PID success, maybe

[YankeeShine]

Well my plan to use a PID to automate a run has come together….sort of. I posted about this a while back but it took me forever to get set up, I had issues trying to get my reticulating cooling working. But now I have a 20 gal pot with a simple copper still head and liebig for cooling with a 55 gallon drum reservoir. It was supposed to be two 55 gallon drums but one sprung a leak. I use a 5500W element controlled by a PID. I taped the PID temp sensor on the middle of the liebig and set the temp for one degree over ambient (after circulation had been on for a couple minutes to establish it). It worked as it should, never overshooting the temp (the safety feature I was trying to implement) and maintained a constant pencil thick stream for the first portion of the run. Then something unexpected happened, the stream started to trickle off about a third of the way in. I didn’t have time to let it go and see if it came back, I bumped the temp up another degree and the stream came back. This happened two more times before I called it quits. During the run I realized that I had a vapor leak at the still head (I did a poor job on making an easy flange).

So I’m not sure if it was the leak or rising temps in my 55 gallon reservoir (I assume they were rising, I couldn’t tell by touch so it was slight) that caused the stream to cut back like it did. Or if there’s something in the physics of the distillation that I still don’t understand because I was really expecting that I shouldn’t have to change the temp at all once I found the degree that produced the size stream I wanted. The leak is now taken care off, so we’ll see what happens this weekend with another run. And the run after that I should have a larger reservoir which should eliminate that variable. Even if I don’t get it to where I don’t ever have to adjust temps like I want, I’m pretty pleased with being able to do a whole stripping run and only adjusting three times and being confident that I don’t have to be right there to cut the heat back to avoid alcohol vapor spewing out the liebig because I wasn’t paying close enough attention. I intend to always be there to cut the foreshots, but for anyone who’s ever let the temps run up to high too fast for whatever reason, I’m sure they can appreciate this safeguard). So if I had to choose between a $3 potentiometer and a $20 PID, it’s a no brainer. $17 is cheap money for insurance against a potential disaster at worst, or wasting a bunch of product at best.

Anybody hazard a guess why I needed to turn up the temp those couple of times?

[kaziel]

I can't read English that long (your post is too long 4 me ) but I think maintaining const temp on a pot still head will make you adjust temp because when more etoh evaporates temperature of vapor raises. Trying to keep it steady will cut output. Maybe I'm wrong I don't know if that's answers yours question.

[Dnderhead]

http://homedistiller.org/forum/viewtopi ... 65&t=16635
http://homedistiller.org/forum/viewtopi ... 65&t=13294

[heartcut]

Good thing is most PID controllers have a manual proportional out setting where you can set the heat (by % of output) and it'll pulse your SSR and ignore your temperature input.

[YankeeShine]

My directions say that it has a manual mode but I can't get it to work. It also says that I can use Celsius or Fahrenheit but I can’t, its stuck on Celsius. You don’t really need a manual mode though. Not much difference between bumping up power out a %, or bumping up the sensor temp a degree.

If the PID won’t work the way I want, does anyone know of an inexpensive flow meter? That would be pretty cool, set it for a drip and just let it run all day. Is there such a thing as running a still too slow?

[acfixer69]

My directions say that it has a manual mode but I can't get it to work. It also says that I can use Celsius or Fahrenheit but I can’t, its stuck on Celsius. You don’t really need a manual mode though. Not much difference between bumping up power out a %, or bumping up the sensor temp a degree.

If the PID won’t work the way I want, does anyone know of an inexpensive flow meter? That would be pretty cool, set it for a drip and just let it run all day. Is there such a thing as running a still too slow?

There is a huge difference and need to understand that first.

AC

[YankeeShine]

My directions say that it has a manual mode but I can't get it to work. It also says that I can use Celsius or Fahrenheit but I can’t, its stuck on Celsius. You don’t really need a manual mode though. Not much difference between bumping up power out a %, or bumping up the sensor temp a degree.

If the PID won’t work the way I want, does anyone know of an inexpensive flow meter? That would be pretty cool, set it for a drip and just let it run all day. Is there such a thing as running a still too slow?

There is a huge difference and need to understand that first.

AC

How so? As I understand it, if my manual setting worked, I could set % of power going to the element. But since it doesn't, I just set a value over ambient temp so that the PID cycles the element on and off (at full power) to raise the temperature. So if I want more power, I could either increase % output, or increase set value, which causes increased fire frequency. Either way introduces more power to the system, so why is it a huge difference? Granted the quickest way to get distilling would be to have manual setting on the PID, or use a potentiometer, so I could just have the element fire full on until I decided to back it off, but the benefits of not having a set temp overshot is worth the 15 minutes extra heat up time too me.

[MuleKicker]

If you are in manual mode and you set the controller at 50% and lets say you are running 10k watts, your output will be 5kw of power to your element. If you just bump the set point up 1deg like you are thinking, the controller is going to throw 10kw at your element until you reach the set point. You cant chase temp, you cant chase temp, you cant chase temp. Get it through your brain. Did I mention, you cant chase temp? Get the basics before thinking you can just automate everything.

[heartcut]

What you're proposing is trying to get a constant heat output by adjusting the setpoint in automatic as the temperature changes- not too accurate, lots of little adjustments and you'll never get the stability that a manual setting would achieve, making for a little less quality in your distillate. Might consider learning how to operate your PID or picking up one that works. You're getting good advice from a number of experienced stillers, don't think they'd all be saying the same basic thing to you if some of them hadn't been there already. Good luck.
Edit: posting while MK was, we're still both singing the same song.

[bellybuster]

did you say "you can't chase temp"?

[YankeeShine]

If you are in manual mode and you set the controller at 50% and lets say you are running 10k watts, your output will be 5kw of power to your element. If you just bump the set point up 1deg like you are thinking, the controller is going to throw 10kw at your element until you reach the set point. You cant chase temp, you cant chase temp, you cant chase temp. Get it through your brain. Did I mention, you cant chase temp? Get the basics before thinking you can just automate everything.

I don't think you understand the math. 10kw for 30 seconds is the same as 5kw for 60 seconds, right?

Also, I'm not talking about controlling the temperature of the boiling liquid. I'm talking about controlling temperature of the condenser by adjusting heat input. It worked to the point where, despite a vapor leak, I had a successful run with the exception of I had to adjust temp on the pid one degree three times. So unless people can be helpful and explain why temp adjustment was necessary if they can, I'm not interested in "it wont work because I said so". Because to a significant degree I've already proven that a PID is a great tool. I know the basics, I get my friends and family hammered drunk on good tasting likker all the time, that's about as basic as it gets right? Now I'm talking about changing my method for its own sake, not improving quality or anything. I like what I'm drinking.

I don't mean to be a pain, I just wish that that one, someone would acknowledge that temp in the condenser is different than temp in the boil, and two, every distiller monitors temperature. If you do it the old fashioned way of touching the still head and watching the thickness of the distillate stream, or use a thermometer, its the same, your getting information so that you can make a judgment about how much heat/power to input into the system. Again, I'm in no way saying that I'm trying to monitor or control the temp of the boil so all the long threads and great information about why that won't work are not helping me. Has anyone else ever put the PID sensor in the condenser? Those are the results, good and bad, that I'm interested in.

[Dnderhead]

"" 10kw for 30 seconds is the same as 5kw for 60 seconds, right?""
and if driving a car,,
it will be the same if i go 60 MPH for for 5 seconds ,stop for 5 seconds ,then do it gain rather than gong steady at 30MPH.?

[YankeeShine]

"" 10kw for 30 seconds is the same as 5kw for 60 seconds, right?""
and if driving a car,,
it will be the same if i go 60 MPH for for 5 seconds ,stop for 5 seconds ,then do it gain rather than gong steady at 30MPH.?

Sort of. 60mph for 1 minute is the same as 30mph for two minutes in that you travel one mile. But the PID doesn't fire constant, it turns the element on and off at varying frequency (sensing velocity of temp increase in between firings), so that is why it takes a little longer for the PID to input the desired amount of power than you could get with a manual setting or a potentiometer. The PID does this so that the desired temp is not overshot. But both ways are means of controlling power input. So if you want to impart power on a mix to be boiled, you can set the PID at a temp somewhere over the boiling temp, and it will fire on and off until it gets there. If there is a large variance between the starting temp and temp you set, the PID, aver testing a few times, will fire with a fast frequency and then slow (based on an algorithm) as it nears the set temp so that the temp never exceeds the set temp. I have my sensor on the liebig arm so I set the temp one degree over the libigs starting temp, which is way below the boiling point of the mash, so the the PID fires as fast as its going to, boiling the mash and sending vapors into the condenser. As the temp in the condenser rises, fire frequency is reduced, cutting back power to the boiler. It is essentially running the still the same way I do with propane. I adjust the flame so that the point of cool water going up the arm turns hot is right in the middle of the arm, which is where my PID sensor is placed.

So in your example of a car, driving 60 mph for an hour would get you 60 miles. Driving 60 mph for a half hour, stopping for five, then driving 60 mph for another half hour still gets you 60 miles, it just takes 65 minutes. Its different, but not hugely different because the end result, in terms of distance traveled, is the same.

[Dnderhead]

id like you explain that to a cop..yes i did 60MPH in a 30 zone but i stopped in between times so i averaged 30 MPH. so i was in the speed limit.
no you need steady heat not stopping and starting."to a average"

[Prairiepiss]

If you don't know why you had to bump the temps up. You don't really understand the distilling process. You were provided a couple of links that will help you understand why.

[MuleKicker]

I don't think you understand the math. 10kw for 30 seconds is the same as 5kw for 60 seconds, right?

Also, I'm not talking about controlling the temperature of the boiling liquid. I'm talking about controlling temperature of the condenser by adjusting heat input. It worked to the point where, despite a vapor leak, I had a successful run with the exception of I had to adjust temp on the pid one degree three times. So unless people can be helpful and explain why temp adjustment was necessary if they can, I'm not interested in "it wont work because I said so". Because to a significant degree I've already proven that a PID is a great tool. I know the basics, I get my friends and family hammered drunk on good tasting likker all the time, that's about as basic as it gets right? Now I'm talking about changing my method for its own sake, not improving quality or anything. I like what I'm drinking.

I don't mean to be a pain, I just wish that that one, someone would acknowledge that temp in the condenser is different than temp in the boil, and two, every distiller monitors temperature. If you do it the old fashioned way of touching the still head and watching the thickness of the distillate stream, or use a thermometer, its the same, your getting information so that you can make a judgment about how much heat/power to input into the system. Again, I'm in no way saying that I'm trying to monitor or control the temp of the boil so all the long threads and great information about why that won't work are not helping me. Has anyone else ever put the PID sensor in the condenser? Those are the results, good and bad, that I'm interested in.

I dont think you understand the distillation process. You cant "control" the temp of the condenser. You control heat input to the still, more heat = more vapor to the condenser, not higher temp. So, doing it your way will get you a surge of vapor to the condenser. Did I mention, YOU CANT CONTROL TEMP!!!

[YankeeShine]

But I did control the temp of the condenser? I set it at 21 degrees Celsius and an it never went over that temp until I told it to go to 22 Celsius, and then a while later 23 and 24. If I wanted it at 20 (the temp of the cooling water and the lowest I could make it go) I could have set that temp and the PID would have backed off the power so that the volume of vapor to the condenser would have been reduced and the temp would have dropped. So how am I not controlling the temperature? Again, I’m not trying to control the temp at which something boils, there are laws of physics beyond my control that affect that. But the volume of vapor in a condenser affect the condensers overall temperate, so I control the volume of vapor entering the condenser by controlling the amount of power input into the boiler. I regulate this control based on the temperature of the condenser. I don’t know how you run your still where you don’t have any control over temperature, but I have almost infinite control over temperature in my condenser. I can make it as cold as I want by lowering the temp of the coolant and I can increase the temp by adding volume of vapor to it (via adjusting power input to the boiler). Again, I am not trying to control the temperature at which a mix boils, which is all those threads address. I’m simply controlling the temperature on the outside of my liebig water jacket by controlling the amount of power input to the boiler. Doesn’t everyone do the same thing? When I run propane, I turn it all the way up. Then I constantly feel the still. First the head warms up, then as the vapor builds up, it works its way into the liebig, and I can feel its progress through the arm as it heats up. As vapor begins to condense, I back off the flame to control the volume of distillate to a pencil lead thick stream and then do my best to keep it running that way. The PID appears to be doing the same thing. Instead of my hand on the water jacket, there’s a temp probe. When the probe notices its hotter than I want, it backs the power off (cycles element slower) resulting in a constant temperature at that point in the liebig arm. How is that not controlling temperature?

[YankeeShine]

I dont think you understand the distillation process. You cant "control" the temp of the condenser. You control heat input to the still, more heat = more vapor to the condenser, not higher temp. So, doing it your way will get you a surge of vapor to the condenser. Did I mention, YOU CANT CONTROL TEMP!!!

You are wrong, more vapor to the condenser means increased condenser temps. If you look at your condenser as a whole unit, with no vapor in it, it is 100% coolant and lets say 20 degrees. If you add a volume of 78.37 degree vapor into the condenser, the overall temperature of the condenser increases proportionally. So I can controll the condenser temp indirectly by incresing power to the boiler with increases vapor volume. No?

[MuleKicker]

So you're talking about controlling coolant temp? There are more variables to that then just power input. Still the point to be made is, you cant just turn your power full on and off without surging. You want to be able to ramp up your power based on coolant temp? Whats controlling coolant flow? And when you get your magical fully automated still, then what? Pour in a bucket of wash and walk away. Only to come back to a jug of perfect likker, right?

[YankeeShine]

So you're talking about controlling coolant temp? There are more variables to that then just power input. Still the point to be made is, you can’t just turn your power full on and off without surging. You want to be able to ramp up your power based on coolant temp? What’s controlling coolant flow? And when you get your magical fully automated still, then what? Pour in a bucket of wash and walk away. Only to come back to a jug of perfect likker, right?

No, I'm not talking about full automation of a finish run (yet). I’m talking about automating a stripping run without cuts. My intention was to charge the boiler, set the temp, and then have the still run until it stopped producing. I think that if I find the right placement for the temp probe, and the right length liebig arm, I could do that. To prove it I hope to have a full day to run this next batch. I think the way I have it set up, it should run as it did last time, but when/if the output slows, I will leave it set at the initial temp, and I think I will see the actual temp at the sensor location fall, and the PID will autocorrect and increase power input which will return the output volume to the initial setting. This should happen until all achohol that can be extracted at that temp setting is distilled, and then the system should reach a point of equalibruim and not produce any further distillate.

Also, the fractions of a second the PID turns the element on and off don’t cause any surging that I can identify.

I’m using a 55 gallon drum reservoir and a recalculating pump. Right now, I can’t change the flow rate or the temperature of the cooling water. My original plan was to use 110 gallon reservoir but one of my drums sprung a leak. I figured with 110 gallons the temperature wouldn’t rise enough to be much of a factor. Based on my first run though, I’m pretty amazed because I couldn’t detect by feel any increase in reservoir temp with just the one 55 gallon drum. I’m going to take one of those aquarium stick on thermometers and stick it on the bottom of my liebig so I can monitor the actual temp a bit better out of curiosity.

What other variables do you see affecting my coolant temps other than power input? Is vapor temp one of them? Because I’m wondering if the reason I had to bump the temp up is because the vapor temp changes at those points where the distillate stream slowed (maybe indicating a good spot for cuts?) based on the same prinicpal that causes boiling point to change as alcohol vapor leaves the mix.

I do think with the right tools the entire distilling process can be automated. Is your apparent distain for that that because you are a craftsmen who doesn’t like the idea of a machine doing what he does (I can respect that and have similar prejudices myself about my woodworking), or is it because you know from practical thought and experience that it is not possible? The automation I envision is that if you can automate the still to the point where the output volume is consistently what you want from start to finish (I think I’m almost there) and you can automate the placement of the cuts (a small mechanical challenge), you could set them conservatively enough so that human senses are not needed. Then it will just be blending the cuts based on taste that is up to an individual. But it should be possible to automate distillation so that there is no question that the harts produced are the purest part of the run and though a human may capture more of them, what automation produced should be no different than what was done by hand; up to the point of blending at least.