DIY Vacuum Still Part 1 focused on the concept of process (form) where Part 2 is focused on how to put the hands on the system (function) and make it work. This is where form meets function. It really forced me dissect the process and answer lingering questions I’ve wanted to take look at. To my great happy surprise it turns out this is easier than I thought. This has been a great learning experience, very fun, and a ton of effort but I’m really excited to show you the economical and fun DIY way I solved the expensive ethanol recovery problem for home cannabis and CBD crafters everywhere.
I’m presenting an overview of what I’ve learned for educational purposes. It’s a starting point and launchpad for anyone that wants more control over their ethanol extraction processing science. This post will focus on the hands on aspects of working with a DIY vacuum still.
Explaining concepts and theory that are unfamiliar to readers can get word heavy and confusing. Especially, when done in great detail. Please watch this video demo to see the system, getting it going and that the reality is actually very easy compared to reading about it. This video should breath great life into the words that follow, then come back to it later when necessary to tie the prose together.
DIY THE SYSTEM TOGETHER
Putting together and working with a system like this isn’t for everyone. It takes up a bit of space and requires some hands on attention. However, the advantage to putting your own system together is that it’s almost infinitely customizable to any individual budget and need. It’s for those that want real control over their cannabis extraction craft with great economics. I would suggest it’s better, faster, cheaper and more fun than any alternative!
GREAT VALUE: MORE FOR LESS
Getting started can be as inexpensive as around $350. If you already have a vacuum chamber and pump it can be half of that. Adding more expensive component options can get up to about $950 but that’s pretty much the upper limit. Both price points are for pretty much any volume you want to work with. The integrated in-process-feed makes volume a non-issue!
This is why I’m so excited about bringing this project to the community. First, anyone can start at the low end price-point and upgrade components along the way as you build skills, confidence, and enjoyment. Second, the DIY systems already include components you would have to purchase separately in addition to buying a Source Turbo or EtOH PRO. The vacuum pump can be used for filtration and the recovery vessel and lid can be used with the pump for post processing as well. It’s a lot of bang for your buck!
I can put together a system that performs the same function as the Source Turbo ($600) but also includes more volume, faster processing, mixing capabilities for post process formulation or decarbing, filtering and post process purging capabilities. A magnetic stir plate, Buchner flask, condenser, reclaim and pump set for around $700. Those same capabilities with the Source Turbo would total about $1,300.
While on the topic of value, the starting $350 DIY version that also includes filtering and post processing capabilities is a great entry point and can do everything the $2,000 EtOH PRO can do and more. There are many combinations to work with and great economics to take advantage of.
For best performance, the system components should be arranged to take advantage of gravity with the vapor movement, condensing process and reclaim flowing downhill like a waterfall. The downward flow from the condenser to the reclaim vessel helps the system to self-clear.
Heat source (HS), Loading vessel (LV), Loading vacuum lid (LVL), Transfer tubing, Condenser (C), Reclaim vacuum lid (RVL), Reclaim vessel (RV), Cooling pan (CP), Vacuum pump (VP), Vacuum tubing
Heat source (HS)
This component became much less complicated once I realized the reasoning error I was making concerning heat application assumptions mentioned earlier. Almost anything will work from a cheap analog hotplate to a more expensive IH single cooktop with a low end temp of 140°F or less. Obviously no burners with open flames!
I like the IH option because of it’s very fast, even heat dispersion and it doesn’t have the hot disk heating element (often uneven) that the analog options do. For those unfamiliar with IH it’s worth mentioning that only IH compatible pots can be used with the cooktops. IH tops also have a lower profile and more surface area for stability and safety. Having said that, the inexpensive analog tops work just fine.
Loading vessel (LV)
The loading vessel can be a simple pot from right out of the kitchen cupboard. It’s important to know the vacuum lids are only available in limited sizes. The size of the lid that will be used must be known before an appropriate loading vessel can be chosen.
Non-stick surfaces work the best. Not because they are “non-stick” but because the slight texture allows heat to phase change and release much easier. Smooth surfaces like stainless steel and enamels don’t work well on their own. Smooth surfaces can work with the addition of boiling beads or chips to aid in energy release and boiling. Using beads or chips are a small complexity but it’s not overly onerous if you want to use stainless.
Loading vacuum lid (LVL)
I’ve been working with both polycarbonate and glass lids. Both have been working fine so far but when available I would opt for glass. Glass lids are currently only available with two-holes so I’ve chosen a two piece hardware set. One ball valve for a vacuum break and one straight port for the vapor exit. The ball valve doubles as an in-process fee port to load additional wash without stopping the process.
I’m currently working with 1/4 inch barb hose connectors because that’s what’s available. A larger hole for the vapor port exit would be more beneficial but I haven’t found an easy off-the-shelf option that yet. I hope to have that sometime in the near future.
The LVL seals have extra width for seating error but the vessel must be centered on the seal area. If the lid is too small with the rim of the LV close to the edge of the LVL it can cause a vacuum failure mid process and the lid can fly off and throw ethanol all over. This lid popping can also happen if the lid is the correct size but isn’t centered carefully at the beginning when initiating vacuum. It doesn’t happen when operated properly so take a moment to make sure the lid is centered well before applying vacuum and beginning the process as well as checking periodically throughout the process.
Transfer tubing is the tubing connecting different components that transfers air, vapor, and liquid through the system. All tubing must be suitable to work under vacuum and food safe with ethanol vapor. Most of the tubing I’ve been using is either 1/4 or 1/8 inch inner-diameter (ID) depending on the size of the connections on the lids and condenser. It’s ok if there are points where the tubing sizes don’t match. A simple 1/4 to 1/8 inch connector can be used to easily put them together.
Each section of tubing will be between 1-4 feet long depending the components are set up. Make sure there isn’t tension on the short sections of tubing causing them to kink and cut of any flow volume. I would HIGHLY recommend getting a bunch of butterfly hose clamps to secure the attachment at every connection. This small thing can save a lot of headache if something starts trying to leak.
I would advise not to skimp on tubing. I always buy more than I think I’ll need because like bacon, having too much is always better than not enough.
The single requirement for a condenser is that it provides adequate cooling to condense vapor from the LV into a liquid. Essentially, temperatures low enough for long enough to cool the warm vapor. Various combinations of dry ice, regular ice water bath, coolers, buckets, flexible tubing, stainless steel tubing, preassembled containments with coils, etc. have worked very well. I think it’s better to error on the conservative side that more cooling is better than not enough.
Some of the things I tested:
- 2 gallon bucket with 10 feet of 3/8” silicone tubing.
- Large rectangular cooler filled with ice and 20 feet of 3/8” silicone tubing.
- 5 gallon drink dispenser cooler with 60 feet of 3/8” braided PVC tubing coiled inside.
- Stainless steel 50 feet coil stretched in a cooler dry ice and ice bath.
- Stainless steel 50 feet tight coil in smaller cooler dry ice and ice bath.
- Pre-made stainless 50 feet coil condenser assembly dry ice and ice bath.
I haven’t used a “Jockey Box” made for drink dispensing but it looks like it would work well.
Make sure to clean any coils very well. Stainless coils can have oils and metal shavings in them that will otherwise end up in the reclaim on this run, then in an end product next time with the reclaim is re-used to make a wash.
When using ice it needs to be an ice bath, a combination of ice and water like what’s used in a champagne bucket to keep the bottle cold. If there’s no water the ice clumps together and melts away from the warm coil it leaving an air gap with no cooling effect. You will need a lot of ice so make sure to plan well ahead. “A lot of ice” means one or two large bags of ice sold at the store.
I’ve been experimenting with a hybrid cooling system of adding dry ice blocks in open food saver bags into an ice water bath. I’m clipping the tops together loosely above the waterline so dry ice stays dry and they aren’t in a closed bag that could build pressure and cause a problem. This has been working pretty well to keep the chill going.
Reclaim vacuum lid (RVL)
The RVL is similar to the LVL with the addition of a vacuum gauge. The straight port will receive the condensate from the condenser for collection, the ball valve connects to the pump and the vacuum gauge allows for monitoring system pressure levels. The ball valve connected to the pump opens and closes the system to vacuum during operation. The valve is in the open position when applying vacuum. Then, when the prescribed vac level has been achieved the valve is closed so the pump can be turned off. If the system is in balance almost the entire operating time will take place under this condition with the system closed and pump off.
A vacuum chamber set is a great option for this component because it can also be used for post processing. The set will include both the RVL and the RV at a reasonable price. If you already have a vac chamber set you may be able to use that! Just make sure the lid material and hardware is compatible with ethanol.
*Because this lid has three holes it seems polycarbonate is the only option currently.
Reclaim vessel (RV)
The RVL can be paired with about any RV that holds the required liquid volume, withstands vacuum pressure and fits the RVL properly. Another pot from the cupboard can work well for this, or, as mentioned above the RV and RVL can be purchased together as a vacuum chamber set.
Cooling pan (CP)
Nothing special needed for this component. All that is necessary is something plastic or metal that is shallow and that can hold the RV with some ice under it. Only the bottom portion of the RV needs to be cooled or sitting on ice. Do not submerge or make it difficult to access and handle the RV during process.
Vacuum Pump (VP)
When I began working with this project I was worried about the pump selection and expense. Once I came to understand a few important things I worried no longer. First, the pump down speed, even with slow pumps, is quite fast. Second, it’s possible to process with only about 80% of full vacuum. Third, and most importantly, working with the system in balance substantially reduces (nearly eliminates) in-process VP activity so vapor intake isn’t eliminated but much less of an issue.
Added bonus, these pumps much quieter than ExtractCraft equipment and there’s no frequent pump failure (and expensive replacement cost) like with the EtOH PRO.
Two categories of pumps can be considered; rotary vane oil pump or diaphragm oil-less pump. Both have distinct strengths and weaknesses.
Rotary vane oil pump:
Advantages – They are very fast and have a low ultimate vacuum.
Disadvantages – Ethanol vapor will eventually ruin the VP or catch it on fire.
A cold trap can be used to mitigate the amount of vapor intake but when the pump is running the vapor moves so fast eliminating the problem 100% isn’t possible.
Diaphragm oil-less pump:
Advantages – Resistant to damaging effects of ethanol vapor and simply vent it.
Disadvantages – Typically expensive, slow, and don’t have a low ultimate vacuum.
For the wet environment operation of a vacuum still a diaphragm pump rated for solvents (ethanol) works very well. Fortunately, with a suggestion from member of IchiBan’s Extraction Lounge FB group I found a powerful, little, inexpensive diaphragm pump I had not known about. It’s pretty impressive so far. It gets almost to full vac which is the first I’ve seen from such a small and inexpensive pump.
Tubing typically is included along with the pump. This tubing does not come into contact with the ethanol that will be reclaimed or reused so the material isn’t as important as with the transfer tubing. It is important to make sure the tubing is proper size to fit onto the RVL valve or prepare a connector that will fit appropriately.
OPERATE THE SYSTEM IN BALANCE
I was dreading this part for fear of having to explain a long list of great variances, complications and solutions. I’m VERY happy to say I’m still shocked at how simple it is to run this system. To be perfectly clear, it’s only simple if you pay close attention to the set up and follow the operating guidance carefully. Paying attention will result in balance and easy operation where inattention will become wrestling with the system the whole time.
Operating in Balance – General Idea
What does having the “system in balance” mean? It’s not a proper scientific term but I’m sure there’s a calculation for it out there somewhere that I wouldn’t understand. Balance is what happens when the amount of heat, vacuum, and cooling work in a condition to create a smooth and acceptable rate of producing vapor, condensing it and collecting the liquid condensate at a reasonable rate without having to run the vacuum pump hardly at all. This balanced state allows vapor to be fully contained in the closed system with nowhere to go. Contrarily, if it is out-of-balance either the production rate will be slow (maybe so slow as to be non-existent) or too fast so that the pump has to run far too much and is constantly intaking and exhausting ethanol vapor. The more the pump runs the more ethanol is wasted, vented into the surrounding air possibly making a dangerous situation as well as putting wear on the pump. We really want balance. It’s simple to achieve and when it hits, it’s like an incredibly Zen-like cruise control.
When in balance:
Heat and vacuum levels are working well together when there is an aggressive but even boil, smooth flow of vapor from the LV into the condenser, condensed ethanol in steady and easy movement from the C into the RV and the RV pressure remains unchanged with no need for vacuum level adjustment. The exit tubing from the C and the RV itself should be room temp or cooler to the touch.
When out of balance:
1) Nothing is happening.
Either heat and/or vacuum is significantly inadequate. If there’s a leak it will be easy to see by a steady decline in vacuum on the vacuum gauge. If the vacuum gauge is holding steady the problem should be fixed by adding heat, increasing vacuum (lowering pressure) or both.
The process will also go flat when uncooperative smooth surfaces like stainless steel or glass are used in the LV even when the temp and vacuum inputs would otherwise be in balance with a nice production rate.
2) System is overly active.
If the boiling action is out of control, vapor aggressively shooting through the condenser or air and vapor shooting through the tubing into the RV there is probably too much heat and/or vacuum. The exit tubing from the condenser will become warm to the touch as may the RV itself. Lowing the heat or releasing just a touch of vacuum should get it back to balance.
Understanding how much heat to apply can be a little tricky without first understanding how boiling points are effected by vacuum. The best example most people are familiar with is boiling water. Water boils at 212°F (at sea level) and is impossible to get any hotter when boiling away on the stove regardless of how much heat you throw at it. When more heat is applied the boiling becomes more vigorous, indicating the rate of evaporation increases but the temperature won’t actually rise. The extra heat is absorbed and exhausted through the phase change to gas and dissipated away with evaporation. The evaporation dissipating the extra heat also keeps the pot from getting too hot until the water is all gone and then the pot is able to absorb so much heat it may eventually melt on the stove.
Keeping the water idea in mind, take a look at this photo. Even though I’m pumping 224°C (435°F) into this beaker of ethanol under vacuum the beaker itself and the ethanol is only around 100°F and I can handle the beaker easily. If the beaker was empty it would be incredibly hot.
It’s just like the water on the stove but by using vacuum I can lower the boiling point of the ethanol to around 100°F and as long as the vacuum level is held steady, the ethanol should stay at that same temperature. Apply more heat and the rate of evaporation increases. Less heat lowers the rate of evaporation. Either taking place with the ethanol at the same temperature. Below is chart from information collected by Gilles Desormeaux with ethanol boiling points at various pressure/vacuum levels.
I made the frustrating and silly mistake of trying to find heat sources that would hold temp at around 100°F to process at a target temp of 100°F. At first glance this makes sense, but when you think of boiling points, vapor pressure and heat transfer it turns out that in the reduced vacuum environment that 100°F heat input would be inadequate for a reasonable rate of production at 100°F. This is why we don’t really have to worry about having a heating source that holds around 100°F but something with a low temp of around 140°F to start.
You can see in the example where Low is registering about 135°F under normal atmosphere, but I’m working at a higher setting between Low and Medium at only around 90°F under vacuum. Even on Medium it’s 190°F at ambient atmosphere and only around 100°F under vacuum.
Vacuum is applied to control the temp of the LV contents. The boiling points table above shows that 101°F will be at 80% of full vacuum. Adjusting vacuum to this level should be pretty close to this target. At a fixed heat input more vacuum will lower the processing temp while less vacuum will raise it.
DIY systems don’t maximize lab vapor flow efficiencies from the LV and through the condenser so more vacuum than 80% is probably going to be necessary. (Especially on a speed run: more on that some other day)
This is much easier than it sounds and there’s a lot of room for error and experimentation. For making dabs around 100°F is necessary but working at 110°F-120°F can be fine for end products that will be decarbed for edibles and such.
Pressure gauges equalized at elevations above sea level read 1 in Hg lower for every 1,000 feet of elevation.
Monitoring temp during the process is a little tricky. To monitor the heat it’s best to have a thermometer in the LV as well as an IR heat reading gun for the outside. Having both allows you to cross check and confirm against each other.
Getting a thermometer in the LV is not easy….but I’ve made it easy. On the LVL vacuum break a 1/4 inch section of tubing can attach easily and hang down a little ways. A donut shaped 3/8 inch piece of tubing can go around the 1/4 inch tubing like a collar and slice the thermometer in between the two and situate to be visible through the lid.
The IR heat gun is a little tricky as well. There are a few things that are important to know when using one to monitor in-process temps. First, they don’t work on reflective surfaces. They reflect and give inaccurate reading. Mat black surfaces work the best. Second, hold it very close to where the temp reading is wanted to avoid error. Second, the contents of the LV cannot be taken through the LVL. Only the temp of the LVL will register and not the contents. Third, they aren’t always accurate so make sure to check and know the deviation.
Pacing For Condenser
The last piece of the balance puzzle is to make sure the condenser is able to keep pace with cooling the amount of vapor traveling into it. The amount of heat a condenser is able to displace depends on factors like the material, length and width of the coil/tubing, how fast the vapor is moving and the temperature in the condenser containment (heat shedding, surface area, time, temp).
All the possible combinations to put a condenser together can be overwhelming. I worked with silicone, PVC, and stainless steel tubing ranging from 16 ft to 60 ft for the coil and used both dry ice and regular ice (ice bath) for cooling. The faster you push the process or the hotter it’s running will require more cooling at a given point of time, but over longer term slower runs will require about the same overall cooling. I hope that makes sense? Basically, you can creep the processing rate (pace) with shorter coil and less cooling, it’ll just take longer. Or, you can push the pace faster with increased coil surface area and more drastic cooling.
Bonus Cheat For Maintaining Balance!
I found a very simple cheat that appears to be incredibly helpful in maintaining balance even when the vapor is moving too fast and would otherwise build pressure and require more pump activity and vapor venting. This is the cooling pan on the component list. By cooling the reclaim vessel any vapor that shoots through the condenser has nowhere to go and the RV acts as a secondary condenser. With the cooling and vapor in check, the pressure in the system remains stable and continues processing without the need for vacuum to be applied. In some cases a good fan providing some cooling can work, but in heavier cases setting the recovery vessel on ice works best.
STEPS FOR PREPERATION & OPERATION
These are the basic steps for preparation and operation of the vacuum still. With experience everyone will find tricks and preferred methods that work better for them and differ from these basic steps, but this is a good base to start with.
Put all the components in place and attach tubing.
Vacuum test the assembled, empty system and make sure it holds vacuum with the system closed and vacuum pump turned off. If it doesn’t hold vacuum the system will be troublesome to work with so stop here, find the leak, fix it and test again before moving on.
Any threaded port piece is a possible point of compromise for a vacuum leak.
Teflon tape can/should be used to avoid any problems at those points.
Load the condenser with ice or dry ice and put the RV on ice. Let set for 30-60 minutes to cool parts.
*In humid climates you may want to have the system under vacuum to purge the air during this cooling time to limit the amount of moisture condensation/contamination.
Load wash into LV with plenty of head space so boiling contents are not sucked into vapor exit on the lid. If some boiling wash does get sucked through by accident it’s not a big problem and it won’t damage anything but the reclaim won’t be clean.
Bring temp of the uncovered LV and contents to around 100°F.
*this step can be skipped an heat can be applied after vacuum has been applied as well. The advantage of using this step it that there is immediate feedback when vacuum is applied in the next steps.
Put LVL on. Make sure it is centered well and confirm the release valve is closed.
Engage the pump with the RVL valve connected to the pump open and watch the pressure gauge as the vacuum level lowers.
Watch the LV as the contents begin to boil.
When the boil gets to a reasonable rolling boil, close the valve on the RVL and turn the pump off.
Observe and make small adjustments to heat and/or vacuum to get the system into balance and the pressure remains steady when the system is closed.
This sounds like there is a bunch of tinkering to get into balance but once the boil is going and the system is closed in step 5 it should be pretty close to the target levels.
Don’t be alarmed if there’s little action coming out of the C. That’s often a good thing indicating the vapor is slowing and condensing properly in the C. There’s a lot of tubing for the vapor and condensate to travel through and it takes a lot of vapor to make a little liquid so the condenser can take a while to fill up and purge itself. As long as the temp in the LV is steady with a nice boil and the pressure in the RV stays even the ethanol will slowly creep out and into the RV.
Things to watch for to ADD heat or vacuum:
- Reduction of boiling activity.
- Low temp of LV and contents.
- No vapor flow in tubing going into the condenser.
- Losing vacuum on the RV gauge.
Things to watch for to REDUCE heat or vacuum:
- Overly aggressive boiling.
- High temp of LV and contents.
- Vapor shooting through tubing too fast.
- Vapor clouding RV.
- Exit tubing from condenser warming.
Add wash with the In-Process-Feed option.
The vacuum break on the LVL can be used as an in-proceed-feed to add more wash into the LV without breaking vacuum and interrupting process. Attach one end of a tube that attaches securely to the vacuum break port and the other into a jar with wash to be loaded. Open the vacuum break and the lower pressure in the LV will suck the wash into the LV.
- Always operate the vacuum break VERY slowly.
- Keep the tubing submerged in the liquid to avoid losing vacuum.
- Adding cold wash to the warm environment can stop the boiling action until the newly added wash has come up to heat.
The in-process-feed control ability to manage volume with LV surface area is an amazing advantage over the EtOH PRO.
Decided when to stop the process.
Stop early for collection ease
For easy collection of the finished product the process can be stopped early in a somewhat liquid state and finished with open air evaporation on a silicon mat.
Go all the way to oil
To go all the way to oil at the end you have to pay attention because things change quickly at the very end and if it goes too far the oil will be difficult to collect.
Catching it just right requires attention:
- Keep an eye on the level at the end.
- Pick up the LV and slowly and gently tilt and swirl the contents around to get a good idea of the viscosity.
- If it moves more like cooking oil than water it’s time to stop. If it’s not quite there yet set back on the heat.
- If it looks like it’s close to being finished turn the HS off, add more vacuum and swirl around a little more until you see the viscosity you like.
The ability to control the heat and vacuum with the LV in hand is an incredible control advantage that allows crafters to skip the significant stall of the EtOH PRO and avoid the very large temp overshoot of the Source Turbo at the end of the process.
Stop process by turning off the HS if you haven’t already and break vacuum from the valve on the LVL SLOWLY.
Open LV to collect concentrate and save reclaimed ethanol for reuse.
Post process if necessary and enjoy the fruits of your crafting!
COMPONENT SHOPPING LIST & DISCUSSION
The component shopping list will be available in the FB user group IchiBan’s Extraction Lounge. The shopping list will be in the files section in the group. If there are any questions or something you would like to be share please hop in the group and get involved. This is going to be fun and save people a lot of dollars.
SAFETY CAUTION – PROCEED AT YOUR OWN RISK
Safety considerations and cautions must be taken very seriously! Ethanol is highly flammable and can be very dangerous. Anyone deciding to work with these operations DOES SO AT THEIR OWN RISK. What I present here should not be considered as fully comprehensive nor exhaustive and everyone is responsible for their own safety review. The minimal safety considerations required without compromise are having very good ventilation and no ignition sources in the area and general fire prevention for highly flammable liquids must be carefully observed at every step. Very importantly, NEVER leave the system unattended or any component plugged in when not in use.
It’s incredibly important to know exactly what material is being used in tubing and confirm that it’s rated to work with ethanol at the processing temp. Failure to do so can result in leaching of possibly dangerous chemicals into the finished product. Anyone buying equipment to use for distillation has the responsibility to check and double check for themselves that every component used is safe before using.