Thanks for the question!  This is a pretty common question with a fairly complicated answer, so check out this related question on the Cannabis ECN: https://cannabis.expertcallingnetwork.com/ask/what-are-the-safest-and-the-most-effective-ways-to-extract-and-produce-cbd-rich-cannabis-oil-co2-oil-or-ethanol

Thanks Gina for the question!

There are multiple options for extraction equipment manufacturers to choose from, but I would advise the use of caution on 2 fronts when evaluating a company to build a custom system:

Are you building a custom system because you are ready to scale?  If you have been doing extraction for a while and are ready to scale, congratulations!  However, only a relatively small number of cannabis companies are large enough for something like this.  It is always important to have a solid understanding of the parameters that you need to operate your equipment with BEFORE you try to scale.  Large scale equipment is very expensive, difficult to modify and generally lacks flexibility.  Make sure you have plenty of experience with the pilot and mid-range scale commercially available equipment before building a large, custom system so you know EXACTLY what to expect! 
Make sure your equipment provider is actually the manufacturer!  Many extraction equipment companies in the cannabis industry today are just assemblers – they order parts from other companies and just piece them together.   While this approach can work, it increases costs, lead times, and makes quality control much more difficult.
 

Ready to go?  Check out Apeks Supercritical – we can build just about any extraction system you need in our ASME and UL certified manufacturing facility!

Also, check out this related post for qualities to look for in an equipment manufacturer (hint: lots of experience, legitimate certifications, long warranty, bonified claims, etc) : Claims by Manufacturers

Apeks Supercritical developed the 1500-1L in 2013, primarily to provide caregivers and small collective groups the ability to control the extracts they were providing their patients. Back then, extracted oils were still a fairly new concept, and there were a lot of extracts on the market that had residual solvents, heavy metals, toxins, pesticides, etc. Caregivers wanted an affordable way to provide clean medicines to their patients, and a way that didn’t involve hydrocarbons because of the explosion hazards. 

Between 2013 and 2016, we sold about 125 of the 1500-1L units into 20+ states. Many of our customers have taken advantage of our 75% return credit program and were able to “trade up” their 1500-1L systems for one of our higher throughput Mid-Range or High Production systems, allowing them to increase their ability to extract more efficiently and also grow their businesses!

The cannabis industry has grown significantly since 2013 and has also experienced significant price depression which is forcing extraction companies to be bigger and more efficient in order to survive. Also, most new medical marijuana states are following the “small number of large operator” structures. As a result, smaller throughput systems like the 1500-1L just aren’t big enough to be competitive. As a result, we decided to discontinue the 1500-1L in 2017 as we continue to focus our R&D and production efforts on the higher throughput systems we have developed over the past 3-4 years. 

Regarding the post from 2 years ago: We strive for 100% customer satisfaction. Unfortunately we missed the mark on that customer. It isn’t representative of the satisfaction level of the other 125 customers.

Thanks for the question - I'm glad to help clarify!

I recently saw an article from an equipment manufacturer with a very generic statement about how back pressure regulators, valveless expansion systems and other types of pressure controls that are tied into an automation system are particularly prone to failure. Unfortunately, this type of misleading and irresponsible generalization is all too common in the cannabis industry today.

As with most equipment, generalized comments about components or portions of a piece equipment are not accurate or representative.  You must get into the details before you can make an assessment about the component's ability to perform in service.  A common example is a comment like "stuff made in China is junk".  A general statement like this is simply not true - some cheaply made items from China will not last very long, however there are many high quality components manufactured in China that will perform as good or better than their US counterparts.  The devil is in the details! 

There are thousands of components that can be utilized in an extraction system, but since your question focuses on valves let's dig into the details on those:

• Pressure regulators and needle valves - two types of pressure regulation valves are commonly found on extraction systems: Back Pressure Regulators and Pressure regulators. BPR's are most commonly used as relief valves, but are also used to maintain a constant back pressure in extraction vessels and also in cascading separators.  Pressure regulators are commonly used in control air systems, but not commonly found in the CO2 portion of an extraction system.  Regardless of the type or quality of regulator, any regulator used to control decompression in an extraction application will have the same problem - they will clog due to the extracted oil precipitating out in the valve internals. This issue isn't restricted to pressure regulators, as needle valves, throttle valves, and any other kind of valve will always have the same problem. It is important to note that the quality of the valve has not bearing on this issue. It is simply the wrong tool for the job.

Automated Valveless Expansion systems inherently do not clog because there are no valve internals to be affected. Automated Valveless Expansion provides a stable pressure platform that can adjust to varying environmental conditions such as line voltage, ambient temperature, heat transfer changes due to vessels filling up with extract, etc.  Manually operated systems, especially those with manually operated needle valves, will not automatically compensate for these changes which can cause flooded separators or allow the liquid pump to run dry – both requiring significant downtime and maintenance costs. 

• Solenoid valves - Solenoid valves were invented in 1910 and are widely used in industrial applications, including highly critical applications such as nuclear power plants, oil drilling platforms and aerospace applications. As you might expect, there are cheap solenoid valves and there are high quality solenoid valves. Manufacturers of automated platforms with experience will have a track record of success with a particular solenoid valve. First time automation won’t have the experience and can suffer failures.  
• Pneumatic valves – air operated valves are very common on CO2 extraction system.  It is a common misperception that air operated valves will fail faster than manually operated valves.   Regardless of the operation mechanism the construction of the valve is exactly the same!

A final thought on automation "failures": It is common for operators to identify any kind of notification as an "error" or a system "failure". Notification signals are part of the automation safeguards to notify the operator when the system experienced a condition outside of the normal operating parameters, not just a failure. A great example is notification that the CO2 supply bottles are empty – it isn’t a failure of the system, however the system will not operate properly unless the CO2 bottles are changed. The benefit of the automated notifications is that the operator doesn’t have to visually monitor the system or even be in the same building – they can be notified remotely if there is an issue that needs to be resolved!  

Dial face gages are also known as analog pressure gages.  They use a mechanical gear/needle and a bourdon tube to indicate pressure. 

The most important reason for a dial face gauge on an automated system is for safety – this allows the operator to determine if there is pressure in a vessel or a portion of the system without the need for power – i.e. during a power outage. 

The second reason for a dial face gauge on an automated system is for redundancy - this allows the operator to compare the analog readings on the gage to the digital readings on the automation system.  Both mechanical and electrical instruments can fail.  Having redundant indications allows the operator to identify an issue and quickly correct it.  Without redundancy there is no way for an operator to know there is an issue and thus can operate with incorrect readings!

A note on accuracy - automation platforms with digital indications will always be able to display a significantly higher resolution and responsiveness than a typical analog dial face gage found on CO2 systems.  There are high accuracy pressure gages that are typically used for calibration, but are not typically used on CO2 systems.

I also addressed a similar question in a separate post that can be found here: https://cannabis.expertcallingnetwork.com/ask/i-understand-that-fully-automated-machine-sensors-can-get-gummed-up-with-oil-causing-machines-to-fail-or-for-readings-to-be

Great question!  There is a lot of misinformation being spread around these days so I appreciate the opportunity to clear this one up!

These kinds of issues are indeed possible due to the high viscosity of cannabis oils and highlight the reason to seek out an automation platform that has a track record of success.   Manufacturers who have never automated a system or have very little automation experience may not be able to identify these issues and will encounter these problems frequently.   Manufacturers who have been through several iterations of the software and automation will have identified these types of issues over time and dealt with them through proper probe design (i.e. use of snubbers and diaphragms), proper component placement (i.e. not in a place where oil or resin can accumulate), and proper cleaning/maintenance procedures.

A common misperception is that a manual system is not susceptible to these types of issues - this is not true.  Pressure gages and temperature probes can be affected in a manual system the same way pressure transducers and temperature sensors can in an automated system.  As described above, proper placement can help avoid these issues, but this is where redundancy becomes very important.  Automated systems will have backup analog or mechanical instruments for safety (i.e. to verify a vessel is depressurized without electrical power) as well as to double check that an instrument reading is correct.  When there is only 1 level of instrumentation, such as in a manual system, there is no way to tell if an instrument is clogged or has failed!

Thanks for the question! I believe the video that you were watching is here: https://www.youtube.com/watch?v=Nt5xlJSo5Hc

Around 6:00 in the video we are discussing our welding and fabrication procedure for our vessels - full penetration welds on stainless steel components that are subsequently ground and polished on the interior to produce a smooth, continuous surface. This process is a TON of work – using component designs such that the interior surfaces are the same diameter, welding designs for full penetration, and then grinding and polishing the weld takes a significant effort. All of this effort drives costs up.

In an industry that lacks regulations for food safety, the temptation to cut corners and costs is strong. The FDA is a federal agency and thus does not acknowledge the cannabis industry. However, that doesn't relieve the manufacturer of the responsibility to produce equipment that complies with known regulations that will eventually be applicable.  

Equipment that is being utilized for food product should comply with The Code of Federal Regulations (CFR) Current Good Manufacturing Practice (GMP) for the Manufacturing, Packaging, or Holding Human Food. This code details the requirements for manufacturing, preparing and holding of food to prevent its becoming adulterated and unfit for human Consumption. Specifically, CFR 21 - §110, Subpart C - Equipment, §110 .40 Equipment and Utensils states seams on food contact surfaces must be smoothly bonded or maintained so as to minimize accumulation of food particles, dirt, and organic matter and thus minimize the opportunity for growth of microorganisms.

CFR 21 also states food processing equipment must be stainless steel, or at least have stainless steel parts that are in contact with food product. There are multiple grades of stainless steel - 304 and 316 are most commonly used. 

Our extraction equipment is designed for the future. There will be a time when GMP will be required in cannabis operations - our equipment is manufactured with this in mind: stainless steel vessels with polished interiors that are accessible from both ends for cleaning.  

Below is an example of what to look for. On the left is a properly fabricated vessel, with no trace of the weld or surface imperfections from the pipe mill. On the right the vessel has mismatched interior diameters, an unground weld and a rough surface from leaving the mill scale on the pipe as received.  

Thanks for the great question!  There are really 2 questions here, so I'll try to answer them separately.

First question: What are the safest ways to extract? When it comes to extraction, safety is an important issue and has multiple areas to consider. The list below represents some of the major areas that need to be addressed with the popular solvents being used in the cannabis industry today:

• Design - The extraction equipment needs to be designed to handle the solvent being used  
  • Materials of Construction- Stainless steel materials for food/consumed oil applications
  • Electrical for Flammable Solvents - Class 1, Division 1 (explosion proof) electrical components for compressed flammable gasses, Class 1, Division 2 for ethanol/alcohol  
  • Electrical, Non Flammable Solvents - NEMA 4x wash down electrical enclosures
  • Pressure Rating - usually 300 psi for hydrocarbons, 2000 or 5000psi for CO2. 
  • Overpressure Protection - non-isolable relief valves set to 110% of maximum allowable working pressure
• Purpose - The equipment needs to be constructed for its intended use
  • Food grade - welds in contact with extracted material should be ground flush and polished
  • Accessibility for Cleaning - vessels and piping should be accessible from both ends to allow proper cleaning
  • Storage tanks - should be stainless steel to prevent corrosion
• Facility - In addition to the equipment considerations, the facility must also be appropriate for the extraction solvent
  • Compressed Flammable Gasses - Class 1, Division 1 facility. This includes electrical fixtures, and also monitoring and evacuation equipment in the event there is a release of flammable gas into the area around the equipment.
  • Ethanol/Alcohol - vent hood or equivalent walk in vent area
  • CO2 - asphyxiation hazard. Monitoring and audible alarm to warn of leaks.
• Human Consumption - The solvent should be safe for consumption by humans
  • CO2 - Generally Regarded as Safe (GRAS) by the FDA for consumption
  • Compressed gas - GRAS for use as a propellant, states differ on safe residual solvent levels
  • Ethanol - GRAS for food products, states differ on safe residual solvent levels

So the answer to the question about safety really doesn't have anything to do with the solvent, rather the equipment chosen and the facility where the extraction is performed determine safety. The solvents commonly used in cannabis extractions today all have pros and cons, and all can be operated safely as long as proper guidelines and regulations are followed.

I addressed the safety question in the first part of the answer, in the second part I'll address the efficiency question: What is the most effective way to extract CBD-rich cannabis oil?

A major problem facing the cannabis industry today is a lack of commonly accepted standards - as evidenced by the question referring to "CBD-rich". Does "CBD-rich" mean 40% CBD?  99% CBD? And CBD in what form, CBD, CBD-A or some combination? There are groups that are working towards creating standards, such as FOCUS and ASTM, but they have not been widely accepted yet. Without standards, quality becomes difficult to determine because the only standard is personal subjectivity. 

That being said, there are some generalizations about extraction methods that can be made. Keep in mind - every extraction method has pros and cons. Each method will shine in certain applications, and perform poorly in other. No method is great at everything.

• CO2 Pros
  • Selective and tunable for different molecular weights
  • No residual solvents - great for vape pens
  • Cold extractions and separations are good for temperature sensitive extractions (i.e. terpene preservation)
  • Minimal facility safety requirements/costs - just signs and CO2 monitor
  • Automation is easy and available
• CO2 Cons
  • Equipment is expensive 
  • Manually operated systems are complex to learn
  • Extraction rates are slow on less expensive equipment (competitive rates for more expensive equipment)
• Hydrocarbon Pros
  • Less expensive equipment than CO2
  • Fast extraction rates
  • Great for dabbable products like shatter, honeycomb, crumble
  • High potency levels
• Hydrocarbon Cons
  • Facility is very expensive - Class 1, Division 1 requirements
  • Automation is difficult and expensive due to C1D1 requirement
  • Heavily scrutinized by local regulators and inspectors
  • Not selective or tunable
  • Scaling challenges from limits on amount of hydrocarbon that can be on site (150#)
  • Residual solvent testing required (adds additional time and expense)
• Ethanol Pros
  • Very inexpensive equipment
  • Facility is less expensive than hydrocarbons
  • Less power intensive than CO2
  • Very fast extractions - great for distillate products
  • Easily scalable
• Ethanol Cons
  • Not selective or tunable
  • Will freely extract chlorophyll
  • Requires significant levels of secondary processing
  • Flammable - requires fume hood or equivalent

Thanks for the question - an unfortunate reality in the cannabis industry today. I have said for a long time that the cannabis industry is an entrepreneur's dream. Explosive growth combined with just enough risk to keep the big players out has allowed many Ma & Pop companies to flourish. It has also made for a very competitive space with new entrants almost daily. Choosing a vendor can be tough in this environment, but there are definitely some criteria to look for to help choose the best equipment vendor.

Despite the number of new entrants in the cannabis space, the same rules for purchasing equipment in more established industries apply - i.e. oil & gas, aerospace, automotive, etc. Look for the following criteria, and don't just take the vendor's word. Make them back up their claims with data! One of my engineering professors in college always said "In God we trust...all others bring data!".

A short list of items to help vet an equipment vendor in the cannabis industry:

• Look for experience, both in the cannabis industry and out. Look for an equipment supplier that started outside of the cannabis industry and then entered with an already proven product. The cannabis industry is new, but most of the equipment that is being used is not! Every new company has to spend money and time developing their product. Don't let them use your money as an interest free loan for their R&D project unless you are getting something in return for it (discounted price, future tech, etc.)
• Ask for certifications, and make sure they are real. Is the equipment UL/ETS certified? Does the equipment comply with manufacturing and building codes like ASME or NFPA?  Do your homework and follow up on the claim. Most equipment certifications will have a listing on the corresponding site, like ASME.org, UL.com. There are a couple of industry specific certifications as well, such as Engineering Peer Reviews.  
• A long warranty is a good thing. Carrying the liability of a multi-year warranty over hundreds of pieces of equipment can be a major balance sheet burden to an equipment manufacturer. Only a well-established (and well financed) company can carry such a burden, which means stability for their customers.
• How many systems have you sold? This question should be asked of every equipment vendor. While the cannabis industry is new, it isn't that new. Established vendors should have hundreds of customers, even better several hundred. 
• Outrageous claims? RUN! Generic claims with no data to support them are a major red flag. A quick Google search on "CO2 extractor" provides some wonderful examples of just a few of the outrageous claims to sort through. They can't all be the fastest, but it doesn't seem to stop them from claiming it!
  • "The fastest and highest-yielding extraction and distillation systems on earth"
  • "The highest flowing CO2 extraction system on the market today"
  • "Our machine is the industry's fastest, producing extracts at record speed"
  • "The only SFE system on the market to incorporate three collection vessels with independent pressure control"
  • "Highest Yield On Market"
  • "Our equipment is the fastest and most energy effiecient CO2 extraction equipment available anywhere"

First - all mechanical systems will eventually break down. Automation is an additional level of equipment complexity and thus does increase the possibility for breakdown because there are more components, but to say that just because automation has been utilized that breakdown is inevitable is an overstatement. The corollary would be that manual systems will never break down which we know is not true. Why would manufacturer of manually operated systems offer parts, warranties and service programs if they never break down?

So if automation increases the possibility for break downs, why do it? The simple answer - automation never sleeps. Automated systems remove the variable of the human element, allowing more stability, consistency, safety and reliability. Automated systems don't take breaks, don't fall asleep, don't find better jobs, and don't hold knowledge hostage. Automation will never bother your HR department and will never have a bad attitude.

Here at Apeks, we get the question "I'm and extraction artist, and I need flexibility.  Can your automated system give me the flexibility I need to create my artisan product?". It's important to point out that while not all extraction systems are alike, so too is automation.  Automation can be done well, and it can also be done poorly. This begs the next question: "How do I know if automation is good?" Properly automated systems will not restrict the operating capacity of the equipment. It will allow the operator the same freedoms and flexibility of a manually operated system while not exceeding the safety margins of the equipment. This allows an operator who is less experienced with the equipment the ability to safely experiment without fear of damage. Our response to the Extraction Artist - not only do you have the flexibility, you also have safety and consistency.  In addition - once testing is over and it is time for production, automation allows consistent production without the operator having to monitor the equipment - freeing up time to create other products!

Breakdowns are inevitable, but fortunately the cannabis industry is not the first to rely on automated equipment.  We can learn from our friends in automotive, aerospace, ship building, oil and gas, pharmaceutical and other manufacturing industries.  Just like the rest of the manufacturing world, we recommend all of our customers implement a strong preventative maintenance program to mitigate unscheduled down time, and also recommend carrying a supply of consumable components.   These are proven techniques that are utilized frequently throughout all manufacturing industries.

One final thought on Apeks Supercritical automation - if the equipment does break down unexpectedly we have the ability to look back at the operating parameters to troubleshoot what happened.  Much like patients at a doctor’s office, operators of manual and automated systems report symptoms, and often will incorrectly self-diagnose problems.  The ability to download operating history, alarm conditions and run data gives us the advantage of quickly diagnosing the real issue and getting the equipment back up and running.  Without the data, we would have to rely on the operator to recall exactly what happened...relying on humans is never a good idea!