JBJ Arctica 1/10 hp Chiller Modification for External Control


Premium Member

JBJ Arctica Chillers have a +/- 1 degree Fahrenheit hysteresis. That means that if the chiller is set to 77 degrees, it will start chilling at 78 and stp chilling at 76. That is a 2 degree swing. In the case of my tank, even the 1/10 horsepower chiller brings the temperature down in about 30 mins. I do not like the large 2 degree swings throughout the day.

If the chiller is plugged into an external controller, then it fully powers on/off with each cycle. This means that each time chilling is called for, the chiller has to turn on, wait 5 mins (built in timer to prevent short cycling) and then chill. It also means that the internal controller must be set low enough to be always on, so it can't really act as a fail-safe. Lastly, the display (obviously) is dead when the chiller has no power.

My solution is to bypass the internal relay, and substitute a relay controlled by my Profilux controller where I can set hysteresis to as low as +/- .135 degrees.

Here is the published schematic for the JBJ 1/10

Simple enough - Cut the "white" wire and put a relay (controlled by the Profilux) between it and the SKY wire. Might as well add a DPTD switch to be able to switch between "internal" and "external" control too...


Pretty straight forward right? In one position the switch takes the internal controller's output OUT of the circuit and in the other position it takes the NEW relay out of the circuit and restore the connection to the internal control. Notice on BOTH position, the internal display of the chiller is still powered!

There are other ways that this could be done... allowing both to operate at the same time (below) or have them in series, etc. The problem is that operational modes would be confusing. In the schematic below the control would be "external only" or "both" depending on switch position. I opted against his (again it os confusing).


In most devices, best practice is to always switch the "hot" leg and NOT the "neutral" leg. In fact, the schematic above, shows the hot leg (SKY) traveling to the "controller" and being returned (WHITE) as switched to the compressor and fan. It shows the NEUTRAL wire (BROWN) being tied directly to the compressor and fan, as well it shows the FUSE on the line/hot (SKY) wire (where it should be). The reasoning is simple. When the device is OFF or the fuse is blown, one would expect no live (hot) power to the components that are not ON.

Upon disassembling the chiller to install my modifications, I quickly realized that the FUSE was on the BROWN (NEUTRAL) wire and that the SKY (LINE/HOT) wire was tied directly to the compressor and fan and then fed to the 4 pin plug that powers the display. THEREFORE, the "WHITE" wire coming from the controller and display module that is the "switched" line is actually NEUTRAL. That means that even with the compressor and fan are not running, or in the case of a blown fuse) that live 120V power is present throughout the device. This is insanity... but without rewiring the entire device and modifying the on-board controller, can't be "fixed". I will reach out to JBJ at some point, but would venture to say that they don't know or care, as these are not likely "designed" in-house.

The actual schematic and my modifications are show below.


Left factory connection shows SKY coming from the power cord, going out to the 4 pin connector AND connecting to the compressor (brown) and fan (black).

Right factory connection shows WHITE coming from the 4 pin connector and going to the compressor (white) and fan (black).

The is exactly opposite (left connector) of what the factory schematic shows!

Not show (brown from power inlet) goes directly to the 4 pin connector. The left connector was left in-tact and the right connector is the one that I cut, placing the switch in between.

The photo below shows the relay installed. I used an existing screw on the condenser case.

Waterproof ring terminals crimped (below) before heat shrinking. These will connect to the terminals of the DPDT switch. A note here, you MUST use a ratcheting or commercial grade crimper. Using a pair of pliers or cheap "crimper" never results in a reliable termination.

120V 15A DPDT toggle switch show n with one terminal connected.
Last edited:
I saw the title of the thread and thought’ “what is Bean up to now?” 😀
Anyway: (ignore the PSU - it is ~35, but the probe is dirty and reading low.

Before - see the chiller spikes (7 days) so there were 2 or 3 sharp 2 degree swings per day, sometimes 3 with they system fairly steady during nightime and spiking a few times during the photoperiod as the chillers fights the lights.

After - nice steady temperature. that varies +/- .2 degrees for the last 24 hours.
I don’t any reason to regulate tempatures so closely. It will increase the wear and tear on the compressor do to a lot more on/off cycles. I suspect your electric bill will also increase.
Tempatures can change much more than 2 degrees very quickly in the ocean. Especially in shallow areas. It is obvious while snorkling even in deeper water.
I think this will be great for all the JBJ chiller owners where the controller has failed. This is the most common failure with them from what I have seen.
I don’t any reason to regulate tempatures so closely.
I don’t mind a 1 or 2 degree swing over 24 hours. I did not run a chiller for 20 years, and for a long time no heater either. During the summer, the AC keeps things fairly stable during the day and in the winter the basement is cold enough to buffer things. It is spring and fall where the AC is not running but the basement is warmer and temperatures in the tank stay elevated. I added a chiller to combat this but really dislike the rapid pull-down/swings that happen several times a day as a result.

It will increase the wear and tear on the compressor do to a lot more on/off cycles. I suspect your electric bill will also increase.
Yes, it will certainly be more wear on the compressor. With the current hysteresis settings, the chiller runs for 8 minutes about every 40 for during the photo period. So maybe 10 times per 24 hour period and yes that is more electricity than allowing 2 or 3 runs per 24 hour period with the ~2 degree swing.

I can adjust the hysteresis as needed though, and will adjust moving forward to find the balance that I am looking for seasonally. In the end, being hamstrung by the chiller settings with thousands of dollars of automation equipment sitting next to it was frustrating. So I did what I always end up doing... going down a rabbit hole to find a solution.

Tempatures can change much more than 2 degrees very quickly in the ocean. Especially in shallow areas. It is obvious while snorkling even in deeper water.
I don't disagree at all and would lean toward temperature changes being more natural and likely creating a hardier captive set of organisms. I have made the very point that you just made to countless people over the years and still firmly think that temp swings are not a bad thing.

That said, I got back into SPS not long ago and am trying to eliminate as many variables as possible to see what I can achieve with stability on all fronts.
Last edited: