Freeze Protection Concerns on Outdoor Chiller

[d5708c]

I'm concerned about the risk of freezing damage occurring on a packaged air cooled scroll chiller. The unit was spec'd with factory provided heater for it's brazed plate heat exchanger, and heat tracing for the hydronic pumping package. The manufacturer claims the unit will be protected to -20F/-29C. Having examined the unit after it was installed, I'm a bit skeptical and apprehensive about this claim. Piping inside the chiller is copper with 1/2"? epdm insulation. The manufacturer heat tracing does not cover all of the unit piping, but instead seems to only be wrapped around each of the two pump volutes for the dual pumping package. Although there should be some conduction along the axis of the piping, I'm skeptical that this is adequate. We've left the pumps operating continuously at the moment to keep fluid circulating. The hydronic package also included a small diaphragm expansion tank attached with a length of flexible hose. The expansion tank has epdm foam insulation, the flex hose does not, and neither have heat trace or fluid flow through them. This expansion tank is somewhat redundant, as another expansion tank was provided inside the building sized for expansion of the full system volume. There are isolation valves between the chiller and the rest of the system though which would leave fluid trapped inside the chiller when closed.

Spec's called for 20 vol% ethylene glycol solution as a working fluid, which should provide protection below the temperature corresponding to the minimum evaporator suction pressure, but not to winter low ambient conditions. To maximize dehumidification potential, the system was designed to give 39F leaving air temperatures with 35F on the waterside. This could not be achieved with stronger glycol solutions. I pushed for a split chiller with the evaporator and all hydronic equipment inside the building, but the architect would not give us any floor space in a suitable location. I really would have preferred a dx solution for efficiency, but because of distance between the air handling room and where we were allowed to place heat rejection equipment, this was not viable.

The location where the chiller is located is subject to an average of 5h a year of temperatures below -30C/-22F, with a record low of -34C/-29F according to weather records. The sales rep for the chiller indicated that the -20F protection was based on water as a working fluid, and that it should be able to maintain 52F delta T at more extreme temperatures.

 

[EnergyProfessional]

outside chillers always need glycol in the % needed to not burst at lowest expected ambient temp. What if that heater fails or you have a power outage????????

Keep in min freezing and burst temps are different.

 

[d5708c]

Using antifreeze of sufficient strength for burst protection is a common approach to providing freeze protection, but there are many applications where active freeze protection is relied on exclusively. Open and closed circuit cooling towers, as well as evaporative refrigerant condensers are commonly used in our climate and come with basin pan heaters and heat tracing for make-up water piping or condenser water. Active freeze protection is also used for shallow buried water and sanitary drains, and for utilidors above grade in arctic environments with permafrost.

In the facility in question there is a standby power system that will provide power within 7 seconds of an outage or poor power quality event(single phasing, under or over voltage,etc.). The building is located in a fairly wet area, and has subsoil drainage pumps cycling on and off 24h/day for most of the year to keep the basement dry. There are two means of freeze protection provided: electric heat trace on external chilled water piping and as provided in the chiller by the manufacturer, as well as cycling one of the two redundant chilled water pumps on for 15 minutes every 3h in extreme cold weather. Either one of these means alone should be adequate, so there is some redundancy. My main concern now is that I'm not sure I trust the factory insulation and heat trace on the chiller if the pumps should fail, and I'm concerned that the expansion tank in the chiller might freeze given it doesn't have flow through it or heat trace. I guess we can add more heat trace and insulation on piping inside the chiller cabinet for increased peace of mind.

The trade off in this case is that a lower brine strength allows lower temperature difference between the waterside and airside, increasing chiller efficiency and allowing lower dew points to be achieved. The capability for lower dew point discharge air reduces the volume of air circulated to meet latent loads, and the quantity of reheat required to maintain the temperature set point. Cooling coils can be a bit shallower with lower air side pressure drop. As a result, air handling equipment and ductwork sizes were reduced, as well as operating fan power. This is not typical comfort cooling, but serves a cool storage space in an archive facility (60F 50% RH year round).

 

[Drazen]

wrapping heating cable in spiral pattern is ok, as heat transfers through conduction over the whole perimeter.

 

[EnergyProfessional]

I'm not sure I understand your concern. On the one hand you dismiss the proper solution or using correct % glycol by saying you have all these electric protection features. Great... if you have all these features and think they are 100% reliable and Murphy's law doesn't apply - why do you ask a question then indicating you have a concern? Are you just looking for validation for your existing solution?

You realize your 20% glycol doesn't give you burst protection in your climate? 20% gives you ~5°F burst protection. So all your electric heaters and the pump are on entirely independent electric feeders not shared with anything else?

An HVAC sales guy once told me about outdoor chillers that as an engineer he really wants us to use glycol. But since he lives on selling parts and service calls he really wants us to use water in freezing climates. I'm sure he would have been happy to sell you all these electric gadgets and a new chiller in 5 years when the inevitable happens.

I now have to go out to my car again and start it. I only put in antifreeze for 0°F to save $5 and have to start it every 2 hours to protect it from freezing. No big deal... easier than to just put in proper glycol %. Especially since I always have time to do that, never run out of gas...

 

[cdxx139]

Then there is the train of thought that the facilities guys will not maintain it properly and let the % drift to the lower teens then the single digits.

The glycol seems to be the better, passive way to deal with it, but it sounds like it is too late? You already designed for the 39 degF DAT, and you will not be able to meet that with higher %EG? Is there fluff in the AHU to increase airflow which may allow an increase in DAT?

Go back and super insulate and heat trace the outdoor chiller piping. Shouldn't be too much work/money.

Is there a generator? Put the pump on the generator.

Or add a little freeze pump (sized for chiller and minimal piping static) just inside the building to rat race the chiller piping if/when the main pump dies



knowledge is power

 

[GMcD]

And who allows the use of ethylene glycol in building systems any more? Inhibited Propylene Glycol (Dowfrost)is what the industry standard is, and has been for the last 25 years in most of the North American jurisdictions I'm aware of. In 35 years in the business, it is standard good practice to use a glycol solution for air cooled chillers in cold climates, with the % appropriate for the winter temperatures anticipated. Heat tracing can fail, the power can be out (especially in winter) so what are gonna do?

 

[EnergyProfessional]

GMcD: ehtylene glycol is very standard and ever seminar etc. or study (liek the Trane ENL) use ethylene as standard.

Polypropylene glycol is used where there is chance for contamination in case of leaks (i.e. solar system with single wall HX, geothermal system). i heard rumors of more push for propylene glycol but nothing solid. I like to hear more detail?

Ethylene glycol has less pressure drop etc. so if contamination is not a concern, it is better.

 

[GMcD]

Most Canadian WorkSafe (Workers Compensation Boards) standards have virtually mandated the use of propylene glycol for building systems related antifreeze solutions since it's a lot safer for facilities maintenance staff health and safety. All of the School Districts and Institutional standards also require the propylene glycol solution. Most Provincial Ministry of Environment policies also require environmentally safe freeze protection solutions for geo-exchange systems as well.

 

[EnergyProfessional]

GMcD: oh I see, Canada. Here south of the border there is only remote talk about that. so my guess is we still have ethylene glycol for 20 more years (judging by how long the US kept R22...)

 

[d5708c]

The standby power system is a liquid cooled, commercial lp generator with block heater and battery warmer, installed to the standard of an emergency power system.

I'm somewhat skeptical about the freeze protection provided by the chiller manufacturer for hydronic components inside the chiller cabinet. The field piping is adequately protected with insulation and self regulating heat trace. For peace of mind I'll have this redone to match the field piping.

For a design condition of 60F 50% RH, the dew point is 41.4F. In order to handle internal latent loads the supply air dew point needs to be a bit lower. There is a desiccant wheel dehumidifier at this facility as well, but it's more economical to do as much of the dehumidification as possible with the cooling. React heat for the desiccant wheel is electric heat. Energy savings from this design were found to greatly exceed the energy required for heat trace. The generator was already being provided for subsoil drainage pumps.

Ethylene glycol solution was selected because it had the best fluid properties for this application. While ethylene glycol does have acute toxic effects, exposure risks are easily mitigated by following sensible handling procedures. Ethylene glycol will bio-degrade much more rapidly than propylene glycol when released to land or water, although neither have been found to accumulate in the environment. Last year over 5,500 tonnes of ethylene glycol were released over land from de-icing operations by the main ground support contractor at Toronto Pearson International Airport. If ethylene glycol presented a serious hazard to human health or the environment I would expect controlling aviation sources would be a much higher priority. While the goal of reducing the risk of workplace injuries is noble, banning ethylene glycol seems like a disproportionate solution to an exaggerated risk.