aga
Forum Drunkard
Posts: 7030
Registered: 25-3-2014
Member Is Offline
|
|
PID Distillation
So, i was rigging up my chinese PID controller today because a distillation of methanol from something that says it is methanol
wasn't going well, at all.
The stripping run removed the Blue colour and some water.
After drying the distillate some over 3A molecular sieves, it was distilled again, but kept on reaching up to 75 C which is wrong for methanol (should
be about 65 C).
So, i hacked a hole in some wood to hold the PID thing and wired the hotplate into it.
Unfortunately the PID thermocouple cable was not long enough to reach the stillhead, so i lobbed it into the boiling pot instead.
Set at 84 C the Head temperature remained at 65 to 66 C until the boiling pot went totally quiet !
Never used a PID for distilling before.
So, a PID controller is definitely a good facility to have.
 
|
|
|
Chemetix
Hazard to Others
Posts: 376
Registered: 23-9-2016
Location: Oztrayleeyah
Member Is Offline
Mood: Wavering between lucidity and madness
|
|
So, I read how you found distillation with a PID to be of use. And definitely agree with you. I made a box with an industrial PID and
an SSR wired to a standard outlet plug. In the same box I put a fan motor dimmer wired to another standard outlet plug. Whodda thought that controlled
heating and stirring would be useful?
|
|
|
aga
Forum Drunkard
Posts: 7030
Registered: 25-3-2014
Member Is Offline
|
|
So, Seek and you shall find : PID
Solution to many problems.
|
|
|
Ashot
Hazard to Self
Posts: 79
Registered: 9-7-2016
Member Is Offline
Mood: No Mood
|
|
Hi aga, to be honest, I had to google what is PID, and after about 20 min of reading, I only have the somewhat approximate idea what it is. It looks
like a slightly over complicated device to use for distillation. In my digital mantle I have a digital thermal controller where you can set the
desired temperature, obviously it is not the same but works really well for the purpose.
It is interesting why the temperature was reaching 75C, could this be a ternary azeotrope?
In any case, looks like you have a very professionally assembled fume hood and it can be used for many interesting experiments. I wonder why don't you
make videos of your experiments, I would totally enjoy watching your lab work.
[Edited on 29-10-2016 by Ashot]
|
|
|
CharlieA
National Hazard
Posts: 646
Registered: 11-8-2015
Location: Missouri, USA
Member Is Offline
Mood: No Mood
|
|
PID's are incredible. I mounted a Type K thermocouple in a toaster oven, and control it with a PID (courtesy of my electrical engineer
brother-in-law). Without disconnecting the oven's internal thermostat, I can reach temperatures of over 200*C. This is a great, cheap drying oven!
|
|
|
aga
Forum Drunkard
Posts: 7030
Registered: 25-3-2014
Member Is Offline
|
|
Quote: Originally posted by Ashot  | | ... had to google what is PID, and after about 20 min of reading, I only have the somewhat approximate idea what it is. |
When applied to a Heater, a PID controller is basically a more accurate thermostat.
https://en.wikipedia.org/wiki/PID_controller
http://www.ebay.co.uk/itm/100-240V-Digital-PID-Temperature-C...
|
|
|
Brom
Hazard to Self
Posts: 94
Registered: 19-7-2015
Member Is Offline
Mood: No Mood
|
|
SO if you don't like the way some people post aga , (which i would have to agree with you) replicating that behaviour is not the way to stop it.
|
|
|
Sulaiman
International Hazard
Posts: 3723
Registered: 8-2-2015
Location: 3rd rock from the sun
Member Is Offline
|
|
aga
"Unfortunately the PID thermocouple cable was not long enough to reach the stillhead, so i lobbed it into the boiling pot instead."
should have read "Fortunately ...."
think about it.
CAUTION : Hobby Chemist, not Professional or even Amateur
|
|
|
Ashot
Hazard to Self
Posts: 79
Registered: 9-7-2016
Member Is Offline
Mood: No Mood
|
|
I read this wiki article, also I found this video on youtube, the guy explains it really well, https://www.youtube.com/watch?v=UR0hOmjaHp0.
I will definitely research it further and might find a use for it in my lab.
Thanks.
|
|
|
Chemetix
Hazard to Others
Posts: 376
Registered: 23-9-2016
Location: Oztrayleeyah
Member Is Offline
Mood: Wavering between lucidity and madness
|
|
I dont think the virtues of a PID have been fully explained. There are so many manufacturers and models out there, but in general these controllers do
more than just on and off at a set temperature. Mine is a BTC9300
http://www.brainchild.com.tw/en/2-1752-27960/product/50-x-50...
And just for example, these are some of the functions that I use regularly:
You can set an upper temperature and hold that temperature for a given period of time. You can set both an upper and lower temperature, if the lower
temperature triggers then the output can control heating and if the upper temperature triggers it controls cooling, a fan or water pump or whatever.
It can heat or cool at a given rate, say 200 degrees an hour, until the set point is reached, hold that temperature within a set range, then cool down
at a set rate. The ramp up and ramp down can be different.
What makes them different from just a thermostat is the way many of them have internal programs that can "learn" the right amount of heating to give
the right rate of heating. So there is very little overshoot and undershoot of the set temperature. You can program the unit to have a certain
hysteresis. Say you want a certain temperature of 78C in a water bath but you are distilling a substance at 54 at the still head.
You set the temperature to 51C with a 3 degree hysteresis to allow for the slower rate of heat going into the waterbath.
These are just the basics, they are serious microprocessors packed in a small wall mounted box.
One word of caution: they usually have only three buttons to program all of these functions, the menu structure to set up all these functions is often
hard to interpret. A little LED panel display uses all sorts of cryptic characters to abbreviate the function or selection in the menu. It can be hair
pulling to work out. And the manual is often wrong or has typos or relates to an older model where the model you have now has a slightly different
character set.... again, more hair pulling.
|
|
|
aga
Forum Drunkard
Posts: 7030
Registered: 25-3-2014
Member Is Offline
|
|
This one has 3 buttons and came with no manual.
Looks pretty much the same as your BTC9300 apart from the buttons.
As per usual, just pressee-holdee-long-time button one and up came the P and I variables, plus the alarm threshold.
I don't really understand what those settings do outside the context of a quadcopter, so left them as per default, and it worked fine for this
particular distillation.
|
|
|
Chemetix
Hazard to Others
Posts: 376
Registered: 23-9-2016
Location: Oztrayleeyah
Member Is Offline
Mood: Wavering between lucidity and madness
|
|
After playing with a few brands and models, you sort of get a feel for what should be the right button combination to access which menu. But an online
manual is invaluable when you need that extra function. The BTC 9300 is probably a bit too complicated for the home hobbyist/madscientist, there are
other simpler versions that are pretty much point and shoot sort of things. Set the Temp, thermocouple type, select Proportional or Integrated
control, then play with the setting values to work out what works for the situation. Just like you have done.
When you are controlling a decent amount of current I recommend going for a solid state relay, they are very reliable in my experience. Very clean
shut off and gentle to the equipment being controlled.
|
|
|
aga
Forum Drunkard
Posts: 7030
Registered: 25-3-2014
Member Is Offline
|
|
If it were 100A+ i'd go for a big solonoid style relay any day.
Might be why you'll still not find an SSR in your car for the starter motor.
SSRs fail badly under fault conditions.
|
|
|
Chemetix
Hazard to Others
Posts: 376
Registered: 23-9-2016
Location: Oztrayleeyah
Member Is Offline
Mood: Wavering between lucidity and madness
|
|
Decent current vs. Huge current, yeah, go the solenoid relay.
|
|
|
Sulaiman
International Hazard
Posts: 3723
Registered: 8-2-2015
Location: 3rd rock from the sun
Member Is Offline
|
|
110 A thyristor £25.47 + http://uk.rs-online.com/web/p/thyristors/7005554/
but they do not turn off easily when using d.c. 
CAUTION : Hobby Chemist, not Professional or even Amateur
|
|
|
Fulmen
International Hazard
Posts: 1725
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
I have and I still disagree. The boiling point in the pot should depend on the composition and pressure, nothing else.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
aga
Forum Drunkard
Posts: 7030
Registered: 25-3-2014
Member Is Offline
|
|
I thought about that for a second and still thought that measuring the vapour temperature at the top of the vigreux would still be the best place to
put the thermocouple.
As far as i can see, doesn't really matter as the head temperature and boiling pot temperature are intimitely related.
|
|
|
Fulmen
International Hazard
Posts: 1725
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Head temperature also depends on the power input, pot temperature does not.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
aga
Forum Drunkard
Posts: 7030
Registered: 25-3-2014
Member Is Offline
|
|
Huh ?
Depends entirely on the composition of the stuff in the pot dunnit ?
Edit:
For reference, in these particular circumstances it was a hotplate directly heating a 250ml RBF with a 300mm vigreaux column on top, standard
stillhead and 200mm liebig condeser.
Pot contents were one-time (simple) distilled 'Metanol' which clearly were not 100% methanol.
Blue for starters, and contained a lot of water.
The stuff left in the pot after the first simple distillation was left outside for any volatiles to bugger off. Insignificant volume decrease after
two days (not so hot on the Costa del Sol at the moment) so i guess it's water.
[Edited on 30-10-2016 by aga]
|
|
|
Sulaiman
International Hazard
Posts: 3723
Registered: 8-2-2015
Location: 3rd rock from the sun
Member Is Offline
|
|
I thought that distillate vapour temperature should be constant
= the least useful control signal
the pot temperature should rise as the lower bp fraction is reduced,
but at any particular concentration the b.p. is constant
= not a good control signal
I think that what is required is a constant boiling rate
which is best achieved by controlling the heater temperature
or by using simple variable power regulation.
[Edited on 31-10-2016 by Sulaiman]
CAUTION : Hobby Chemist, not Professional or even Amateur
|
|
|
Fulmen
International Hazard
Posts: 1725
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
Lets start from scratch. The boiling point of any liquid depends only on it's composition (and the pressure). So pot temperature will change
throughout the process as the more volatile compounds are boiled off. At the head the temperature will also depend on composition, and since the goal
is a pure product it should be kept constant (distillate BP).
The actual temperature (and thus composition) at the head also depends on the power input due to changes in reflux ratio. As the power increases more
vapor enters the column while the cooling power is fairly constant. This means that less of the heavier fraction condenses and returns to the pot.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
Sulaiman
International Hazard
Posts: 3723
Registered: 8-2-2015
Location: 3rd rock from the sun
Member Is Offline
|
|
so for a feedback controlled pot heater,
where would you put the temperature sensor for best product fractionation ?,
and why?
CAUTION : Hobby Chemist, not Professional or even Amateur
|
|
|
Chemetix
Hazard to Others
Posts: 376
Registered: 23-9-2016
Location: Oztrayleeyah
Member Is Offline
Mood: Wavering between lucidity and madness
|
|
I'm going to throw my two cents worth in and say the still head; getting the parameters right with the controller is tricky because there is less
coupling with the heat source, but the end result will be better. If you set the still head to not go over a certain temp, even if the heating has to
oscillate a bit to achieve it, the end result is that nothing over a certain temperature will come over.
The arguments above that the temp in the pot will rise as the distillation procedes is somewhat correct, the main factor is the volume in the flask
diminishes and so there is less surface area in contact with flask walls that are directly being heated with the water bath. And not to mention the
surface area from which to evaporate reduces too. Heating requirements for the bath are generally non-linear and require more heating for the smaller
volume left, if you want a reasonably constant rate of distillation that is.
If you have a column involved then the rate of evaporation from the liquid in the flask can drop below the rate of heat being lost from the column and
the distillation stops at some point and just refluxes. More heat is going to be needed to keep the distillation progressing and at an amount that
doesn't compromise separation.
|
|
|
Fulmen
International Hazard
Posts: 1725
Registered: 24-9-2005
Member Is Offline
Mood: Bored
|
|
At the head. Your end goal is a product with a given composition, this means a product with a given boiling point. In this case you want pure MeOH (BP
65°C) from a water/MeOH-mixture. If the head temperature is higher than 65°C you're not getting pure MeOH, if it's lower then nothing is passing
over.
We're not banging rocks together here. We know how to put a man back together.
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
