Ice makers and the Art of Integration

By: Russell Muren

Hey there, it’s Russell again. If you haven’t read our previous post about sending our demonstration Unit to Southern California Edison, check it out here. In the meantime, I want to follow up that post with one story I believe sheds a lot of light on the challenges of developing IcePoint™ and the approach we take to solve it.

I want to begin by showing you two videos.


This video is boring and that is just the point. In IcePoint™, ice falls into the tank about 8-12 times every hour the system is charging. This video was taken during the afternoon right after we had run through all our ice during a storage period. You can see how little ice is left in the tank. Now lets go on to Video Two.


This video is extremely exciting for us at Rebound. What it shows is paramount in our quest to commercialize IcePoint™. You may have some questions at this point like: What is this showing? Why is it bad? Why did this happen? How many hours have you guys spent watching grainy videos of ice makers? Trust me, we will answer all of those questions.

To quickly contextualize these videos, we placed cameras inside our ice tank that show us liquid level, ice levels, and, most importantly, ice quality. What is ice quality you ask? That is a nebulous term we use to describe how well our icemakers are freezing our water. This can range anywhere from 100% rock solid ice to 100% unfrozen water. What you are seeing in these two videos demonstrates the extremes of this concept. In Video One, you are observing a well-frozen slab of ice falling into the tank. In Video Two, you are observing pure water being poured directly into the tank. As you can imagine, this change can be pretty detrimental to IcePoint’s performance.

As a refresher, IcePoint™ operates by changing the concentration of a freeze point suppressant (FPS). The cooling IcePoint™ provides is based on two ice tank characteristics: the amount of ice and the FPS concentration. Adding pure water into the tank hurts both of these metrics. Not only does it dilute the FPS concentration, but it precludes any real ice from entering the tank. Specifically, it will result in an increase in refrigeration temperature and a decrease in the overall system duty.

Quantifying water’s impact is an important step in addressing this issue. In the figure below, we can see how the system performance deteriorated when this water was put into the tank, and how the system recovered once it stopped. 

Screenshot of demonstration unit control screen showing impact of water dumped into tank during Video Two. An unrelated system shutdown is also shown when we were pushing a controller update.

Screenshot of demonstration unit control screen showing impact of water dumped into tank during Video Two. An unrelated system shutdown is also shown when we were pushing a controller update.


This is, in essence, the issue Rebound faces again and again: IcePoint’s™ FPS process is so new that we must solve a bunch of problems other manufacturers don’t. A little water falling into an ice bin is just another day for most icemaker manufacturers, but for us, it's a killer. That is why our current demonstration unit has been so critical. We always knew our cycle would work, but it was an open question if we could work with existing process equipment.

What we have learned from our internal testing is that, yes, we definitely can use existing equipment, but it is not as simple as hooking up parts supplied by manufacturers. To explain, lets look at the icemakers. You might be tempted, as we were, to think that all applications for ice making are basically the same. Thus, icemakers should already be more or less optimized by the passage of time and the hundreds of engineers who have spent their lives working on them.

Instead, we found that most of the process equipment we used was optimized for such a narrow band of real world applications that they only marginally worked in our slightly different application. This obstacle occurred a handful of times, including working with the pumps as well as the icemaker. What’s worse is that the issues occurred despite manufacture confirmation that their product would work for our specific set of operating parameters.  

What the demonstration unit has shown us is that we cannot rely on manufactures to know how their equipment works outside its normal application. While this may seem disheartening, it is actually freeing. IcePoint™ is unique and we now have the data to support our R&D efforts on several key pieces of process equipment. This work will now a higher efficiency product with significant economic impact.

Most equipment is designed for a narrow set of applications, which in retrospect, makes our struggles understandable. Commercial icemakers are designed to make ice that is: potable, clear, glass-sized, and made by a compact machine. The custom icemaker Rebound is currently designing, on the other hand, can make ice that is: non-potable, cloudy, awkwardly sized, and made by equipment with a larger footprint.

While we work on our own designs we will continue to use commercial icemakers that, with a few modifications, work well enough. But looking forward, it is an exciting time for Rebound’s design team because the demonstration unit has shown us so many small ways to improve our equipment. Time to get cracking!

P.S. We have watched and recorded many and more hours of ice falling into a tank. Let us know if you need any clips.