Biopowered - vegetable oil and biodiesel forum
Biodiesel => Chemistry and process => Topic started by: Glycer-rides on May 08, 2016, 10:57:15 PM
-
At least it was a nice day to spend a la shed...doing 4 stages. :)
This oil, from a new source, looked good. Clear reddy-orange with a spicy nose to it.
Settled for a month, bottom 10% drained off, glyc. washed. That should be good to go, easy peasy...
I will spare the details, but:
S1 27% uc.
S2 18% uc.
First ever S3.... 15% uc.
S4 :o......14% uc!
Low glyc yield all round. On the latter stages it wasn''t drained before the next stage.
In the dim and distant past I recall reading of curry oil probs? Any ideas?
-
Perhaps it's a mix of veg and mineral oil?
Whenever you think you've got it all in hand, there's a batch that likes to show you you're not as much in control as you thought ;)
-
Thanks Tony, that thought crossed my mind.
Too true!
I suppose I might ask N***, in a nice way, where he gets it and why he didn't filter it for his wvo use.
-
I have had similar batches in the past, probably due more to wet oil, but not as bad as that.
it is never a good idea leaving the glyc in between stages, it 'holds on' to some of the next stage chemicals.
You say low glyc yield, but if you added so much excess chemicals surely you would end up with more glycerol than normal.
-
I'm sure it wasn't wvo wet, Dave.
The glyc volume decreased 35% after the glyc wash. Something's transfered across, hopefully not all soap...
With wet wvo, the glyc wash turns 'brown and bubbly'.
-
OK, I understand now.
I wasn't suggesting it was water that is your problem in your case. I suppose it could be mineral oil, but how did it get in there.
I wonder if anyone has ever done a dropout on mineral oil, just to see the phase split and to see if any oil dissolves into the methanol.
I will do it now and report back.
-
Well, used engine oil doesn't split with glyc, so I'd expect mineral oil to perform the same with meth?
So how did 25L of clear (not used) mineral oil end up in my wvo?
I've now asked the guy I got it from.
-
I added 10mls 10-40 mineral oil to 90 mls of methanol. Gave it a good shake. An emulsified layer of about 40mls settled. The methanol phase is cloudy. The emulsified layer is semi-solid, if the tube is held horizontally, the meniscus hardly alters. It has now been 15 mins.
If nothing alters, could we assume that if your bottom phase goes clear then mineral oil is not present?
I have done tests with glyc and WMO, to see if the glyc removes any suspended carbon. As you say there does not appear to be any split. However, due to the SG difference between the two, I think they do split, but the WMO transfers so much carbon to the glyc that it masks any visual split.
-
It has now been 75mins since the test started, the bottom layer has reduced to 25mls and the methanol phase is nearly clear. The emulsified layer looks like dessicated cocanut, only a dirty orange colour.
i have just started another 10/90 which has 1.5mls of 10-40 motor oil and 8.5mls of finished bio, with 90mls of methanol. (I think this more replicates what you may have)
At the moment there is 8mls of the same bottom layer. (dessicated cocanut)
As said if your bottom dropout layer is clear, from the above results I would doubt the reason for your unusual reaction is mineral oil.
-
Thanks Dave, useful tests there.
The wvo comes from a trustworth source and he has confirmed it was his normal stuff.
It must be me! The drop out is exactly the same shade of orange as the wvo was.
With the excess KOH added having not much effect, I would have expected jelly territory but there was no sign of this.
I gave up last night and stopped the inaction with a 5% wash. I will drain tonight. Probably should have left it 'live'.
-
Certainly a strange one Andy.
I've just checked the tests again, just mineral oil shows a little under 25mls of 'whatever' and the bio/mineral oil shows a little under 8mls. I don't think there will be any more change now.
The dropout test is measuring bio/oil in roughly a 50/50 ratio. We rely on the soluble proportion of the diglycerides and all the monoglycerides to roughly equal the amount of bio contained in the oil dropout. (the reason why the test usually works)
If for some reason these di's and mono's are in a much higher proportion in your partly converted oil it would account for what you are finding, but I have no idea why this should be.
-
I left the 5% wash 24 hrs before draining it. What drained out looked unexceptional.
Heated it back up to around 50C and glyc washed the bio. Again, what came out looked normal, no real loss of yield.
I did a 10/90 on the bio and got a similar result, 15% u/c, rising to 17% u/c the next day.
Tonight I did a 500 mL Dr. Pepper on the bio to see if I could get a full conversion. (measuring 3g KOH was hard, no spare batteries left for my 'drug dealer' mini scales...)
It processed OK, no soap or mono/di layer BUT the 10/90 is showing...................15% u/c, which I expect will rise to 18%.
In summary, still stumped!
-
Are you sure the meth you are 3/27 testing against is pure?
I've had problems with dropout before, and that turned out to be because I had left the top off the jar of meth I used for this testing, and it must have absorbed some moisture from the air. It seems only a tiny amount is needed to mess up the test.
Alternatively, have you got some "known good" bio to double check your 3/27 test with?
There was a chap I met locally that called me over to look at an odd batch. 3/27 gave dropout that formed into balls. Couldn't get any further conversion with a Dr Pepper, and his meth was pure enough. Never did figure that one out.
-
It's pretty new decanted Trinity reclaimed, so I think it's OK. I'll try another cubie.
That's a good idea, Tony. I'll try testing some known good bio.
My drop out also has a ball-like, 'spawny' look to it initially, which later evens out to look normal.
I'd can this batch...if I had any cans!
-
I was worried about this exact same thing when I started using Trinity reclaimed. I did some comparative tests with virgin stuff and the dropout was exactly the same.
-
If the meth turns out to be OK, how about drying it thoroughly (maybe heat to 120C to make absolutely sure there is no trapped water) before reprocessing?
I'd try burning a bit of this dried batch in a jam-jar lid to see if any residue is left (beyond soot).
How does a 50:50 water shake look? Lots of of emulsion? (I'm not sure what that would tell us beyond known partial conversion - but, you never know, it might come out clear in which case more thinking is required!)
-
B0llocks, I've just spent 20 mins updating the thread without being logged in.
Clear pass obtained!
I will type it out again when the will to live has been re-established...
-
Stop teasing, how did you fix it?
-
I had the Dr P methoxide left, so I re-dosed it very heavily, just to establish conversion was possible. It was!
Based on the 18% u/c in the batch, I mixed up 20% meth @ 6 g/l, bearing in mind my unwisely depleted start point.
No dice after 90 mins.
Being late, I mixed up a 'pint plus a handful', ran for 60 mins, drained a sample and sodded off to bed in a mood.
Tonight, the sample looked a good split, drained the glyc and got a normal quantity and a 10/90 clear pass.
I will post up my 'learning points' later...
-
I mixed up a 'pint plus a handful'
I'm glad it's not just me that does this.
"Just a dash of catalyst, and a splash of methanol to taste"
As used by all the finest television chefs.
-
The guy that I collect the chemical rich glycerol from has never weighed his catalyst or measured his methanol. When I asked him how he measures his KOH he said X number of scoops, however the mood takes him.
He has never done a dropout test. I once (couple of years ago) took him a 3/27 tube and explained how to do the test, he never bothered.
-
sounds a bit like mine, just under a 20L drum of meth (inc 25% recovered) and then a tub and a half of NaOH if the oil is good, or a tub and three quarters if its not so good ;)
-
I'm usually as accurate as I can be with the vols and masses.
It didn't show results in this case, so the artisan flair was an uncharacteristic moment.
Points:
As Dave has mentioned, I'm thinking I've glyc washed with excessive soaps being carried over.
I'm finding KOH glyc takes much longer to settle vs NaOH glyc, which might have stalled the stages. Not enjoying the K experience so far...
5 g/L KOH doesn't seem enough for my process.
Not having a stable source of wvo makes things harder!
-
Update!
Whilst my crystal clear pass 10/90 is still crystal clear some days later, there is some small amount of white drop out which started to appear after 24 hrs.
It comprises tiny, white soapy flakes. Not wvo coloured as would be expected.
I did a 1/9 on someone else's bio (water washed and settled for an age...gives a much nicer 50/50 than mine) - same white drop out.
Obviously, the only common denominator here is the meth used for the tests. This is quite new, Trinity reclaimed meth.
Ideas??
-
White fallout could be waxing, but it seems a little unlikely at these temperatures...
-
Here's an update, as we like updates...
Well, this bio has been fine in my new 250bhp/540Nm brown trouser CDi.
Just started another batch.The processor was pre-loaded with more of this orange wvo, so I had no feedstock choice.
Same lack of reaction crapness transpires.
At this point, all I'll say is that this Trinity reclaimed (that smells of solvent) when mixed with Trinity KOH, turns the colour of bio.
That's not right, IME.
Off to the pumps mid week, then.
-
The Trinity reclaimed methanol is certainly wierd stuff as far as appearance and smell goes. The last IBC I had also had a colour to it. Having said all that, the reactions are completley normal.
When processing, if the last reaction glycerol isn't given enough time to settle and be removed then any chemicals added from the 10/90 findings will result in a lot of them being absorbed by the glycerol.
Using low temperatures for the first reaction/s and a minimum of methanol seems to slow the glycerol dropping out of the process. If I leave each reaction overnight so most of the glycerol can be removed then the chemical addition from the result of the 10/90 is quite accurate. This is one of the reasons that I use no heating at all for the initial reaction/s at this time of year.
-
Thanks Dave.
I started at 60C, minimal method 11% and 6.5 g/l of KOH. Eductor mixed for 2 hours.
With nothing happening, the meth was incrementally increased, ending up at 13.5% with nothing happening.
I then did what I did before. About 250g in a litre or so. An hour of mixing and a good split looked evident before I went to bed.
I hope to find that I've single staged it, when I check the conversion later!
I suppose it's taken 9 g/l thus far. This KOH is not old and has been well stored.
An ode to titration...?
-
Not enough meth handy in my 10/90 test supply bottle, so I had to do a syringe 1/9 for indication only with what was left.
Looks to be c. 70% converted.
-
Bit of an update for y'all.
The subsequent batch of good wvo went fine, no glyc wash was done.
This weekend's batch was fair wvo and was K glyc washed...and ****, we're back to 3+ stage buggeration.
Despite long settling and done at 60C, my K glyc. just doesn't seem to drop out fully, then when processed I get the knock on effects.
I can't wait to get back to Na...
-
Have a go doing what I do;
2/3 Na (this can be solid or ASM or a mix, depends on what type of methanol I'm using )
1/3 K
It has worked really well for me, better than using just one or the other.
If the K is dissolved in the methanol first, so the methoxide gets quite warm, then any solid Na that is added dissolves quite quickly. Any ASM is added last and mixed well.
-
Cheers Dave, I might try that when I get some Na.
That was kind of you to offer to help Beryl's mate many miles away.
I realised I knew Beryl locally from 2000 to 2005 on Classic car related stuff. Last time I saw him in 2005, before he moved County, he'd just fitted a Merc engine to a Sherpa and was running wvo. First time I'd seen it actually done.
So he's to blame...!
-
Haha, it seems as though Jim has it covered anyway.