Heatwaves and CO2 refrigeration

Following this summer’s extreme high temperatures in the UK, Daniel Clark, MD of Isentra, asks whether 130bar transcritical systems have now become a necessity?

This summer’s heatwave in the UK has created havoc for many supermarkets with multiple stories of refrigeration system failures, lost stock and lost sales. 

All refrigeration technologies have been affected by this extreme weather, but now, with so many retailers using CO2 systems, Daniel Clark owner and MD at Preston-based CO2 systems manufacturer Isentra, asks whether 130bar transcritical systems have now become a necessity?

As an industry we need to rise to this ever increasing challenge of climate change. 32°C design ambient was commonplace when I arrived in the industry, but we are now moving up the temperature scale quicker than a standby engineer submitting his heat wave time sheet. 

I can see specified design ambients moving up to 42°C. That’s 10°C increase in the span of my career so far – and I’m a long way off retirement age.

So is poor maintenance to blame or is it a lack of understanding combined with system shortfalls for how CO2 really needs to work in high ambients?

We all know being heat wave ready through cleaning gas coolers is a must, as are coalescent filter changes and stripping and cleaning strainers, etc, but this will only go so far in the battle of operating in 40°C ambients.

The step change

The fundamental problem is gas cooler pressure – or a lack of it. Compressor selection software tells us that the optimum gas cooler pressure in such high ambients needs to be >96bar. Running at these pressures provides a massive step-change in the capability of a transcritical CO2 system. 

However, I suspect most CO2 systems are not running at this high pressure, because, practically, it’s very difficult to achieve. To save nuisance pressure trips and keep plant running (for as long as possible), most systems are probably set with a maximum gas cooler pressure of around 85bar or less. Unfortunately this is the exact opposite of what really needs to happen: gas cooler pressures need to be increased. 

Transcritical systems are generally designed with 120bar pressure relief settings. EN378 dictates that high pressure switches need to activate at 108bar. This only provides a maximum of 10bar head room between the optimum gas cooler outlet pressure and the dreaded compressor high pressure cut out – it’s just not enough in extreme conditions. 

The effects of low gas cooler pressure just spiral out of control – it’s a bit like trying to put a fire out with petrol. 

Lowering the gas cooler pressure in such ambients causes unmanageable amounts of flash gas. Ultimately when far too much flash gas is created, this can cause the MPV (receiver flash gas valve) to become undersized and the receiver goes high in pressure. 

Even if the MPV can cope, the compressors become far too busy recycling flash gas and have a drastically reduced actual refrigeration capacity. Ultimately the compressor suction pressure goes up and consequently the CO2 mass flow rate increases dramatically above design conditions. This causes much higher pressure drops through the oil separator and gas cooler and with the HP switches being on the compressor side and the gas cooler pressure set point being controlled at its outlet, this pressure drop easily brings the compressors up to and over their trip pressure. 

These higher mass flow rates also increase compressor power demand so circuit breakers and motor starters often trip too, especially in hot conditions – another form of trip where it presents a manual reset nuisance.

So what stops us running at optimum high pressure in the first place? The answer is gas cooler valves (HPVs) and controlling them. HPVs need to be very dynamic, especially when compressors stage in. If you are already operating at >90bar and a compressor stages on, many HPV control algorithms just don’t respond in the correct way to open the HPV and the 10bar head room disappears in an instant. This is the real issue that needs addressing in most instances.

The move to 130bar systems will get over this problem, having pressure switches set at 117bar, ensures operating gas cooler pressure at around 100bar are much more practical and, ultimately, reliable. At isentra we have built a few 130bar systems – but we will be offering them as common place going forward.

So what about all the 120bar systems already out there? I see two options. The first is adiabatic water conversions but this isn’t simple, with control, legionella safety, frost prevention, mineral deposits, corrosion, installation and maintenance all causing management and budget headaches. However, that said water feeds onto gas coolers in such ultra ambients does work very well, as systems benefit from wet bulb temperature.

The second option is refining control of the HPV valve, developing advanced HPV control practices is essential. I think it’s a necessity for control manufacturers to address this issue and provide detailed guidance on this subject. It is the foundation of the problem, albeit controlling so close to the upper limits is not easy. Also, every system has different operating characteristics. 

Making UK CO2 refrigeration systems work reliably in extreme ambients is entirely possible, we only have to look to Spain and Italy. Testing ultra ambient operation can be done in normal ambients and time should be spent with control manufacturers and live sites to get reliable operation with gas cooler pressures at around 100bar with gas cooler outlet temperatures at around 43°C. 

This will solve the problem that exhausts our industry and exacerbates our customers every time we get these blasts of heat; which we have to assume will be happening again in the near future, and, in all probability, more frequently.