Eric Paulin: The HVAC Presentation

My name is Eric Paulin, co-founder of Cultivation Warehouse.  For over 20 years we have been suppliers of all types of horticultural equipment.  From lighting, to controls, racking, irrigation, greenhouses, fertilizers, substrates, and just about anything related to cultivating a plant.

I’ve had the benefit of being involved in the design & construction of several hundred commercial cannabis facilities in nearly every active US state.  Ranging from craft grows that are several thousand sq ft, to greenhouses larger than a million sq ft.
While I love to geek-out on all types of cultivation technologies, today I’m going to focus on the biggest CapEx item you will have when building an indoor grow: HVAC.  Specifically I’d like to address:

1. How to save as much money as possible, while still selecting a system that performs well.
2. How to avoid the common pitfalls you may encounter when trying to determine which system is right for your facility.

Before we dig into it, I’d like to review some HVAC fundamentals

When sizing an HVAC system, there are two types of heat loads that must be considered; sensible and latent.

• Sensible heat is what we traditionally think of when we imagine heat, and it’s a straightforward thing for your engineers to calculate. Take total watts of all equipment in each room, and multiply by 3.412 BTUs per watt.  12000 BTUs per ton.  In a cultivation room the vast majority of your sensible load comes from your lighting.

• Latent Heat – While this isn’t the technical definition, for our application latent heat is the heat that results from an increase or decrease in the amount of moisture held by the air. When water changes phase from a liquid to a gas, it retains latent heat as stored energy. When we cool air to reach the dew point, we are converting the water from a gas back to a liquid and effectively removing the latent heat.   Plant transpiration is going to be almost 100% of the water vapor in the air of our cultivation spaces.  Your engineers know both of these facts; what latent load is, and that plants are producing it.  What they may not know is how to estimate cannabis transpiration rates.

In my experience it’s really common to see the MEP team just turned loose, without much guidance on how the system should be specified.  They’re engineers!  They should be able to figure it out.  Well, to an extent.  They can determine:

• Sensible load
• Your building’s spatial & structure constraints
• As well as your electrical & utility constraints
• And of course your geographic location and worst-case outside environmental conditions

But your engineers will need input on other criteria that they may not have insight on.
Your cultivation team should provide:

• Ideal daytime and nighttime environmental setpoints (temp & RH%)
• Their irrigation and transpiration estimates to help determine latent load

Ownership or executive team should provide:

• Budget range
• Acceptable ROI if considering an investment into a higher priced system with lower OpEx

And 3rd party rebate consultants can provide:

• Local energy code requirements
• Ability to qualify for utility rebates

The most commonly used systems in the field are either Package DX or Water Chilled.

• Package DX are refrigerant-based systems that are quite literally, all in a package. Meaning all of the normal HVAC components are housed in a single outside unit, rather than separating the condenser coils and placing them outside, and the cooling coils and placing them inside.  Sometimes these are called RTUs (or roof top units), though they don’t have to be placed on the roof.  Package units have the simplest (and most economical) installation of all the commercial HVAC options.

• Water-chilled HVAC is a broad term used to describe many different types of systems, but similarly they all pump water through a piped distribution system to each of your cultivation rooms. There are many options of chillers utilizing different types of compressors; rotary screw, centrifugal, scroll, etc.  There are two piped systems, and four piped systems.  Some integrate boilers, some do not.  We don’t have the time today to get into all of the differences, just know that generally speaking water-chilled is more energy efficient than most other types of systems, and they can maintain tighter environmental setpoints compared to most other HVAC options.  But the material costs tend to be a bit higher.  And the installation and controls costs can be much much higher than other HVAC systems.  Unless you are planning on building a very large facility and have access to a large pool of capital for your build-out, these can be cost-prohibitive and may not generate a return on investment very quickly.  But if you are planning on building a fairly-large facility (say at least 50k sq ft or greater), and the project is well-funded, water-chilled may be the best choice for your project.

Some other types of HVAC you may commonly see in the field are: residential, split, or VRF.  We won’t get into the specifics of each type of system today, but just know that these types of units were developed for residential applications.  Which means that they aren’t sized to handle the heavy loads we should expect from commercial cannabis cultivation.  Nor have they been designed to remove large amounts of moisture.  Some people try to make these work by using many units per room, sometimes dozens!  And because they aren’t designed to handle large latent loads, people have to supplement with additional dehumidifiers to keep the humidity levels down.  On paper these types of systems seem economical, but the hidden install costs can be quite high when you have to install so many units.  And they are not great from an energy-efficiency perspective, or at maintaining tight environmental controls.
There are several other niche technologies that can work really well under certain applications.  Some of these include; liquid desiccant systems, absorption chillers, or package units with integrated dry desiccant enthalpy wheels (that’s a mouthful…).  I won’t get into the specifics of these today, but just know that there are certain situations where they can be the best choice and can demonstrate a quick return on investment. 
Now that we’ve reviewed the types of heat loads and the most common HVAC options, let’s get into how to save time & money.

Plan for a pre-engineering review of the types of options and relative costs.  Is your team planning for a water-chilled system, package DX, split/VRF, etc?  Why?  What are the pros and cons of each, and approximate relative costs.  Not just material costs, but also installation estimates and operational costs such as energy usage and maintenance.  It’s possible to change the brand post-engineering, but changing the type of system would likely require the entire thing to be reengineered.  They may get defensive that these things can’t be determined or even estimated until after they finalize engineering.  But assure them that you are only looking for very-rough estimates and comparisons between the options.  Anyone who has extensive experience with each of the different types of systems should be able to provide ballpark estimates.

Well of course the answer is “both!”.  But sometimes we can’t have our cake and eat it too.  Energy efficiency unfortunately comes at a price, so it’s important to consider the following:
Is this a short-term or long-term investment?  Are you trying to build a production facility and flip it in two to three years?  Or are you trying to build a brand and have a growth strategy of 5 or more years?
This is the first step to identifying your budget range, and acceptable ROI.  There will be many options of equipment, brands and technologies that will all “work”, but will have varying levels of energy efficiencies.  If you plan on a quick-exit, you may be in the market for something that is cheaper on the front-end even if it doesn’t have a fantastic operational efficiency.  Or on the other hand, if one option costs an additional $600k compared to another option, but reduces monthly OpEx by $10k, does your business plan, budget, & financial model support a 5 year ROI?

There are many reasons why your engineering firm may just pick one horse and run with it:

• It’s less time and work for them.
• There could be financial incentives (kick-back from certain manufacturers).
• The manufacturer may provide support and engineering to your MEP team, reducing their workload.
• They may just choose a cannabis-specific HVAC company/solution because of their reputation, not realizing other “commodity” HVAC brands have gotten up to speed and can offer similar solutions for pennies on the dollar.

Before engineering begins, have a discussion about what other companies can provide a similar solution.  Review the differences in; cost, lead time, capabilities, controls, & support. Does one option work better than another in your geographic area and worst-case climate conditions?  Is there a strong network of technicians that support that product & brand in your area? What about maintenance on the equipment? How much will that cost?

I can’t count how many times I’ve heard “We’ll just use residential units and supplement with dehumidifiers, it’s cheaper than sizing the HVAC to handle the dehumidification”.   While it may be a tad more economical on the equipment side (but usually not by much), the scope of installation just increased significantly.  Lots more refrigerant lines, lots of additional equipment to run power and condensate lines.  And more units to service/maintain.  When you look at the combined cost between materials and install, it very well could be higher than moving forward with a “costlier” system that can handle the load without supplementing with dehumidifiers. Hence the reason for commercial sized systems! It’s about the right product for the right application.

The lower you go with your temperature and humidity setpoints, your tonnage will increase exponentially.  For example, even though 45% RH is only 5 percentage points lower than 50% RH, it could require 50 to 100% greater tonnage to continue operating at the same capacity while running at lower setpoints.  From experience I’ve seen growers do this quite often.  “I may not need to ever go that low, but I want the ability to do so just-in-case”.  That is likely a very, very costly contingency plan.

Utilizing modern lighting like LEDs we’re able to gradually increase the light intensity from the start to finish of a cultivation cycle, so the power draw also gradually increases as does the sensible heat load in the space.
And we know that as the canopy develops, transpiration (or latent load) will also gradually increase from start to finish.
These load changes will also impact at what capacity your HVAC system is operating; the energy required to run your modulating compressors, the amount of dehumidification that can be achieved by hot gas reheat, and the amount of supplemental electric heat you may need to run.  While the person specifying your system may not be able to perfectly predict these changes, they should be able to roughly estimate load changes throughout the growth cycle of your plants.  Anyone who assumes a static load from start to finish will overinflate their energy usage estimates, and vastly skew your ROI projections.  Insist that whoever is spec’ing the system creates a model of how they are assuming load is changing throughout the lifecycle of the plant.  Corroborate that with your growers or consultants.  And ask for them to detail how each HVAC option’s electrical load may change on a week to week basis.

The subject of “cannabis transpiration rates” isn’t covered in the mechanical engineering curriculum, so unless your engineer has worked on several indoor cannabis projects with real-world data to back up their metrics, it’s possible they may not know.  The engineers may be relying on your Director of Cultivation to relay that data.  And unless your DoC has managed some very large grows and has an in-depth understanding of how to estimate transpiration rates throughout the day, they may not know either.  So it’s common to see transpiration placeholders such as “well let’s just assume we irrigate 1 gal per plant per day, so assume a similar transpiration rate”.  But depending on your plant size, you may only be irrigating ½ gal per day, or even ¼ gal per day.  This could cause you to oversize your HVAC system by 50 to 100%.  Not only does this mean your system will be far more costly, but an oversized HVAC system may not do as good of a job of maintaining tight setpoints (also known as the deadband) compared to a system that has been sized correctly.

Instead of considering what you may water a single plant within a day, and trying to extrapolate that out across all plants, it’s both easier and more-accurate to just consider transpiration rates by sq ft of canopy.  Whether you have 9 plants that are 1 sq ft each, or one plant that is 9 sq ft total, if both scenarios have a similar leaf density and biomass they will both uptake and transpire water at similar rates.  At the late stages of each cultivation period when we are at greatest leaf & canopy density we can safely assume the following daily peak loads per sq ft of canopy:

• Flower: 2 pints per sq ft
• Mother: 2 pints per sq ft
• Veg: 1.5 pints per sq ft

For the sake of measuring total canopy, just consider the total square footage of either benching or racking within each cultivation space.

We do not consistently irrigate our plants throughout a 24 hour period, and the plants do not consistently transpire during a 24 hour period.  The majority of transpiration happens during the lights-on period of your respective cultivation rooms (12 hours for flower, 18 hours for veg or mother).  It’s safe to assume that in flower approximately 80% of your load will happen during that 12 hour period, and veg and mother approximately 90% of your load will happen during the 18 hour period.  This is an important consideration when sizing HVAC or dehumidifiers.  HVAC systems will commonly be rated in lbs of moisture per hour, which we can easily convert to pints because a pint weighs almost exactly a pound. 

Standalone dehumidifiers are a good option for your dry rooms.  There isn’t much sensible load present, mostly just latent load from your plants releasing their water content.
Dehumidifiers typically have a number next to their name, denoting their daily capacity in pints.  Think Quest 506, or Anden 710.  These units will pull out 506 or 710 pints per day respectively.
But something is commonly overlooked about dehumidifiers, is that they operate less efficiently at lower environmental setpoints.  When you see the pints-per-day capacity of a dehumidifier, those ratings are taken at 80°F, 60% relative humidity.  It’s very common to run dry rooms quite a bit cooler than that, and drier than that.  At 65°F and 50% relative humidity your dehumidifier will operate at less than half of their normal capacity!
This isn’t a suggestion to run your dry rooms warmer, or wetter.  65°F @ 50% RH is an appropriate temp and relative humidity setpoint for a dry room.  It’s a suggestion to explore the capacity of your preferred dehumidifier at whatever setpoints you intend to run in your rooms.  Well your next question may be “how do I estimate my moisture removal requirements in my dry rooms?  How much moisture will my plants emit?”  If you reach out to me directly I’d be glad to explore with you further.

The biggest take-away I hope you have from this, is that communication is key.  No one should expect their HVAC system to be engineered correctly for their application unless:

• They communicate how it should be engineered, and at what cost
• They demand communication from their engineers and staff on how they are specifying the system, and qualifying the options

If I touched on some points that you’d like to investigate further, please feel free to reach out to me directly!  Additionally there were a lot of topics that I didn’t address here that I would love to review with you further, such as:

• Why is redundancy important, and how much redundancy do you need?
• What are the ideal day/night setpoints for different cultivation methods?
• How do engineers typically determine “worst case outside ambient conditions,” and why might their estimates not be sufficient?
• How do you find the best price on HVAC equipment?

Please feel free to reach out to me directly, I’d love to discuss with you further.
[email protected]