Making Very Cold Water

Hey,

Incredibly new to this forum, I keep ending up on it when I'm researching stuff which is usually a good sign that I'm in the right place to ask a question!

We're working on a project at the moment that involves refrigeration. I'm a software engineer, not a refrigeration engineer/designer but I've picked up a very basic understanding of how refrigeration works.

We're cooling ~300 litres of water down to 0 degrees to make sheets of ice and we're running into a few issues:

• Scaling the system up results in worse performance
• Our superheat control works 99% of the time until it doesn't
• We're not totally convinced we're using the right method to charge the refrigerant

There are definitely more things we're new to and need some help with!

We're not looking for someone on the internet to donate their time for free helping us and would ideally like to contract someone who has experience designing and building refrigeration systems from base components (compressors, evaporators/condensers, EEV, sensors).

Remote is totally fine, we're in the UK but a few timezones either side works too!

The client's genuinely a joy to work with and they've got good budgets for product development consultancy.

When you say sheets of ice, what thickness are we talking? 1mm, 1cm?

are you cooling the 300L of water down first, or is it part of the process of making the sheets of ice?

what refrigerant?

can you give a slightly more detailed description of the setup?

are you using of the shelf controllers or are you designing the automation system?

Hey Tycho,

The water’s cooled down to around 0.5C as part of the process.

We’re using R134a with a 1/3rd horsepower compressor, I think the total refrigerant charge is somewhere around 1kg.

The controllers are mainly just superheat-based control of the EEV using return temperature and pressure.

It’s bespoke as it’s got network connectivity and control.

Does that help at all? More than happy to pay for some time going through what we have on a call to get an opinion on where we can improve!

You basic setup seems fine: EEV for a precise control of SH... install anyhow a liquid separator to avoid liquid slugging in case of liquid flood back... as soon your SH is within spec (3 to 5 K) and you see that your EEV starts controlling (closing and opening) , your charge is OK. Adding some more with a receiver in the system doesn't harm at all. Are your sure you read exactly the correct temperature and pressure? Better insert the temperature probe inside the suction line via an immersed copper tube.
How's your evaporator made/looks like?

Thanks Peter! That’s really useful info. We can generally get SH in a good place, if it’s not we’ll definitely look at the charge levels.

Measuring temperatures on the return has been problematic, we’re currently using a temperature sensor that’s attached to the copper pipe using insulating tape which we’ve already seen issues with in terms of variability. We’ll have a look at ways to get it immersed inside the pipe as per your suggestion as I can totally see how this is going to be problematic if it’s inaccurate.

We’re using a suction line accumulator (looks like a metal canister) which I think is what you’re referring to in terms of a receiver and we’re measuring the SH temp and pressure before it goes into it.

The evaporator is the thing I’m perhaps most concerned about at the moment, it’s a long run of copper pipe that goes behind a stainless steel plate that’s backed by insulation and has thermal paste between the steel and the pipe. The surface area of it, while significant, probably isn’t anywhere near what would be achievable with, say, a finned design. Any ideas on how we can improve on that?

Theres another few outstanding items that I could do with knowledgable advice on:

- If the external temperature is higher, we see higher temperature on the liquid line (i.e. after the condenser). Is this a case of needing more airflow and/or a bigger condenser?
- We’d like to know when we’ve built up 1cm/1/3inch of ice or so. One of the ideas we’re investigating is to measure SH after the EEV and then SH after the evaporator to determine the amount of energy present in the refrigerant. Do you think we could use the delta as an indicator of the insulating properties between ice/water to anpproximate ice thickness?