Expertise Needed - Residential Construction Project

I'm a ME but HVAC ain't my bag. But your specific question about controlling water flow can be answered in general terms. Especially since the residence is sounding like a fairly high end one where cost isn't that big a deal.

Modern water flow control usually uses variable speed pumps because the energy efficiency is much better. And the flow control is smoother, without any concern with water hammer-which might not be an issue in hydrionic heating, since the pressures are low.

Actually, it seems to me that the pumps and valves (and control system) should be able to be provided to you as a one-stop shop from the same suppliers that have helped you with the rest of the HVAC system. Can't you get these people to spec them?

Jon.

Thanks for your input. I was leaning towards the variable speed option for the reasons you stated but wasn't quite sure if my conclusions were correct. Most of my experience has been in the commercial end of the construction industry and I have almost always seen air handlers piped with pneumatically controlled 3-way valves so I thought there must be a good reason for it.

Purchasing the pumps and such is following the same basic pattern as the rest of the project in that it is subject to the......idiosyncrasies of the owner who needs to be comfortable that he has learned absolutely everything that he reasonably can about the subject in question before he makes a final decision about anything. This explains why it has taken about 5 times as long to bring this project to it's current state of completion as would normally have been necessary. That said, he is genuinely seeking to do the best job possible within the limits of his budget. Cost actually is a factor, but he tends to spend what is necessary to get it right. What I'm trying to do is find someone with the expertise and the willingness to help us finish out these details so that he is comfortable with the decision. I would prefer someone local but a lot of the collaboration to date has been via telephone and e-mail. That seems to work as well. AutoCAD drawings and any other documentation can be provided in almost bureaucratic quantity if necessary to help the process along.

Well, there must be a phone book full of HVAC engineers in your immediate area, if your owner is willing to spend the money. It sounds like the guy is wanting to understand the design...not necessarily a bad thing. It will mean the HVAC guy will have to do the design and then spend some time explaining it to your client.

Just a thought about the pumping--if I am not mistaken, these hydrionic pumps are usually pretty small-in the 1/8 hp range. During the heating season, at least, if you used constant-speed pumps, the energy lost due to the throttling of flows would be converted to heat, which is what you are dealing with, anyway. And unless you are in an area where electrical rates are heavenly, the payback from the energy saved from a herd of tiny pumps is pretty small.

And thinking about the controls on the zones: Presumably the nice way to set these all would be some sort of central control panel or at least a T-stat in each zone. You had mentioned compressed air controls, which I kind of think may be getting to be a little old school these days--electrical controls lend themselves to computers and night setbacks and all the gadgets that people seem to like nowadays. If you are controlling the humidity as well, then all the more argument for a sophisticated electronic controller. I would suspect you can get electrically controlled water valves for this reason. The controller should also have enough smarts to shut down a zone pump if the heating/cooling load is satisfied for that zone and only restart it when a thermostat calls for heating or cooling.

"Guest" raises a good point, too--if you are cooling a floor there will be some concern that water will condense out of the air onto that cool/cold surface. At least you can control the condensation in the air handlers and route it to a drain. A cold soggy floor wouldn't be too good to impress the house guests with, especially if it has a rug over it! I guess if one is monitoring humidity, then a sophisticated enough controller will be able to decide if "radiant cooling" should not be used if the humidity is too high due to the potential for condensation!

Jon.

I'm civil not mechanical, but do tons of piping and specify pumps all the time. Here's my 2 cents.

As with most things in engineering there is a give and take. You sacrifice one thing to gain another. The constant/variable speed pump question fits into this. The constant speed pump is a cheaper install, but has a higher operating cost. Operationally either system can be designed to achieve the desired performance as you know, it's just a question of would you rather pay at the front of the project or over its life?

If the load will not dramatically swing, one way or the other, the two way valves are the way to go.

If the comfort zone temperature and humidity are critical and tight range than the three way would suit you better.

Hope this helps and good luck

I do get involved in Hydronic heating.

1: Get in touch with a plumbing supplier that has a dedicated Hydronic department.

2: Tell them what you want. Have all the information ready to give. Tell them you want a complete package. Boiler, valves, pump, controls, etc.

3: When dealing with multiple zones, check valves are a must.

I would go with the variable speed pump controlling a minimum loop pressure. The three way valves will eliminate the need for manual zone mixing valves.

You have indicated that the owner is considering the in floor system to augment cooling. As another poster pointed out, this is just not practical. Floor coverings, the temperature of the floor that must be maintained given the emissivity of the floor's finish, etc and your local climate potentially all conspire against such a cooling solution meeting his expectations.

Having some experience in designing similar systems for industry, and having my own radiant floor heating system in my home in AZ (where we need cooling 9 months out of the year), I dispossessed myself of the floor cooling approach during my design phase. I don't use the radiant floor system for cooling. I would offer the following, but please bear in mind that this response is given assuming radiant floor heating, not cooling:

Without knowing the individual zone run lengths & pressure drops anticipated due to the routing of the (I'm assuming PEX tubing in the floor) this part of your problem seems do-able. The TACO line of pumps for small zone systems (typically 1/2" dia & less than 500' long each) work well for me, but your OP indicates 16 zones. This indicates an industrial sized, but still likely a fractional HP end suction or vertical circulator pump, will likely be needed. There are MANY VFD's available for <$300 that will work up to 2 HP and single phase. Beyond this you could easily go to a 3 phase pump motor, and have the VFD go from the single phase 240 VAC to the 3 phase motor, as these devices typically derive the 3 phase from an internal DC bus anyway. The trick is sizing the pump to be able to handle all the zones flowing at the same time, as well as having sufficient "pump curve" available to handle the significantly reduced flow requirements at the slower RPM. You would want to estimate the amount of time at part load to see if this solution makes sense for your geographic area.

Have you considered that the valves might also be throttling type, 2-way valves that control the water flow on the discharge end of each zone (always best) to an adjustable, but locally set, return temperature? In this case, you could still use the VFD to save the pumping power when the zones are at temp.

You also did not indicate the boiler size and whether it has any special requirements for minimum flow, or whether a main supply & return piping header will be used. Depending on these details I could help you out, but I would like to see a statement of work before committing.

There was no mention of in floor cooling, just heating. Water Buffalo is correct to use the air handlers for the AC. This is how I would build my house, just a bit smaller.

The use of two way valves for heating will result in a vary hot section at the start of the zone, as you will be sending in water at what ever temp the boiler loop is. Unless you have a three way mixing valve upstream per zone. Different applications require different temperatures. The air handlers will need a higher temp water than the floor zones. A modulating three way valve controlled by the zone thermostat will provide the correct supply temperature. Note there should be a maximum limit set for all the floor zones.

Depending upon the layout of the in-floor tubing, supply and return header piping, etc, a "very (sic) hot section at the start of the zone" is not necessarily the case. The fact is that the inlet of each zone will always be warmer than the outlet, or no heat is being transferred into the floor. The VFD system allows the most energy efficiency AND comfort, as the delta T between supply & return headers will decay to some minimum value, as will the pump flow, which will serve to equilibrate the zones. And a warmer zone inlet is not always undesirable; it is often useful. It is common to route the first pass along the periphery for exterior zones where the most heat transfer occurs.

The VFD enables upstream temperature control of a common supply header line via a single mixing point. This system works with the thermostatic 2-way zone valves to keep the return water at the set temperature, and thus control the amount of heat going into the floor in each zone by flow rate and not temperature, per se. This type of system enables greater long term cost savings by reducing, but never completely shutting off each zone.

Putting the proper controls in place allows the boiler to operate at its peak efficiency point, as well as the pump and still achieve cost savings with the VFD while maintaining comfort in the floor heating.

You have valid points. A properly designed in-floor tubing system is an advantage at the first call for heat. The problem that I see is that your thermostatic return valve is at the exit of the loop. Temperature over shoot can be a problem as the valve only senses the return and not supply temperature.

You mention the single mixing point. The next problem is that you have multiple in floor radiant heating zones, mixed with multiple fan coils. Either system will require different supply temperatures. Also a single zone, like and entrance way can, have very high intermittent demands. The system would have to divert all available hot water to this zone. A central mixing valve will not work, as the pump volume could not compensate for the low water temperature of the mixing valve.

That is unless you have a massive pump and over-sized tubing, or you are running the mixing valve at a very high temperature. This will negate the energy efficiency and still led to very high inlet temperatures to the in floor heating.

It is best that each zone and or device has access to the highest water temperature from the central boiler feed loop. Use a multiple header system off the loop. Three way valves controlled by a zone stat as close to the load as possible is best for control. This will give you both, the capacity and control needed.

I'm glad to see we're on the right track. At least it appears so from what you posted since we have been over most of those topics repeatedly....except the thing about the 3-phase pumps. We only have single phase available. The system is relatively complex for a residential application but I created a piping schematic to help with the placement of pumps and various valves that would be required. We took the schematic to a local ME and he checked the pipe sizing and flow requirements, and made some notes and comments, but he seemed to think things would work. When we asked him to recommend pumps and valves though, he said we would have to get the manufacturer to to size them. He didn't seem to get the fact that we were looking for him to select a pump or valve that would work in our system and them we could go find a manufacturer with a product to meet the need. I kinda' thought that's what engineers did. The radiant system is going to be tapped off of the loop for the air handlers with it's own pump and tempering valve because it doesn't require either extremely hot or chilled water to function. The air handlers will run with 140F water for heat as opposed to the 103F required for the radiant. The radiant floor disperses the btus over a larger surface area over time whereas the air handlers have to get all the heat into the air quickly from a relatively small coil surface. The cooling side will have a similar difference in temperature with 44F for the air handlers and 59F for the radiant. Humidity is controlled so there should be no condensation on the floors at those temps during the cooling season. There are several comprehensive control systems available but I think we have pretty well settled on a system from Uponor which, I'm told, they adapted from a company called Delta. The Uponor system purports to be able to handle all of the elements we had so far envisioned the system to have. We have the humidifiers on site as well as ERVs and the air handlers and associated ductwork have been nearly completely installed. All the radiant loops have been embedded in the slabs and the manifolds and S&R distribution piping has been run for the radiant system. The S&R for the air handlers is underway, but I don't want to go any further with that until I know exactly what pumps and valving I'm putting in. I find that doing my piping layout is a lot simpler when I know that stuff.

We have been to a couple of local MEs but they didn't seem to interested in this project. I think it's might be due to the unconventional nature of the system and, frankly, the unconventional nature of the client. Most folks in the industry want to do the whole design like they want it and say "here, go buy this stuff and call this guy to install it." They're just not used to clients like my boss. I understand we are not following the typical glide path but we are trying to integrate as much as we practically can of the best modern methods available. That's why we need to find someone who is willing to work a little unconventionally as well.

Hey water buffalo, you're saying all the right things.

It makes sense that most ME's wouldn't want to hassle with finding which zone has the highest DP from which to begin pump sizing calcs.

If you're too far along some of your options may have already been decided, like where to put a pump or pumps. I would need to see your process & instrumentation diagram (P&ID) or piping drawings if you've got them, (or something close to it in the plans?) to avoid buying a small circulator pump, DP gauge (or 2 very accurate pressure gauges), and rising-ball, or rotameter type flow meter and individually testing all loops at their termini for flow rate vs pressure drop. This information is critical to properly sizing the pump for the radiant system. Of course it can be done based on theory, but doing so necessitates a safety margin greater than that of knowing the true system friction head at the design flow rate.

It sounded like you are looking for a single pump and VFD solution. There are killer VFD's out there for the 240 VAC single phase market. Even outstanding quality stuff like this toshiba unit I found on ebay: http://yabe.algebra.com/~ichudov/misc/ebay/Toshiba-5-HP-VFD/ebayhist.html that would be guaranteed to be large enough for your load if the price is right.

Check out this place for great deals on your single phase drives: http://www.driveswarehouse.com/Drives.htm?filter=1&manufacturers_id=10 . Remember in getting a drive that is specifically sized for your load to get a constant torque unit.

From experience, (and as a CEM & installer/operator of similar system as well) I would not want to have to maintain and operate several pumps per system when fewer would do. It has been my experience that the smaller the pumps get, the more often they get replaced, and the lower the overall efficiency. Plus if the system is operating correctly, the water would never be stagnant, for corrosion inhibition considerations, and the fractional current draw of the single pump makes all kinds of sense.

Last thought, again about the floor cooling: Fredericksburg, VA could see ambient dew point temperatures above 59°F and the specifics of the system should be reviewed to ensure the cooling water SP for your floor will be raised in extreme cases to eliminate condensation. Hopefully the structure is designed with sufficient air flow at the floor level to make use of the cool floor surface, either from the AHU's or ceiling fans in every room. We can take any further discussions off line if you like what you hear and want to take it any further.

Otherwise, best of luck.

Water Buffalo-

There is no need to use 2 temperatures or hot water. The fan-coils can operate with 103F water as well. Just use two coils per AHU for heating and one for cooling (with lower temperature water than being fed through the floor system).

As another consideration- why not use return water from the AHUs to be the water for the floors. you could use, say, 120F water for heating yielding 100-105F return water and 45F water for cooling , yielding 60F return water. The return main from the AHUs could be the supply main for the floors, then the return from the floors could mesh into the return from the AHUs for a final total return connection.

By the way- if you are still looking for someone to help you put this all together, I am willing to donate my time- as long as you know that any designs will not be "stamped".

First- I am an engineer who has over 38 years of experience with ALL kinds of HVAC systems, plus I am a Certified Energy Manager.

Based on the relatively extensive zoning that is planned, and the cost of control valves- especially modulating- and their respective control systems, the best, lowest cost and lowest energy usage system is to use individual small circulating pumps for each zone and energize them with line-voltage thermostats.

The pumps will run when there is a call for heat or cooling. When they stop, the amount of residual heat/cool in the coil or floor loop will release lessening amounts of energy. A good anticipating thermostat will account for the residual energy transfer.

Since pump operation will ONLY occur when there is a load present, the pump energy will be the lowest possible- NO background energy use.

Would there be a valve with each pump?

In addition you would probably want a central plc to control the boiler, bypass setup. regulate the priority of the zones within the system & any other tricky functionalities that may be desired

I assume that those pumps have built in check valves?

Garth and Icarus-

Each pump is dedicated to one zone, so only a simple balancing valve (with shut-off capabilities) would be required.

Yes, there would need to be a check valve at the connection to the return header.

If "simple, low cost, effective" controls are the goal- it is still a house- the boiler could be controlled by either an outside air thermostat OR a secondary relay activated by any thermostat calling for heat. As stated earlier, each pump would be activated by the zone thermostat. Two solenoid valves- on each of hot and cold water mains- controlled by space thermostat could provide heating / cooling changeover for feed to the pump.

I agree. There are thousands of multi-zone hot water heating systems which function very well without valving.

___________________

CR4 Moderator: Post Modified Politics/Religion: Part of this post was deleted because it was overly religious or political. While each user is entitled to his or her own opinion on these topics, CR4 is not the place for discussion about them. Please review Section 14 of the CR4 Site FAQ and the CR4 Rules of Conduct.