HVAC

September 2019 Newsletter, 5/11/21

In the last newsletter I described ways that we look to find value in the design of deep foundations—specifically, questions we try to ask ourselves to lead us to the most cost effective and constructible designs before projects ever break down.

Another rewarding place to look for value engineering is a building’s HVAC design. Last quarter I discussed how the results of value engineering a building’s foundation are binary (either the building stands up or it doesn’t). This is very much not the case for HVAC systems. As we make changes to system designs for cost reasons, it is important to remember that we also change the resulting building in a tangible way for our clients and other stakeholders. This could be in terms of energy efficiency, operational cost (not always as intuitively linked to energy efficiency as one might expect), controllability of systems, comfort for occupants, aesthetics, and longevity. 

Frequently, the issues identified in this article appear to be related to the misapplication of energy code or outside sustainability standards. Before reviewing the questions here it’s necessary to first understand the project’s compliance path and sustainability goals—and to make sure the designers are all on the same page on this topic.

The following is a list of questions we ask ourselves when reviewing an HVAC design to save cost.

Is the primary means of heating and cooling selected also the most cost-effective option?

Whether the primary means of heating and cooling apartments that has been selected is the most cost-effective option appropriate for the building type is always the first and most fundamental question when looking at a building’s HVAC system from the perspective of value engineering. There are a variety of different systems we often come across as the primary means of heating and cooling. Ranked from most cost effective to least, the typical systems we encounter are:

  1. Hydronic “Baseboard” and “Through-Wall” ACs

  2. Gas-Fired PTACs (Packaged Terminal Air Conditioners)[1],[2]

  3. VRFs (Variable Refrigerant Flow) - Non-Ducted, Heat Pump (or “Two-Pipe”)[3]—although non-ducted and heat pump need not go together

  4. Water-Source (or “Hydronic”) PTACs

  5. VRFs - Ducted, Heat Recovery (or “Three-Pipe”) –although ducted and heat recovery need not go together

  6. Water-Source Heat Pumps – Non-Ducted

  7. Hybrid Water-Source Heat Pumps – Non-Ducted[4]

  8. Water-Source Heat Pumps – Ducted

  9. Fan Coil Units – “Two Pipe”[5]

  10. More exotic systems such as Four Pipe fan coils, water-cooled VRFs, geothermal options, etc.

It is far beyond the scope of this article to describe the systems above or the various pros and cons from an owner’s perspective in operational costs, comfort, etc. Here we are primarily concerned with cost. With that perspective, the following (extremely basic) principles influence the above ranking:

  • More ductwork is…more work. To the extent we can limit ductwork, we can save cost. All else being equal, the non-ducted options (where equipment is exposed or located immediately behind a partition without any horizontal distribution) are always less expensive. For this reason, a non-ducted VRF is less expensive than a concealed horizontal unit of the same type. The same goes for water-source heat pumps, concealed fan coil units vs. console fan coil units, etc. Of course, no rule works in all cases and it is important to remember that—depending on how much ductwork is being added and the incremental cost of additional equipment pieces—there are cases where adding ductwork will be less expensive than additional pieces of equipment.

  • More piping is…more work. Like ductwork, we can save cost by limiting piping. For this reason, the “two-pipe” options of the systems above will almost always be less expensive than the three- and four-pipe version. Typically, the additional pipes are used to allow for differential heating and cooling between units attached to the same loops. In the case of VRFs, “three pipes” only refers to the quantity of piping leading from the condenser to the branch selector box[6],[7]. In the case of four-pipe fan coils, the “four pipes” refers to the piping leading to each unit—each unit receives a supply and return of boiler water and a supply and return of chiller water.

Some systems scale down better than others. Systems with large centralized plants (all the systems that involve cooling towers, chillers, or significant hydronic heating plants) will be more cost-effective on large projects than small ones because of their higher fixed costs. Systems with greater modularity (such as Gas PTACs, which are 100% decentralized, and VRFs, which can have scattered and very small condensers) will often be more cost-effective on small projects. Even though boiler plants and cooling towers can scale down to some extent, there will always be fixed costs associated with each—especially boiler plants which require rooms, breaching, life safety controls, etc.).

Is the mechanical engineer over-specifying?

Over specifications can happen in hundreds of different ways. We see this most frequently in controls (covered further down), insulation, sheet-metal gauges and connection types, condensation drains, equipment support, vibration isolation, and seismic requirements. Here are some related questions (for the most frequently recurring issues) to think about when looking into over-specification:

  • Are the ductwork gauge requirements appropriate?
    For example, Monadnock recommends 24 ga for toilet and kitchen exhaust; however, drawings will often specific heavier gauges.

  • Are the ductwork connection type requirements appropriate?
    For example, we have run into documents (and projects) requiring 100% flanged ductwork connections when elsewhere we have found “slip and drive” perfectly adequate.

  • Are insulation requirements appropriate?
    It is not unusual to find three or four different, contradictory sets of insulation requirements in a contract document set (between drawing lead sheets, written specifications, and equipment schedules). They are rarely right—and by “right” here I mean cost-effective. For example: at the time of this writing, Monadnock recommends no more than 1” insulation on refrigerant vertical and horizontal piping and no insulation at all on CPVC condensation drain lines; however, drawings will often specify more.

  • Are all condensation drains required?
    For example: we often encounter drawings that require internal condensation drainage systems for all hydronic PTACs that are completely unnecessary (we have never actually installed this).

  • Is equipment support overly elaborate? Engineers have the tendency to call for full structural dunnage on their drawings regardless of cost. Most equipment will not need this type of support—prefab curbs will suffice for most fan, condenser, and RTU applications.

  • Are vibration isolators specified where they are not needed?
    For example: drawing sets will often call for spring-type vibration isolators for VRF condensers when manufacturers recommend only neoprene pads.

  • Is the engineer using the appropriate fire/smoke damper for a given HVAC system or wall rating?
    For example: engineers will often show combination fire/smoke dampers when only a fire damper is needed. (The former also adds to electrical line voltage and fire alarm scope when introduced to the HVAC design.)

The questions above only scratch the surface of the over-specification common amongst MEP engineers. Reviewing these questions and more is essential to value engineering, leveling bidders (who will often ignore such requirements) and understanding what we are buying in an HVAC contract. 

Are layouts and ductwork distribution strategies as cost effective as possible?

Architects and engineers can generate a lot of unnecessary cost in horizontal ductwork if they are not careful. We often focus on the efficient orientation of bathrooms to avoid unnecessary cast iron risers for plumbing, but apartment layouts can also be efficient or inefficient for the distribution of kitchen exhaust (KX), bathroom exhaust (TX), and fresh air supply (for those limited instances where we see it ducted to apartments such as luxury projects or Passive House certified projects) as well.

When architects and engineers don’t speak and coordinate, the results can be wasteful. The architect will email his floorplans over to the engineer (which may not at the outset provide efficient duct riser locations) and the engineer, without discussion or revision with the architect, simply plunks her riser down in any place available and includes horizontal ductwork as necessary (generating cost). Such is the state of the design industry in NYC today.

What we look for when reviewing the above is kitchens that are as “back-to-back” as possible with a KX riser located in between them and limited horizontal “run-outs”—direct riser taps are ideal in most cases. For bathrooms we are looking for the same, although we may get more push-back from the engineer because direct riser taps can generate sound transmission issues between apartments or create moisture issues in the duct if the tap is too close to or within the shower (according to some). 

Although not an issue of comparative efficiency, we often see fresh air supply ducted to apartments when it is not required by code or common practice (and thus try to eliminate it to save our clients’ money).[8] As mentioned above, this is required in Passive House projects and will likely be expected by buyers and renters in the condo or ultra-high-end rental market; however, it is dead standard practice to provide only exhaust for rental apartments otherwise. We sometimes must remind engineers of this.

Is the controls strategy cost-effective?

In recent years it seems that engineers’ and clients’ expectations regarding HVAC controls continue to grow. Besides cost, this can also result in systems too complicated for residents and building superintendents to operate correctly. The following are some of the more frequently recurring examples of controls systems that are worth questioning (as they sometimes may not be required):

  • Centralized Control for PTACs. Clients and engineers will sometimes call for centralized (wired or wireless) controls for PTACs. This can give them a variety of capabilities including monitoring energy use, monitoring trouble indicators or maintenance alarms, and imposing a dead band on heating (the cost of which is typically born by our clients). However, it can also generate additional cost in the PTAC material cost and the installation of wiring or wireless accessories such as nodes and repeaters.

  • CFM Monitoring. Sustainability consultants will often include permanent CFM and other air quality monitoring. This may be related to discretionary bonus points in whatever sustainability rating system the project is pursuing—but it could also be related to the capricious whims of the consultant. If the system is elaborate (many points and capabilities), this type of monitoring can add a non-trivial amount of cost.

  • Interior Space Temperature Feedback for Boilers.
     In years past, a boiler basically only “knew” the outside air temperature, HHW temperature, pressure, and flow (if it even modulated at all…). Today, we sometimes see interior space temperature feedback as a requirement for boiler controls[1]. With this strategy, a boiler will modulate and control itself using data from some sampling of apartments’ interior air temperature. Some believe this can increase the efficiency of the hydronic system from the perspective of operational costs, especially on renovations where the original systems may not have been designed with adequate controllability—but can also generate additional cost in wiring or wireless accessories.

  • Apartment Thermostatic Control. We often see electronic control or zone valves for apartment hydronic baseboard. This can allow a tenant to have a digital programmable wall thermostat to control the heat in their apartment. Certain more zealous engineers will call for this to be interlocked with through-wall ACs so both heating and cooling are controlled from the same thermostat. Setting aside that latter possibility, even a basic electronic control valve and digital thermostat will create a variety of additional work—there is line voltage wiring to a transformer, mounting the transformer in a location where it won’t be unsightly or overheat, low voltage wiring from the thermostat to the valve, etc. And another component has been added to every apartment that can eventually break and need to be replaced. The alternative is a mechanically actuated thermostatic valve—often referred to by the most common brand, “Danfoss”. These valves can also have remote thermostats (although with less distance from the valve) and have no electricity required.

  • Individual Room Control. When engineers or clients require that every room have individual heating control in a hydronic system, they are not simply causing the addition of a thermostat and control valve for every room, they are also causing the introduction of supply and return risers for every room. This can have an extraordinary impact on cost due to the amount of labor and material related to the additional piping.

The issues above barely scratch the surface of HVAC value engineering. But hopefully it has provided a good starting place for analyzing Mechanical drawings in the future.

[1] The cost difference between baseboard with through-wall ACs and Gas PTACs that we have seen is close to negligible and certainly not statistically significant. This is assuming that in the former system the through-wall AC equipment is within scope (sometimes these are purchased directly by our clients or by their tenants).

[2] It is important to note that gas fired PTACs will also generate gas piping costs that is usually not carried in Division 23. It is typically carried in Division 22 and for that reason sometimes people forget to count it in comparisons. We did, however, include it in the comparison that led to the above rankings.

[3] Ranking efficient VRF systems as less costly than hydronic PTACs is somewhat controversial. We are likely the only GC/CM in the city making this claim. We also happen to have done more VRF work in affordable housing than anyone else as well.

[4] These are the same as the previous option; however, they can operate as a pure hydronic heating fan coil unit in the coldest weather—i.e., their compressor turns off and hot boiler water passes through the coil instead.

[5] We are using “Fan Coil Units” here specifically to refer to systems that utilize boiler and chiller water. This is sloppy usage, however, as almost all the systems above involve something that can be—and often is—described as a fan coil unit because all of them involve a fan blowing across a coil.

[6] Also referred to as the branch control box, valve manifold box, branch selector, etc.

[7] Mitsubishi, for the most part unique amongst VRF manufacturers in this respect, only uses two pipes even from the condenser to the branch selector for their heat recovery system.

[8] It is important to note that not all agree that fresh air ductwork is not required. The code question may hinge on the engineer’s interpretation of the use of trickle vents and their willingness to prove or provide calculations for the trickle vents related to the lowest calculated neutral pressure plane.