Softub controller

I have a Softub 300 that is about 10-12 years old. For those that haven't ever seen one, this is a hot tub that uses the heat from the motor to heat the water. I like it, it's simple and very comfortable.

Last week the digital controller died. The failure is in one of 2 circuits; one costs around $300, the other around $400, and right now I don't have the $$$. You can already tell where this is going, but you knew that already given the forum I'm posting in... I grabbed a spare Uno, a proto shield, and an LCD shield. I picked up a 40A SSR and heat sink from Amazon for under $20. A few lines of code and I now have a pretty decent replacement controller for well under $100 (although it's not quite done yet).

This is a temporary lashup - I am working on adding some protection circuitry to limit the maximum water temp and I will also be replacing the SSR with one that has better specs once I have figured out which one to get.

I'm just wondering if anyone else has gone down this road and has any insights to share or anyone has any interest in this project. As I see it the key issues are:

  1. use an SSR or mechanical relay? Each has advantages. The SSR is more efficient and simpler to drive from the 'duino, but it needs a proper heat sink. If I put the SSR inside the hot tub "heater package" I will need a pretty expensive heat sink. With the old controller, which used a relay, the motor drew around 12A; with the SSR I'm seeing closer to 15A now, and the water heats up faster as a result.

There are some pretty scary stories out there of SSR's catching fire when used in hot tub controllers. I have been discussing this with a very helpful gentleman from Crydom and he tells me this is because some SSR's use epoxy encapsulation. Crydom has stopped using epoxy and he says this makes a big difference. Having enough heat sink is also important.

  1. protection circuits - the old controller used 2 temperature sensors. I'm guessing one was a safety backup - it looks like one was connected directly to the MCU inside the heater package and the other was connected to the MCU in the display panel. Either one can shut the motor down if the temperature gets too high - redundancy. I am thinking of adding a button-type overheat sensor and killing the power if it trips. That will shut down the pump unless the SSR fails shorted. I could also use a second 'duino inside the box and use a second temp sensor. I don't know what happens if the tub develops a leak and the pump runs dry - I could certainly test that at the next water change.

  2. This is not a PID application, that would be overkill here. It's a simple on/off thermostat with a small amount of hysteresis. I'm already finding that my little controller has better temperature control than the original one from Softub. I can tell by the data from my power monitor when the tub is on and for how long, especially at night when there isn't much else in the house using power. The Softub controller would turn on for much longer periods, which resulted in wider temperature swings. The total run time is pretty much the same, I'm just turning it on for shorter periods more often. With outside temps around 45-50 F at night it's running for on average 25-30 minutes every couple of hours (a Softub is VERY well insulated, much more so that a normal hot tub).


This sounds like a good project. I too have a SofTub and even though mine is still going strong for the moment, I know it is just a matter of time before the board dies. I was also thinking of how to replace it with an Arduino. I would be interested in any lessons learned that you have and status of your project.

I know this is an old topic… I just picked up a used softub 300 … I had to replace the motor, it was shot from water damage and corrosion. I was able to get it up and running… But it seems I have some flaky electronic issues… I was thinking about going the Arduino path… I built a MendelMax 1.5 with a Ramps shield and arduino, so I’m a little fimilar …

Do you happen to have any photos ? or code you could share from this project ? I really dont want to spend 400 bucks on new softub controls … :confused: