Using 10/90 tubes accurately.

[dgs]

How strange that I'm posting this with 2 other threads ongoing about dropout tubes. It is actually coincidence.

PD1 (Paul) on here, sent me a sample of his bio to test a few days ago and his dropout result prompted me to make this post.

So, down to the sample. Water result was 400ppm- better than most and well in spec. Soap 44ppm again well  within spec of 65ppm.

Dropout 4.5mls very high, even for a car that will cope with a high % of unconverted. After checking his result Paul got a result of 2.5mls. So why the big difference.

Paul had used a pipette to add his sample to the dropout tube and noticed that the level read about 27mls. Now to explain, this was one of Neisels tubes which read double the level on the low end. This isn't a problem as the reading is simply halved. The 10mls dispensed by the pipette should really read 20 on the tube, not 27.

Paul, thinking the pipette was incorrect poured some sample out of the tube until the level read 20, then added his methanol. This then gave the  (incorrect) lower reading of  2.5mls dropout, explaining the discrepancy.

I calibrated several types of dropout tubes a few years ago and at the lower levels ( 0.1 to 2.0mls) the Neisel type tubes were more accurate than all the others I tested. However they arn't so accurate at the 10ml level (probably like most other tubes) so best not to use the tube calibrations for adding the sample.

I always use a pipette to add the sample, regardless of what the reading is on the tube. Even a cheap class 2 pipette is accurate to +/- 0.1mls far more accurate than a syringe. I realised a few years ago that the 10ml syringe I used actually dispensed closer to 11mls.

[PD1]

I can’t thank Dave enough for his help , I sent a sample as I was interested to see if my drying process was good enough , but learnt so much more .
Paul

[WesleyB]

The volume of two liquids that do not ionize in each other should be the volume of the two parts added together .  For example one milliliter of sodium chloride added to one milliliter of water does not produce two milliliters (volume) or material , it's less . But one milliliter of biodiesel  plus ten milliliters of methanol should equal about 11 milliliters of liquid .  I ignored water as being a factor .  So I had a glass company make me a Warnquest measuring tube . The  lower tube is 15 milliliters graduated in 0.1-0.2 milliliter increments . The upper part of the tube is an inverted pear shape with 2 ml graduations up to 30 ml , then 50 ml , then 100 ml , then 250 ml .  It looks like a separatory funnel .  It was made by NDS Technologies in the USA .  If I add 10 milliliters of biodiesel then add enough anhydrous methanol to make a total of 100 milliliters then my 10/90 test ought to be accurate enough . 

[dgs]

Interesting Wesley. I think I would be tempted to use the full capacity of the tube and do a 25/225 ratio test.

May I be so bold as to ask you the price you paid for this 1 off tube.

[WesleyB]

The tube I described is glass , a custom piece .  I paid over two hundred USA dollars for it . I don't recall exactly . It has a  teflon screw down cap on the top .  I figured it was a one time expense , and I might get credit for designing it . If the Chinese manufactured thousands of them , they would be cheap .  There's not an overwhelming demand for this type of thing though . I would  post a photograph of it if I knew how , but I don't . 

[dgs]

WOW, Wesley over $200 - you must be serious about your testing, well done. If I had spent as much as that on my dropout tube I would be sleeping with it under my pillow!

Do you realise that Jan (Warnqvist) is a member on here. He is regarded as biodiesel Royalty amongst some of us. If you pm him a message about your dropout tube I'm sure he will be impressed.

[DavidA]

How do I use the 'no-titration' calculator from the wiki ?
I don't seem to be able to get it to take any input.
Dave.

[dgs]

Don't bother with it, I never have, just work it out  from scratch each time.

Dropout x batch volume= litres unreacted oil. Then use 15% of this as your methanol amount and 5gms/litre if using NaOH or 8gms/litre if using KOH.

[DavidA]

So if I had a 5 litre batch, Used 800 mL meth ( 80% of the 1Litre I would use at 200 mL/Litre,  25 Gram of NaOH (5 g/Litre) and got a drop out of 1 mL on my 10/90 test, how would this plug into your equation ?
In other words, what would my 2nd stage look like ?

(Assuming I removed the glyc from the 1st stage, of course)

[dgs]

So if I had a 5 litre batch, Used 800 mL meth ( 80% of the 1Litre I would use at 200 mL/Litre,  25 Gram of NaOH (5 g/Litre) and got a drop out of 1 mL on my 10/90 test, how would this plug into your equation ?
In other words, what would my 2nd stage look like ?

(Assuming I removed the glyc from the 1st stage, of course)

So your 1ml dropout is 10% of the original oil, so 500mls.

Methanol at 15% is 75mls and NaOH at 5gls/litre is 2.5gms.

This would normally be more than sufficient to get a clear pass after S2. Remember that S1 versus S2 chemicals will always be at a different ratio, It is assumed that the glycerol produced during S1 will leave the mix relatively dry, so base amounts only are reqd for S2 conversion. However S1 is a different animal. ffa neutralisation uses catalyst and liberates water, oil usually isn't down to the 500ppm water, so more than base chemicals are reqd.

[DavidA]

Ah, I see.

The 1mL being 10% of the 10 mL of oil used in the 10/90.
10% of 5 Litre being 500mL.

That helps a lot, thanks.

[dgs]

Ah, I see.

The 1mL being 10% of the 10 mL of oil used in the 10/90.
10% of 5 Litre being 500mL.

That helps a lot, thanks.

You've got it.