Dual-Peltier Dew-Point Control & Curing Chamber aka DUAL-TEC-TEK

Dual-Peltier Dew-Point Control & Curing Chamber
(one thermoelectric module drives relative-humidity, the other drives bulk-temperature)

Purpose
• Hold product (e.g., cannabis flowers, specialty meats, optical coatings) at a fixed dew-point so that moisture leaves the material slowly and uniformly.
• Achieve this by splitting the usual single climate loop into two orthogonal PID loops, each powered by its own thermoelectric cooler (TEC).

Hardware Block Diagram
┌───────────────┐ RH loop Temp loop ┌───────────────┐
│ Sensor set │──┐ DHT-20 / SHT35 NTC/RTD │ µ-controller │
└───────────────┘ │ └──────┬─────────┘
│ PID-A (RH) PID-B (T)
┌───────┐ ▼ ▲ │
│ TEC-1 │ ← PWM driver 1 Air fan │ PWM drv 2 → │ TEC-2 │
└───────┘ (condensing plate) │ (radiant cold/heat sink) └───────┘
condensate tray │ ↑
└──────── Chamber fan ──────────┘

Control Philosophy
• Loop-A (RH) treats the chamber’s dew-point as the set-point.
TEC-1 chills a small aluminum plate; when its surface less than chamber dew-point, water condenses and drips to a tray → lowers RH.
PWM duty is modulated so the dew-point error approaches zero without overshooting below 40 % RH (to prevent overdrying).
• Loop-B (Temperature) keeps the bulk air temperature at the recipe (e.g., 18 °C for cold cure).
TEC-2 operates bi-directionally: forward current for cooling, reverse for heating (or supplements with resistive film).
Dead-band ±0.3 °C to avoid constant polarity flips.
Sensor Suite
• Combined RH/T probe at mid-height (±1 % RH, ±0.1 °C).
• 2 × surface thermistors glued to each Peltier cold plate for closed-loop protection (cut power 70 °C).
• Optical drop counter (opto-interrupter) in condensate drain as a sanity check: if duty-cycle high but no drips for 10 min → alarm (ice blockage).

Firmware Algorithm (simplified)
loop {
read T_air, RH_air → compute DP_air (dew-point)
error_RH = DP_air – DP_set
duty_1 = PID_A(error_RH) // drives TEC-1 PWM 0-100 %
error_T = T_air – T_set
duty_2 = PID_B(error_T) * sign(error_T) // ± value gives direction
apply duty_1, duty_2
house-keeping: thermal-cutout, condensate watchdog, OLED display
delay 1 s
}

Mechanical Notes
• TEC-1 cold plate faces open air; hot side coupled to external heat-sink/fan that vents outside chamber.
• TEC-2 assembly is larger, mounted in the air stream of the mixing fan so it “owns” the chamber’s sensible heat but contributes minimal latent removal.
• Insulate both cold blocks to stop ghost condensation elsewhere.
• All penetrations sealed

Advantages vs. single-loop
• Decoupled latent and sensible loads prevent temperature “see-saw” common in fridge-dehumidifier hybrids.
• Finer resolution: ±0.5 % RH, ±0.2 °C.
• TECs give silent, vibration-free operation-critical for terpene preservation.

Limitations / Safeguards
• Ambient 28 °C or less than 40 % RH reduces condensing efficiency; add pre-cool loop or modest humidifier.
• Ice buildup on TEC-1 below ~8 °C plate temp; firmware caps cold-plate delta-T to 12 K.
• Peltiers age; include 10 k cycle MTBF in maintenance plan.

Typical Curing Recipes
Found on literature released from companies with this device or other users - there is not a perfect formula and it appears to be based on cultivar. In general, holding 58%RH and right below 70°F has worked for me.

This dual-loop TEC arrangement yields tight, independent control of water activity and temperature-ideal for precision curing or any process where the dew-point dictates final quality.

I firmly believe this tek is the future for drying and curing. Currently it is hard to access due to price. Every company it seems is scrambling to build one. Know why this works, and where it doesn’t. For example, some units soon to hit the market use compressor-driven cycles (the entire problem that made this tek needed…). Do your research!

4 Likes

On paper this is awesome! I could probably find ways to make it less expensive but would also widen you range outside of tolerance. The design though is solid and theoretically work super well.

Unless you are showing me something that exists already and is just too prohibitively expensive to mass produce.

Either way thank you for sharing!

@C_in_CT you may find this interesting too!

4 Likes

Thanks! I wanted to give back to the community. I have an image I can dig up of my current drying chamber built this way, kind of a sloppy version with ACI controllers and taken apart household devices/appliances.

The most sound implementation of this would be the Cannatrol. Lots are rushing versions to market. The team at Cannatrol have patents on this technology to an extent. I haven’t recreated anything that isn’t in the public domain, the concept is rather trivial actually… They took a first principles approach and said, “what if everything we know about drying and curing is false?” The current research seems to substantiate their claims…namely, regarding trichome integrity. As many familiar with the anatomy of cannabis and trichomes would know, trichome integrity is the final filter to pass for quality flower. Ruptured trichomes mean exposed (to the environment) cannabinoids, terpenes, and flavenoids. This is where the majority of hours are spent in flower - repairing and building out trichomes essentially.

Their research in a sentence is basically this - “Compressor cycles like AC systems and dehumidifiers wreak havoc on the trichome level by creating gradients/deltas with respect to bound//unbound water, effectively rupturing cuticles which are only 2 molecules thick in the first place for all intents and purposes.”

Even if that is all nonsense, the theory of dew point monitoring as a strict ‘never-going-to-happen’ mold mitigation strategy is worth exploring.

4 Likes

Just noting that in screenshot showing ‘Cannatrol’ is merely my chamber recreating their algorithms from publically posted user manuals.

6 Likes

I kinda did this as a proof of concept. I haven’t actually dried in it. I took a broken dual zone fridge and hooked it to a ~$20 controller that does temp and humidity and it handles it. Was considering adding an additional fan, but haven’t run it in real conditions, so wasn’t sure it was necessary.

At some point I will harvest again and I will try it out.

3 Likes

I agree completely. Hasn’t been on my radar yet but it is now.

1 Like