Why You Can't Control Temperature

[HDNB]

temperature related..

Mars

temperature related vs temperature controlled.

it's funny the war of words that say the same thing. no matter how complex you make the equation, it's still not temperature controlled.
indeed, there is a relationship to temperature.

you could draw the same correlation to the volume of liquid in the boiler. it's related...but no one "controls" the level of boiler relative to the subjective quantity or quality of the still's output.

you could really hurt your brain by trying to math the controlling of temperature of the boiling wash controlled by adding mash via a continuous process using a fixed energy input. arguably you could then state you are controlling the output by temperature, since the energy input was fixed...but then you are really controlling by level, not temperature...?

[OtisT]

For a Pot still I think it would be cool if you could measure production rate with the PID so it could adjust power accordingly for a proper flow. If it were a fractioning column, measuring vapor temp, reflux rate and production rate could allow you to control power and cooling flow for a run. That would be a cool use for a PID. Of course, you need to know how to tune these things manually before you could ever program your system to do it for you and by that time ya gotta think why bother, unless you are commercial.

Otis

[acfixer69]

Hate the page breaks but hears a go again from me. With a pot still all heat going at it is going into the boil and what that temp is, is what it is. Logs, amps, don't matter it still boils at the same temp. For plated stills that is still a true statement sort . All bubblers are not the same. Caps and perfs are different again so be sure we are all talking on the same thing not what you do.
Happy VET'DAY fellow VETS

[Saltbush Bill]

All that said WE don't control the boiling temp the wash ABV does so WE set the rate of the boil.

Thank you AC , exactly my point to begin with, We do not control the temp of a boiling wash, instead it controls us. We must wait and let it do its own thing. When it is ready it will release fores, heads, hearts and tails in order as the ratio of water and alcohols change, and it reaches the right temperatures . We can not force it or control it by pushing a button or turning a knob.

[NZChris]

For a Pot still I think it would be cool if you could measure production rate with the PID so it could adjust power accordingly for a proper flow.

That is a way proportional control is useful for a pot still. Few forum members do it, maybe none except iStill users. I don't bother because I'm not commercial and tweaking the power a couple of times during a run to keep the flow rate up is enough to get the job done and let me get to the pub on time. Sometimes I don't even bother doing that, choosing to run two elements in series and not using a controller for the entire run.

[cayars]

Several commercial distilleries have and use PID controllers.

I'll bet they do too, as there are places in a distillery where they are useful, but I'll also bet that they are not using them to control the heat input to the boiler by reading a temperature somewhere in the boiler like you are telling us they do.

You can find some limited info on how these are used on the ADIforums. PIDs are great for controlling ball valves and the like which are more common on bigger stills then electric elements. PIDs come into their own when you programmatically control them as you can setup complete automation and have other "monitors" influence thins beside just what a single PID could do on it's own.

Can we stop concentration on PIDs? The thread was about temperature control and there are many other ways to control temperature than a PID successfully.

You brought them up as a control option and are the one who claims they work.

Actually I wasn't the first person to mention them in this thread. I was defending or showing how they could be used.

What a PID controls is temperature, not boiling point, so they are excellent tools for that purpose, but when you set a PID a target temperature in a pot still, the PID decreases the heat input to the boiler as the constantly rising boiling point of the charge approaches the set point. What you have controlled is the heat input, not the temperature. It doesn't make any difference what type of PID it is, or how much you spent on it, it will only ever interfere with the smooth operation of a pot still.

Well in this case the PID is controlling heat input to the boiler. It does this by monitoring other variables like temperature of the boiler and flow rates.

Cayars, in you example of a 10% abv boiler charge, in which you predicted the temp and output,
How do you account for the varrying amounts of methanol and acetone and fusel oils (among other things)? You cant. Unless you send a sample of every ferment off to a lab to have it analyzed, what you've proposed just wont work in the real world.

I don't care about that and neither do you when you run any still. You only care about the overall ABV of alcohol which will dictate the boiling point at any given time in the boiler.

Your methanol, acetone and fusel oils will come over as normal just like they always have. No magic going on here.

Furthernore...
When I charge my still I can either do a spirit run or a stripping run. Either way, I get the same volume of the same products, over the same range of temps. Distillate might start coming over at around 80*f and the run is done around 202*f.
If I do this as a stripping run, I apply more heat. The result is more smearing but a quicker run.
If I do this as a spirit run, I apply less heat and the result is less smearing and a longer run.
But either way the temps dont change. The run still goes from 80*-202*. Only things that is changing with my energy input is the amount of time it takes to do the run and how mu h smearing takes place.

Please explain how those temps fail to change at all based on your theories.

Exactly, it's very predictable which is why it can be controlled (manually or automated). When you are just at the boiling point of the fluid in the boiler you will have little to no flow. But as soon as you add additional heat past the boiling point you add more energy to the fluid and will cause it to boil more violently and the vapor coming off will be greater.

Once at the boiling point (which will change throughout the run) the change of heat (staying at or above boiling point) will cause the vapor rate to change. That is the difference between a strip or spirit run and how much smearing you get. It's all predictable, programmable and controllable.

Think about it. If you wouldn't consider turning coolant flow completely on or off, then why would you turn heat completely on or off to control temps?

I do that with all of my pot still condensers

That was an incomplete thought as expressed. That was in relation to a reflux still and controlling the amount of reflux based on controlling the coolant in some way (needle valve, coil height adjustment, coolant flow rate, coolant temp, etc). My point is that you wouldn't want to control the reflux by only turning on/off the coolant (pulsing) and neither would you want to do this with the boiler. Just ignore this as it's trivial and not needed to the conversation.

In agreement with the statement, but it is still temperature related..

Rate of the boil = input energy + boiler load ABV = which generates a temperature level which in turn generates an evaporation temperature.. (as the temperature of the boiler load will always be different than the temperature of the evaporation temperature until they become equal)..

So.. The rate of energy input is just a means to generate an evaporation temperature and therefore temperature driven..

In a pot mode, if one leaves the energy input at the same level through out the run, as set at the start, the output flow will almost come to a stop, and it is the same behavior in a flute when the first plate dries out..

Reason.. the evaporation level of the ABV is depleted, therefore temperature related..

Mars

ABSOLUTELY, couldn't agree more!

[cayars]

For a Pot still I think it would be cool if you could measure production rate with the PID so it could adjust power accordingly for a proper flow. If it were a fractioning column, measuring vapor temp, reflux rate and production rate could allow you to control power and cooling flow for a run. That would be a cool use for a PID. Of course, you need to know how to tune these things manually before you could ever program your system to do it for you and by that time ya gotta think why bother, unless you are commercial.

Otis

Well if you can monitor your boiler temperature and your flow rate you can automate it if you choose to do so.
Actually, instead of directly measuring flow rates you could measure weight instead. Imagine having your collection jar sitting on a smart scale. By smart scale I simply mean a scale of good resolution with a computer interface (think something like a postal scale but better). Now at any given time you can know the weight of the contents of the collect jar.

By monitoring the weight of the collection jar you can determine the rate of output. If weight isn't increasing you don't have enough heat input (if at boiling). If weight is increasing too fast you have too much heat and can cut back until the weight changes to match a specific flow rate you wish to have.

Not only could you use the weight of the collection jar to determine the flow rate but you could also programmatically know when the jar is full or exactly how many ML of content is in it. You could ring a bell/alarm at jar change points. You could decrease heat input at X ML for foreshots, you could take off Y Heads, Z Hearts based on weight as well.

Technically you could probably drive your heat input strictly by the scale if you wanted to as you could easily determine how many ML per second (any time frame) and adjust heat to match a stripping run or a spirit run. Of course the more things you can monitor (that make sense) the better and the more control you would have over the complete process. Hell, you could even add an electric parrot if you wanted which would give you a lot data to work with as well.

Instead of being "temperature driven" it could be "weight driven" or "flow rate driven" design.
Using some type of computer (computer, tablet, cell phone, Pi, etc) to drive it you could setup recipes. Mix up a batch of Odin's Easy Gin, kick off the program and get an alarm at 50 ML to dump the foreshots, then it runs the still until you get to 400 ML and it shuts down the heat for a perfect run of OEG. Of course you might need to shut off at 395 ML and allow the last few drips to come through to get the "perfect" 400 ML but that's just fine tuning for your particular still.

But the point is with only a couple of monitoring instruments you could build automation into any still including a pot.

[still_stirrin]

... if you could measure production rate...

Well...instead of directly measuring flow rates you could measure weight instead...

By monitoring the weight of the collection jar you can determine the rate of output. If weight isn't increasing you don't have enough heat input (if at boiling). If weight is increasing too fast you have too much heat and can cut back until the weight changes to match a specific flow rate you wish to have.

Point taken.

But....what if the density of the collection fluid was changing? Then, the volume would be dynamically changing too. So, you’d need a derivative specific gravity reading as well, and then you’d need to integrate this over time as the jar fills.

Not only could you use the weight of the collection jar to determine the flow rate but you could also programmatically know when the jar is full or exactly how many ML of content is in it. You could ring a bell/alarm at jar change points. You could decrease heat input at X ML for foreshots, you could take off Y Heads, Z Hearts based on weight as well.

This is another “rookie” mistake, thinking that specific gravity (or %ABV measurement) is an appropriate determination of cut points between the chemical fractions. It just isn’t so...and will lead you astray every time.

Instead of being "temperature driven" it could be "weight driven" or "(mass) flow rate driven" design.

Absolutely a better concept than trying to measure temperature and use that as the “scaler” to regulate heat input (production rate motive force). The temperature of the boiler (and correspondingly, vapor temperature) is controlled by physics...the composition of the mixture, which incidently is NOT purely ethanol and water although those will be the primary two constituents. Some consideration should be given to the many congeners also ”along for the ride”. How do you analyze those?
ss

[cayars]

But....what if the density of the collection fluid was changing? Then, the volume would be dynamically changing too. So, you’d need a derivative specific gravity reading as well, and then you’d need to integrate this over time as the jar fills.

I doubt it would change enough to throw any simple use like this off. Maybe if you had a puke during a strip run or something like that but of course you are still there while it's running so you would notice such an event.

You have me curious now how much the weight would change for a jar right off the still. I'm running some George Washing Rye, Single Malt and some Rum today/tomorrow so I'll have to weigh specific ML amounts and see how much this would change.

But for simple use to augment temperature you can certainly see if the weight is increasing or not to know if more heat is needed.

This is another “rookie” mistake, thinking that specific gravity (or %ABV measurement) is an appropriate determination of cut points between the chemical fractions. It just isn’t so...and will lead you astray every time.

I beg to differ. I know upfront where my cut points will be on most of the things I run BECAUSE I've ran them so many times before. I hardly ever taste stuff coming off the still anymore for things I run often as I just know by the proof where to cut and it's like clockwork. I'd find this pretty easy to automate based on ABV of the pot. For strip runs I don't care about cut points and would just run it down to 5% to 10% using an electronic parrot if equipped.

Absolutely a better concept than trying to measure temperature and use that as the “scaler” to regulate heat input (production rate motive force). The temperature of the boiler (and correspondingly, vapor temperature) is controlled by physics...the composition of the mixture, which incidently is NOT purely ethanol and water although those will be the primary two constituents. Some consideration should be given to the many congeners also ”along for the ride”. How do you analyze those?
ss

I would think it best to combine both temperature and this type of simple flow rate monitoring.

[pope]

How do you measure flow rate? Is there actually a mechanism to do so or do you need to do as Otis suggested and draft your own correlations between temperatures at different points in the boil, which also hinges on the starting abv of your charge, no?

I am in the process of experimenting with/migrating from phase angle control to ssr control with PID's. I adore the simplicity of a PSR but a PID or series of PID's opens doors to new possibilities.

I also wouldn't say I control temperature in the boiler or anywhere else, but I am starting to use boiler temperature to inform the PID so as to control the amount and rate of heat input. I wouldn't say I control anything, really, so much as just corralling vapors. More of a vapor rancher than a distiller. Those vapors will do what they want to do based on the particular chemical composition of what's in the boiler, I just prod them with a little heat and then lead them to the glass. Knowing temperatures is just helpful estimating how those vapors might smell and taste without drawing samples or getting my face in there.

[still_stirrin]

I know upfront where my cut points will be on most of the things I run BECAUSE I've ran them so many times before. I hardly ever taste stuff coming off the still anymore for things I run often as I just know by the proof where to cut and it's like clockwork. I'd find this pretty easy to automate based on ABV of the pot...

This may work successfully for YOU depending on the equipment you use, the experience level you have, and the controls you manage with the recipe and fermentation processes. Consistently controlling this “dog’s breakfast” of variables is key to repeatability for which you rely to “assume” your cut points.

However, other conditions may change the constituents and their ratios in your low wines. And, if using a reflux column, the proof offstill will not always signal when you’re through the heads nor when the tails come in.

For example, on my reflux column, I can push the %ABV to azeotropic throughout the run. In fact, often the %ABV (measured by density of the distillate, not a spectrometer) will fall slightly after the fores and during the heads portion before rising again when into the early hearts. The proof offstill will stay at azeo through the balance of the run until my VM stops producing. It is then that I know (not by proof offstill, but by flowrate) that I’m to the tails. You see, a VM is very good at pushing the tails to the end and then simply reduce production nearly to a stop (without adding more heat to the boiler).

Similarly, the LM will compress the heads towards the foreshots and allow me to take those off early, managing the heat input with the distillate take off rate. When I’ve collected the portion I anticipate for fores + early heads, I usually switch over to the VM control to navigate to the backend.

My point is (as you will agree), that the portions (volume in the jars) are “estimated” from the quality of the low wines and experience of operation. This variable is difficult to quantify into a computer program and can and will vary run to run and operator to operator. But the cut points for my keep depend on my senses and the “blending” of jars on the outside of the obvious “hearts keep” jars.

Perhaps a computer automated system will help you “monitor” progress, but I still believe that your senses are the best “tool for the job” of selecting what goes into your liquor cabinet and what gets recycled in the manner that you do.

ss

[still_stirrin]

How do you measure flow rate? Is there actually a mechanism to do so...

It can be done.

In fact, I believe NZChris developed a liquid flow measurement mechanism several years back. It used a cup of known volume and a timing device to measure volume/time. I don’t have the thread handy, but an exhaustive search should find it. Not sure, but he may even employ the equipment now (still).

But commercial flowmeters do exist. Laboratory flowmeters have a floating ball in a clear tube with a weir (metered flow outlet) that would fill to indicate the liquid flow throughput. Also, in petroleum production plants, flowrates are measured within a valve body that has a rotating vane. The output is a proportional voltage to send feedback to monitoring systems. As a note, this type of flowmeter is what measures your fuel at the gas pump too.

In the end, flowrate of the distillate would be a better feedback loop to control heat input, and the corresponding distillate production including consideration of the reflux ratio and vapors condensed and returned down the column.
ss

[Hillbilly Popstar]

Seems you're saying the same thing we are.

What you have control of is RATE of boil. But you cannot immediately change or control the temp of the boil or the vapors anymore than you can tell the sun when to rise.

Someone you seem to equate that control over vapor speed or boil rate to being able to control the temp.

That's fine and all if that's how you see it in your head, bit the way you word it could be very confusing to a novice distiller.

Also, methanol, acetone, acetyl alcohol, propylene, fusel oils, and all the rest have as much effect on the boiling point as the ethanol content. They are the very reason a run starts at 180 and finishes at 205 (give or take...) So yes, I do care about them when I run my still.

[NZChris]

In fact, I believe NZChris developed a liquid flow measurement mechanism several years back. It used a cup of known volume and a timing device to measure volume/time. I don’t have the thread handy, but an exhaustive search should find it. Not sure, but he may even employ the equipment now (still).

It wasn't me. My flow sensor is an old eyeometer I've had for a while.

[OtisT]

How do you measure flow rate? Is there actually a mechanism to do so or do you need to do as Otis suggested and draft your own correlations between temperatures at different points in the boil, which also hinges on the starting abv of your charge, no?

You have me confused with someone else. I was not advocating what you suggest. I would want to know actual production rate, regardless of the method, to make a PID even remotely desirable.

My point was that without your PID knowing actual production rate when pot stilling, it would be nearly useless to me IMHO. I say “nearly useless” because by knowing temps it could still be used for a few simple things when pot stilling . 1) by knowing your charge temp it could warm up the charge at high power then turn power down to your desired level just before the boil. 2) shutting things down when you are done, simply by knowing boiler or vapor temps. All this is painfully easy to do manually and I would not consider making any cut decisions by temp, even on a recipe I know well. Yes, I use temps and I use the % location by total run volume to give me a ballpark as to where cuts may be located but that is all it is, a ballpark.

Otis

[pope]

If it were a fractioning column, measuring vapor temp, reflux rate and production rate could allow you to control power and cooling flow for a run. That would be a cool use for a PID.

I just misinterpreted what you said is all. I found a PID that does 1 and 2, as you pointed out those are pretty easily controlled manually, but they're nice added features along with other bells & whistles. I think for now though I'll ultimately break out the ol' beaker & iPhone timer and check my flow rate once or twice per run. It didn't take long to get a feel for good flow. BUT, if anyone here gets a proper flow-driven controller working, I want to see pictures!

[Yummyrum]

Sorry of this is off topic but this thread went that way long ago

How do you measure flow rate? Is there actually a mechanism to do so...

It can be done.

In fact, I believe NZChris developed a liquid flow measurement mechanism several years back. It used a cup of known volume and a timing device to measure volume/time. I don’t have the thread handy, but an exhaustive search should find it.

I had this wacky idea , was this the one Still_stirrin?
viewtopic.php?t=70272&p=7518434#p7518434
It would fill a 50ml tube until the float triggered the controller to flip the feed to a second tube . It would then open the drain valve for long enough ti empty it then close it again . Then the same thing happens ti the other tube before swapping back to the first .

The second idea was to have the output fill jars with a servo controlling the feeder to the jars .
Never built it though .

I thought about using weight too but the density problem put me off it but with an Eparrot , corrections could be made yo keep it correct .

[OtisT]

If it were a fractioning column, measuring vapor temp, reflux rate and production rate could allow you to control power and cooling flow for a run. That would be a cool use for a PID.

I just misinterpreted what you said is all. I found a PID that does 1 and 2, as you pointed out those are pretty easily controlled manually, but they're nice added features along with other bells & whistles. I think for now though I'll ultimately break out the ol' beaker & iPhone timer and check my flow rate once or twice per run. It didn't take long to get a feel for good flow. BUT, if anyone here gets a proper flow-driven controller working, I want to see pictures!

Ya, no problem Pope. I just didn’t want folks think I am advocating cuts by temp. I do think it could be used to help you tune the system for max production while maintaining desired ABV outputs.

I would also like to see if/how reflux and production rates could be measured on a hobby scale. I’m sure it can be done, but at what cost? I worked in the pharma industry and helped with the build of a .5 billion dollar batch production plant from scratch. lots of cool flow control devices but boy do those things cost major bucks.

Otis

[6 Row Joe]

If it were a fractioning column, measuring vapor temp, reflux rate and production rate could allow you to control power and cooling flow for a run. That would be a cool use for a PID.

I just misinterpreted what you said is all. I found a PID that does 1 and 2, as you pointed out those are pretty easily controlled manually, but they're nice added features along with other bells & whistles. I think for now though I'll ultimately break out the ol' beaker & iPhone timer and check my flow rate once or twice per run. It didn't take long to get a feel for good flow. BUT, if anyone here gets a proper flow-driven controller working, I want to see pictures!

Ya, no problem Pope. I just didn’t want folks think I am advocating cuts by temp. I do think it could be used to help you tune the system for max production while maintaining desired ABV outputs.

I would also like to see if/how reflux and production rates could be measured on a hobby scale. I’m sure it can be done, but at what cost? I worked in the pharma industry and helped with the build of a .5 billion dollar batch production plant from scratch. lots of cool flow control devices but boy do those things cost major bucks.

Otis

Interesting stuff Otis. I love looking at that stuff and figuring out how it works.

[cayars]

Just curious how many of you guys have seen this iStill video where Odin shows some of the temperature based set points used during the run.
Of course this is in reflux mode but it's an interesting video to watch if nothing else but to see how cool his iStill technology is:

[pope]

I would love to be able to measure reflux rates. Especially as I take the final leaks out of my new dephleg, refluxing an appropriate amount is of sudden, great importance to me. I come from a non-technical field so if its technical I learned it here first, but I have a posh coffee 'dosing' grinder and a buddy made a comment that it was smart to calculate dosing using time to estimate volume because it was reliable. Unfortunately there are more variables in our systems than an otherwise inert grinder with uniform inputs.

What about a little ss water wheel in the reflux chamber, and we can just count the rpms?

[NZChris]

There are many tricks you can use to automate various functions associated with running a pot still, (I use some of them), but those that are not related to the OP would be of more value if discussed in other threads already existing, or new, so as not to muddy the waters in this one.

Rant over

[Yummyrum]

There are many tricks you can use to automate various functions associated with running a pot still, (I use some of them), but those that are not related to the OP would be of more value if discussed in other threads already existing, or new, so as not to muddy the waters in this one.

Rant over

You are bang on NZChris
I thought about slicing and dicing this thread up ... it has become so off topic it’s not funny ..... it would need multiple new or existing topics to deal with what has been discussed .

What I want to say as a Mod is how proud I am of all of you for how this has progressed without the usual crap that has gone down in the past .
You are all a good bunch of fellas

It is great that we can discuss things civilally here again .

[cayars]

Personally, I'm thankful you have let it run without splitting the thread as it's as it's all mostly inter-related.

What I find amusing is how we all mostly agree on the principles of how things work, but then disagree if it's "temperature" controlled or not depending on point of view. It's like the "which came first the chicken or the egg" which is a matter of perspective.

This thread has got me thinking of something I'm playing with and testing today, that I likely wouldn't have done otherwise. This is the collection jar weight idea using a scale. It will be interesting to see how much the density (or weight specifically) changes based on what is run through the still. Strip vs Spirit run. Rye wash vs single malt vs Rum. Then being able to determine if the density change matters or not for "hobby" use or if the flow rate can be pretty accurately calculated based on this weight. I'll be using a postal scale I have with USB interface but it's a start.

[kimbodious]

If you’re using something like an eparrot you wouldn’t need to measure the weight, and the temperature and volume to calculate the density. You can extrapolate that from the ABV.

[StillerBoy]

It will be interesting to see how much the density (or weight specifically) changes based on what is run through the still.

Weight specific would be more accurate to work from, but I'll stick to using time factor for my flow rate adjustments.. it's work well for me as I well trained at it now..

Furthermore.. I don't need more toys to play with.. lol

Mars

[cayars]

What I have now works for me as well and don't plan on changing anything either.
I'm just kind of curious how much difference the density or weights will change based on different mash bills and strip vs spirit runs.

I know the differences in mole weight to volume of pure water and pure ethanol but I'm more curious at our typical ABV how much this will matter using common "smart scales" anyone can purchase and if the resolution would be "good enough" to use to calculate flow rates.

It's one of those things that will bug me till I know the answer. LOL

[StillerBoy]

It's one of those things that will bug me till I know the answer. LOL

Hear you on that one..

Some years back, played around temperature sensor probes and software until a got a handle on vapour behavior, now I just use three regular meter/sensors points, and that provides me with all the data I need on a run..

What I notice was that the low wine run of some of wash/mash was noticeably different, enough to required a water flow adjustment for better reflux.. otherwise the three point temperature points give all the data needed for a quick setup..

Mars

[Acrylic]

Re: Cayars....>Compare that to just cranking up the heat quickly bringing everything to 100C and getting everything smeared very badly across the whole run. Surely the temperature is at play here as a higher than needed temperature will increase the vapor rates. Many of the still automatons make use of boiler temp, column temp, flow rates, etc and adjust things on the fly specifically to control different temperatures to get cleaner cuts and to compact heads and tails.<

Forgive me I'm new... I've read and re-read but I just cannot comprehend this- how is it possible to bring everything up to 100C? There are alcohols in the mix that would boil off at lower temps which would hold the mix at that temp, more or less to keep it simple, and once those alcohols boiled off, the mix would rise in temperature to the boiling point of the next alcohol. Yes, I'm obviously missing something but I just cannot understand quickly bringing everything to 100C. Perhaps, if it was a sealed boiler and under pressure?

thx

[cayars]

Forgive me, didn't mean to sound like you could get 100C right away as the boiler will get to boiling temp much earlier than that with alcohols and other things in it. Remove the reference to "100C" if that helps. I was likely referring to being able to run more heat than needed which would run the whole boiler down to nothing and end up at water boiling temp (over driven) which of course would smear everything along the way. Very poorly worded on my part.

No you aren't missing anything. The boiling temp will rise as the run progresses until it gets to 100C or 212F (or close) and no alcohol is present theoretically.