last bit of the
reaction to go and settle out the remaining glycerin. Of course
this is well known already. Kenji and many others do straight base
catalysis as a two stage deal. You can do a methanol test of sorts
and the unreacted oil will settle out. Then you can use the
measured amount of unreacted oil in the methanol test vial to
estimate the percentage unreacted oil in your batch and dose
accordingly with the stoichiometric amount of methoxide. Assume
neutral oil for this calculation. Rod and I do this regularly if
the batch fails the QT and it works like a charm. Will save you
settling time in the long run.
Well, settling time is free.
Acid-base aside, there's the two-stage base-base process, which quite
a lot of people use and like, but otherwise why do more than one
stage? Do you mean two separate stages, with a methanol test in
between? So you process it twice? Plus extra methanol.
Why not do it in a single phase? Todd Swearingen once suggested this
here (discussing mixing pump sizes):
To judge an appropriate reaction time, pull an exact amount of fluid
(200 ml would suffice) out of the reaction stream every half-hour or
hour after an arbitrary initial ~1 hour reaction period.
Presuming that the contents of the reactor are kept homogenous from
the pump flow, the volume of the glycerol cocktail that settles out
of each sample will give you a fair gauge as to when your reaction
completed.
The suggestion would be to continue the reaction for ~1/2 hour
beyond the point where your glyc cocktail volume stabilized.
That works. Then, surely, you can standardise the process, with the
only variable the amount of lye according to the titration level.
Then if you do one-litre test batches first, especially with iffy
batches of oil, and you have a clear idea of how your test-batch
processing relates to your full-scale processing, life should be
easier and there shouldn't be any batches failing the QT.
What did I miss?
_______________________________________________
Biofuel mailing list
Biofuel@sustainablelists.org
http://sustainablelists.org/mailman/listinfo/biofuel_sustainablelists.org
Biofuel at Journey to Forever:
http://journeytoforever.org/biofuel.html
Search the combined Biofuel and Biofuels-biz list archives (50,000 messages):
http://www.mail-archive.com/biofuel@sustainablelists.org/
_______________________________________________
Biofuel mailing list
Biofuel@sustainablelists.org
http://sustainablelists.org/mailman/listinfo/biofuel_sustainablelists.org
Biofuel at Journey to Forever:
http://journeytoforever.org/biofuel.html
Search the combined Biofuel and Biofuels-biz list archives (50,000 messages):
http://www.mail-archive.com/biofuel@sustainablelists.org/
