- Tech Tips
I reached out to Jeff Neiman, our resident HVAC School chiller tech and he answered it. Here is the question
Thanks for all the good material you provide. I mostly work on the commerical building side of HVAC where chilled water is used as cooling medium and cooling towers provide condenser water. We have chillers as well as heat pump and air cool splits throughout facilities. Most of your diagnostics and troubleshooting methods are for air cooled units. Can they be applied to water cooled evaporators and water cooled condensers? My thinking is yes and no, because with cooling tower 85 supply and return 95 is maintained and 45 supply and 55 return chilled water is provided. Since there is not much change in these temps as opposed to outdoor ambient temperature there won’t be much pressure change in condenser. And as long water is regulated at proper flow to evaporators and condenser then all should hold steady. Do you have any input on this? I’m in NYC. Went to 2 year hvac school and worked almost 3 years in field starting out as a helper in service van as experience and learned as much then got into the building side for about 8 years now. I like listening to your podcast and reading your material as it keeps me refresh with field work as the building side is a little different but the basics and fundamentals are the same. Thanks!
The answer is yes.
Some of the measurements can be applied to chillers as well. Just some of the verbiage is different and the values differ.
The numbers for chilled water (44 out 54 in) and condenser water (85 out 95 in) are industry standard values at full load conditions. Most chillers regardless of manufacturer will have a 10*F delta T on the cond and evap. Machines that operate outside of those ranges are chillers that were ordered specifically to provide a lower temp or larger delta T.
Many people look at the compressor motor RLA% as the chiller capacity, which is not accurate. Chiller capacity is measured by the evap delta T. If the chiller is designed for 10°F(5.5°K) delta, and is currently providing 44°F(6.66°C)) water and the return water is at 49°F(9.44°C), the delta T is 5°F(2.75°K). So that chiller is currently running at 50% of its total capacity.
Subcooling is still measured the same, although the reading that you get will change as chiller capacity changes. At low loads your subcooling will be lower and will increase as capacity increases.
Suction superheat is a value that I really don’t look at because the reading on a flooded type of system will usually be very low or even 0. Rather discharge superheat (discharge temp – cond sat temp) is a more accurate reading and will be a direct result of you suction superheat. High suct SH, there will be high dis SH and vice versa. Again, this value will change as chiller capacity changes.
However if the chiller is a DX type, the suction superheat is just as valid as on a residential system
One of the values that was described in the podcast was temperature difference (supply air temp – coil temp).
In regards to air handlers with chilled water coils you can do the same thing. Measure your supply air temp minus the coil leaving water temp. This will tell you how well the heat is transferring to the water from the air going across the coil.
In chiller lingo this measurement is called approach
There are two different approach temps that i look at on a chiller:
Condenser approach (cond sat temp – lvg cond water temp)
Evaporator approach (lvg water temp – evap sat temp)
Approach values should range in 0 – 3°f(0°K – 1.65°K), given that your flows are correct.
Just like on air cooled units where proper airflow is needed across the evaporator and condenser, you need to verify that you have proper water flows.
In air to air applications you are measuring static to identify airflow issues. In water applications, I’m measuring pressure differential across each barrel. If I know my design pressure drop on the evap and cond, I can compare to my actual to know if my flows are proper. Keep in mind though that most chiller manufacturers will give the the design pressure drop in ft/hd. You will need to convert your real time reading to ft/hd to have an accurate comparison if you are using a gauge with a psi scale.
Even if your water temps stay pretty constant while in operation, your pressures will veer off as problems arise and your approach values will increase.
The chiller will always try to maintain at 44°f(6.66°C) chilled water out (or whatever the setpoint is) as long as it can do so.
The refrigeration cycle doesn’t change, stick to the basics and don’t over think it
When running building, try to get your condenser water as low as possible when running. But stay above 65°F(18.33°C).
Anytime you can provide condenser water lower than the design of 85°F(29.44°C) you will lower your condenser pressure and lower the lift (cond pressure – evap pressure). This will result in less work the compressor has to do and lower KW. This is a common method called condenser relief.
— Jeff Neiman
Sometimes we can focus on the more complicated aspects of system and parts installation like evacuation and flowing nitrogen and forget the simple and critical common sense steps to keep contaminants out of the system.
One of these is pre-cleaning tubing and connection before cutting or unsweating.
If you are replacing a coil, use sand cloth or a wire wheel on your drill to pre-clean the suction and liquid line connections before cutting. After cleaning with the abrasive, wipe down the area with a cloth to remove any loose particles and then make your cuts.
You would do the same when installing a line drier, clean the cut spots, wipe and then cut.
For something like a compressor it may be easier to get in there with wire brushes or a drill mounted wire wheel to do the pre-cleaning before cutting out or unsweating (I really prefer cutting things out if at all possible)
By cleaning BEFORE cutting you greatly reduce the odds you will get little pieces of copper or cleaning abrasive into the system.
A word of caution with cleaning copper – On new components, the copper may be a thin copper plating over steel. You will not want to overclean copper plating or you risk making it thinner or rubbing/burning through the plating. Obviously, if you are removing a failed component then this isn’t an issue but just make sure not to over clean new parts.
This is a subject that even many commercial guys don’t have to consider. For the majority of equipment, even refrigeration equipment, all that is required for proper oil return is to size the suction line properly, trap the suction line as needed, and allow for proper slope towards the compressor.
Then we get into larger equipment. Due to what can be extreme swings in load that result in wide swings in suction line velocity, oil return isn’t always what we’d like to see, even with proper trapping and line slope, so rather than allowing that oil to load up the evaporators and affect heat transfer among other problems an oil logged evaporator can cause, we install systems to prevent the oil from ever leaving the mechanical room.
I’ll try to lay this out in a step-by-step manner, adding layers of complexity as needed.
Since oil will be entrained (mixed / carried) in the discharge gas leaving the compressors, we’ll first want to install an oil separator in the discharge line to capture this oil, then we’ll work on managing its return to the compressor crankcase where it belongs.
That’s step one: separating any oil from the discharge gas leaving the compressor. There are 3 basic methods used for this (in order of effectiveness)
Impingement . In this method, all of the discharge gas passes through a screen where oil vapor gathers into larger droplets and drips off into a vessel where we can deal with it later.
Helical. In this method, the discharge gas enters the vessel at an angle and swirls around plates within the vessel. The oil droplets entrained in the discharge gas, being heavier than the vapor itself, are flung outward and hit these plates and drain to the bottom of the vessel as before.
Coalescing . Here, the discharge gas is forced through a filter where the oil droplets are captured and, again, drain to the bottom of a vessel.
Now that we’ve captured the oil, half of the job is done. We’ve prevented it from going out into the system, now we basically have a bucket full of oil under discharge pressure we’ve got to manage.
One thing to remember is that oil tends to accumulate the worst of the garbage in the system, so a quality oil filter is necessary. To prevent problems with clogging fine orifices, needle valves and pressure regulators we’ll encounter in our oil management systems, that filter should be installed as close to the oil separator as possible.
Things start to get interesting from here, so I’m going to try to explore the simplest methods first and dig into more and more complicated oil management strategies as we build an understanding.
Probably the simplest management strategy is one of the most modern ones. A direct oil level management system. An electronic float at each compressor monitors level in each crankcase and, as that compressor pumps out the small amount of oil it normally pumps, the electronics package energizes a solenoid valve to let that oil back into the crankcase. There will typically be a small orifice within this valve so that feed happens rather slowly but fast enough to prevent the level from dropping low enough to cause a problem.
Most equipment that is out there however, isn’t quite so simple, direct and easy to understand.
None of the other systems use electronics to manage oil flow, so from here on out, all controls are mechanical.
The next type of system uses mechanical float-type regulators bolted to each compressor to monitor the oil level in the crankcase. As before, when the level drops, the regulator needs to add oil back into the compressor. Much like a toilet tank or other float controlled device, the float opens a needle valve to allow oil into the regulator . The actual oil level within the regulator is adjustable within a fairly narrow range.
For this control to regulate properly, we need to reduce the oil pressure to a safe level. If we fed these regulators oil at discharge pressure, the high-pressure differential would force the small needle valve inside open and allow the regulator to overfeed and overfill the compressor. Instead, we install a valve between the separator where the oil is at discharge pressure and the regulators on each compressor to reduce the pressure down to typically about 20# above crankcase pressure.
Adding a couple layers of complexity to the system, we arrive at what is probably the most common type of oil system in use on parallel refrigeration equipment today.
Oil drains to the bottom of the separator vessel, as the oil level rises there, it opens a float valve. Oil passes through the float valve into a reservoir tank. The reservoir tank serves two purposes. First, it simply holds the oil until it’s needed. Second, through a special check valve installed between this reservoir and the suction header, the oil pressure is lowered to that same 20# above suction pressure figure. These check valves are available in different pressure differential settings, but 20# is the most common.
From the reservoir to the compressor, the system is the same. An oil line sends oil from the reservoir out to the mechanical float devices that control the level of oil in each compressor.
One other common feature in oil management systems like this is an equalizing line. We all understand that 2 containers of any liquid will have the same level in them due to the self-leveling nature of liquids. This equalizing line, in theory, connects the crankcases of all compressors together to create a self-leveling system. It doesn’t always work quite as well as hoped for because there can be different pressures in the crankcase of a running and non-running compressor. We’ll dive a little deeper into that as we move into troubleshooting oil problems.
While these systems seem complicated, and they have a lot of moving parts that can fail, they really boil down to oil level and pressure differential. We need to maintain a level of oil in the compressors and in the reservoir or separator and maintain enough pressure differential to keep that oil moving. Lose one or the other and you’re not going to stay running for long.
Let’s kind of walk step-by-step through a troubleshooting process until we find or eliminate all problems. First, I look at all compressor levels and reservoir level. If I’ve got a lot of oil in the compressors, I want to check equalization lines. If they’re all consistently low, I’m going to start looking at the oil management system.
To evaluate the oil management system, start by checking the temperature leaving the oil separator. The line leaving the separator should be warm to the touch (100F). I like to put a thermometer that logs min/max temps and observe it for 10 or 15 minutes. You’ll see the temperature climb and drop as the oil float inside feeds. No feed? Time to consider the float inside as a problem.
Next, let’s check the pressure in the reservoir or the outlet pressure of the unit’s pressure reducing valve against suction pressure. 20# above and we’ve got level in the reservoir? OK. No level in the reservoir? Lets try to find that oil before we go adding oil. A system with too much oil can be as problematic as one with too little oil… If we don’t have differential in the reservoir, I’ll isolate the inlet and outlet of the reservoir and bleed some hot gas from discharge into the reservoir with my gauges to see if the differential check is faulty. Since we’ve already demonstrated that the separator is feeding, we need to see if the differential check valve has failed.
Next step for me is to start checking each individual oil level regulator. I’ll normally uncap the adjustment stem, turn it CCW to the top stop, counting the turns then adjust it down (CW) to a midpoint which is typically 5 turns. If any are wildly out of adjustment, I single that compressor out for some additional attention. It is very important to not adjust these controls more than 10 turns from the top stop. The adjustment range is limited and adjusting beyond that limit will ruin the control, regardless of its condition before you worked on it.
I mentioned crankcase pressure earlier, and this is an issue that can be problematic with oil issues. As the rings wear in a compressor, we can see some discharge blowby into the crankcase. Not enough to warrant replacement of the compressor necessarily, but enough to sometimes cause issues with oil. First, if we put additional pressure into one compressor, that unbalances the oil equalization system by pushing oil in unwanted directions. Second, by increasing the pressure in the crankcase over the suction pressure, we reduce the net oil feed pressure, slowing the oil feed rate.
To check a compressor for pressurized crankcase, install a gauge on both the suction port and the crankcase. If the pressure within the crankcase is more than about 2# higher than suction, you may have some problems.
A few other, kind of random thoughts on oil failure trips and trouble.
These are 3 phase motors. A contactor with severely pitted points or a contact that doesn’t make good contact can cause a temporary single phase, prevent the compressor from running and create a situation where the oil control is energized but no pressure is created by the compressor. Always check the contactor.
Screens and filters. Since the oil system tends to collect all of the garbage in the system, oil systems tend to have a high concentration of filters and strainer screens. From the impingement screen and coalescing element in a coalescing separator, screens are a huge problem. Add to that the screens, an oil filter, the float at the bottom of the separator, the pressure reducing valve, the screens and valves in each individual oil level control and, often an oil pickup screen in the compressor itself and there are many points that can become obstructed by the debris in an oil system. Regular oil filter maintenance is important for a reliable system.
— Jeremy Smith CM
P.S. – Henry Technologies has compiled everything that you can possibly need to know into a handy manual and, most importantly, a quick-reference chart with some basic diagnostic readings and measurements to take HERE
Both wet bulb temperature and air enthalpy are extremely useful to understand when calculating actual system capacity as well as human comfort. Dry bulb temperature is a reading of the average molecular velocity of dry air, but it does not take into account the actual heat content of the air, or the evaporative cooling effect of the air.
Like we mentioned in the last tip, when air is at 100% relative humidity the dry bulb, wet bulb and dew point temperatures are all the same. This is because at 100% relative humidity the air is completely saturated with moisture and can have no evaporative effect.
When air is less than 100% RH it will provide an evaporative cool effect and wet bulb temperature is a measurement of that effect. In fact, wet bulb temperature is the temperature a damp thermometer bulb will read when exposed to a 900 FPM (Feet per minute) air stream. If you have ever seen someone using a sling psychrometer, that is exactly what is happening (Hopefully you have a wrist that is well calibrated to 900 FPM). The lower the wet bulb in comparison with the dry bulb (This differential is called wet bulb depression) the lower the relative humidity and the greater the evaporative cooling effect.
Enthalpy is the total heat content of the air and is represented in BTUs per lb of air. By converting lbs of air to cfm we can calculate the amount of heat in an air mass as well as the change in the enthalpy across a coil to calculate the heat moving capacity of a coil, BTU losses/gains over a length of duct and much more.
You will notice that wet bulb and enthalpy are slanted lines, descending from left to right and they are equivalent. This means that a particular wet bulb temperature is also equal to a particular enthalpy (At 14.7 PSIA at least). In the chart above you can see that a 62.8 degree wet bulb mass of air contains approximately 28.4 BTUs per lb. The tricky part is reading at this extreme level of resolution, because 28.4 vs. 28.6 can make a significant difference when it is multiplied out over a large air mass. This demonstrates why VERY accurate tools and careful calculations are required for enthalpy calculations in HVAC/R.
For a full WB Enthalpy calculator go HERE and look for the enthalpy chart
HVAC School shows you the new SW-HVAK05 and SW-HVAK15 kits from Solderweld and everything they can do including copper to copper, aluminum, copper to aluminum, steel to copper and much more. Featuring Bryan Orr and Sal from Products by Pros.
Another follow-up article by Michael Housh… Enjoy!
In this article I thought I would show a pump curve and match it up with our system head-loss curve that we created in the last article, but before I do that I thought I would talk a little bit more about the system head-loss curve and why I said it was so valuable.
When we design a hydronic system we must match flow-rates and output according to our Manual-J and the specifications for the radiator/panel that is being utilized (sort of like Manual-S on the air-side of HVAC). Here’s an example output specification for a baseboard radiator that I found on the web.
You’ll notice that it gives two different output rows (1 GPM or 4 GPM), you may also notice that the output is relatively close for both flow rates. It is safe to assume that we can move 1 GPM with less power consumed than moving 4 GPM. Let’s check the difference using the Head-Loss Equation from the last article.
This system is so small, that most circulators are going to produce more than enough flow, but you could maybe imagine how this could greatly affect a much larger or more complicated piping arrangement.
Next let’s look at a pump curve, the following is for a Taco VT2218, this pump fits a ton of applications, is variable speed, and maintains a constant Delta-T across our system. In today’s world, it doesn’t make sense in most applications to not utilize an advanced pump, in my opinion. Now using our 4 GPM and 5.13 ft. of head, we can see that we fall somewhere between speed 1 and 2 (or well within the shaded green area on the right), so this pump will be great for this application (and hopefully sized right for future exercises where make the piping arrangement more complex).
To end the article I thought I would follow up with more detail on the Flow Coefficient (Cv) mentioned in the last article for our air-separator. It is not uncommon in design for items to have a Cv value instead of equivalent length. If you remember from the last article, Cv is the amount of flow (GPM) required at 60° water (important because of density) to create a 1 psi pressure drop across the device. We can use the following equation to convert Cv to pressure drop for our design conditions using the following formula.
Given the above equation, we can solve for our pressure drop using the design criteria of 170° and a flow rate of 4 GPM. The density of water at 170° is 60.81 (I have also created an interactive Water Density Chart online).
Using the equation from the determining pump flow article we can convert this pressure drop to feet of head.
Finally we can solve for the equivalent length for a given pipe size coefficient using a rearranged version of our system head-loss equation from the previous article.
I hope this has helped gain a deeper understanding of selecting a circulating pump and how to convert a Flow Coefficient (Cv) into an equivalent length. In future articles, I hope to create a more complex piping arrangement to show the pump selection and equivalent lengths for those applications.