Bizarre VAV problem
Bizarre VAV problem
(OP)
I am having problems with a VAV terminal box serving a row of offices along the exterior of an office building. The VAV box discharges 83F air. The duct discharges from the box into about a 5 foot run of rectangular duct and into a dovetail wye fitting. The duct then runs each way over the office spaces. A 6 ft. length of round branch duct taps off each duct to a ceiing square cone diffuser for each office. The air discharging from the diffuser is 5 degrees lower than the discharge are from the VAV box. The temperature off the air at the entrance to the wye fitting averaged 81F. The duct is wrapped with 2" blanket insulation and vapor barrier. Branch ducts are insulated as well. The insulation is well sealed. I'm measuring the 5 degree drop out of a diffuser that is only 15 linear feet away from the VAV. The diffusers down the at the end of the duct trunk are discharging 76 degree air. Also, I've had the airflow measured at the diffusers and they match the minimum design settings. I'm absolutely baffled and would appreciate any insight or ideas as to what might be happening.





RE: Bizarre VAV problem
RE: Bizarre VAV problem
Like imok2 said your supply air volume may also be too low in the heating mode
RE: Bizarre VAV problem
Tony
RE: Bizarre VAV problem
The hot water flow to the box is metered by a Griswold valve. It's actually getting more water than the coil was designed for. Here are the design parameters:
min. cfm = 755
gpm = 1.06
EAT = 55F
LAT = 88.3F
EWT = 160F
LWT = 107.8
MBH = 27.15
Here are the field measured parameters:
cfm = 755
gpm = 1.5 (Note: auto-flow control valve. Pressure drop across valve was within manufacturers limits)
EAT = 57.8F
LAT = 81.0F
EWT = 166.2
LWT = 142.6
I examined the coil and it was clean. It was also hot from top to bottom.
Also, the ambient temperature around the duct is 71F.
Question: what's a better indicator of coil performance, water temperature drop or air temperature rise?
RE: Bizarre VAV problem
RE: Bizarre VAV problem
RE: Bizarre VAV problem
RE: Bizarre VAV problem
RE: Bizarre VAV problem
Tony
RE: Bizarre VAV problem
RE: Bizarre VAV problem
I'm surprised to see you recommending additional air flow to raise the delta T. Sure it will work. Throw more air at the coil and the delta T will rise to make up for the poor control valve performance. Pretty typical band-aid for a low delta T issue, and a waste of energy if implemented on a larger scale.
This Griswold valve is clearly passing more flow than it should to address the load. You should ask yourself why this is the case. At EAT 57.8, LAT 81.0, and CFM 755 the load is about 70% of design. With a clean coil, I expect the hot water delta T to be 75 deg F or higher given the EWT and LAT shown.
DouginMB, you may wish to trend a bit of the data so you can see what is happening over time. Check into a pressure independent modulating control valve (Flow Control Industries or Belimo) before settling for additional air flow and low delta T.
In addition, I would take a look at the central plant for the building. If condensing boilers are in use, you want much lower LWT in the system than the 142.6 deg F LWT you are getting at this coil. On the other hand, if conventional boilers are used, this may be about right.
RE: Bizarre VAV problem
The real problem I found so strange was how much heat was lost from the discharge of the VAV to the diffusers (5degF). I was hoping someone could give me some insight into that part of it as well.
RE: Bizarre VAV problem
The heating of air Btu/hr = CFM x 1.08 x (°F LAT -°F EAT)
The cooling of water = gpm x 500 x (°F LWT - °F EWT)
Accuracy, calibration of instruments & parallax in reading of instrument readout scales would be factors in the difference. The more accurate instrument in general should be more trustworthy.
RE: Bizarre VAV problem
2 Questions:
1) Is their a larger AHU upstream serving multiple VAV boxes? I am wondering if there is not significant variation in leaving air temperature off the hot water coil.
2) Were the temperatures measured at the same moment in time? Is it possible that the discharge air from the VAV was measured then fell before you measured temperature at the diffuser.
RE: Bizarre VAV problem
RE: Bizarre VAV problem
I would expect near steady-state conditions to eliminate any time lag. Which lead to the answer to Rhonda's question....
Yes there is a central air handler serving multiple boxes. The discharge air bounces around a little because the boiler is controlled on return water temperature and fires on and off. The water temperature can vary +/- 10 degrees, but at the time I was measuring, we had probes in the duct, pipe and diffuser neck and the water temperature was remianing fairly contant at design temp. The readings i took were instantaneous and were consistent using the same probes at each point.
Getting back to the water system, the T&B report does demonstrate consistently poor heat transfer across the entire system. And that's really aggravating. I really tried to design as a simple system as you can get: turn on the pumps and its self balancing. Go figure.
RE: Bizarre VAV problem
The VAV box hasn't been out of heating mode since the system was started-up before Christmas. The ambient temperature in the plenum was 71F. If it was considerably lower, I would be be concerned.
On the other hand, at minimum airlfow the duct area is considerably larger for the airflow. But, like I said, the duct is well insulated and sealed.
As far as sensor being out of calibration, we were using digital thermometers with probes. I was pulling the probe out of the discharge duct (83F), turning around on the ladder, and sticking it in the flex duct connection to the diffuser. I watched as the temperature dropped to 78F. The duct distance between the VAV and diffuser was about 12 ft. It was strange.
RE: Bizarre VAV problem
RE: Bizarre VAV problem
I agree with Rhonda up to a point but for practicle reasons if you need the heat then I still believe increasing the CFM is as good a solution as any for the time being. Rhonda, how did you determine the control valve was performing poorly?
AT this point I beginning to think the coil may be too small for the application even if the manfg says otherwise Sooo it seems to me you only have 2 options
1. increase the CFM or
2. increase the TD
Dougen I hope you did't buy these boxes on the cheap, please say you didn't.
RE: Bizarre VAV problem
RE: Bizarre VAV problem
Coils only set the delta T at design flow. In all other cases, control valves set the delta T. This coil is clean.
When you read DouginMB's specs for the coil, you will see that its design parameters are 52.2 deg F delta T on the hot water with 160 deg F EWT.
Actual measurements show that the control valve is overflowing so that only 23.6 deg F delta T is achieved while delivering about 19 MBH (70% of design).
Delta T should rise, not fall, at part load for the same air flow. I've now got projects delivering 80 to 100 delta T across the hot water coils designed for 50 deg F.
RE: Bizarre VAV problem
RE: Bizarre VAV problem
RE: Bizarre VAV problem
RE: Bizarre VAV problem
Rhonda is almost correct. There's a modulating three-way control valve on the discharge. When I took the above readings, the valve was manually forced open to allow full flow through the coil.
RE: Bizarre VAV problem
RE: Bizarre VAV problem
I believe that there may be markings on the cartridge that can be cross referenced by griswold.
RE: Bizarre VAV problem
RE: Bizarre VAV problem
Rhonda did I miss something? I didn't see anyone say that there was a 3-way valve on this coil? I always assumed it was a 2-way!. Oh I believe I have that stragity at least once in the pass few years or is this something new.Just kidding!
RE: Bizarre VAV problem
RE: Bizarre VAV problem
RE: Bizarre VAV problem
imok,
Piping connections appear to be made per manufacturer's recomendation. Griswold is piped correctly. There's an arrow cast into the valve body that indicates direction of flow.
RE: Bizarre VAV problem
in a 15 foot run ????
Is the duct ripped open to the ceiling plenum
somewhere on the duct run? Im assuming
this is a flex duct not hard ducted
to the diffuser. Try hard ducting the
15 foot run from the box to the diffuser.
By the way, isn't Griswold the family
name that they used on the National Lampoon
Comedy series with Chevy Chase ?
I'll bet the valve manufacturer didn't
appreciate that.
RE: Bizarre VAV problem
I appreciate everyone's responses. I'm getting the VAV box manufacturer involved and will take more field measurements. I'll keep you posted on what I find.
RE: Bizarre VAV problem
The other half of the problem is determining why the air temperature is dropping so quickly from the VAV dischage to the grilles. I found out that the building perimeter has a 2.5" gap where the structural steel rests on the framing beams. The architect didn't show any draftstopping of insulation on his plan to cover this gap so, in essence, the plenum is open to the outdoors. I imagine this is inducing a convective draft that is causing heat to be drawn off the ductwork. Insulated or not, the heat loss would be significant.
RE: Bizarre VAV problem
I glad you have sorted out a portion of the issues. Sometimes the most obvious things are right there in front of us.
I'd encourage you to address the low delta T and capacity issues head on and consider all the options. There are an awful lot of small fixes proposed out there that will only treat the systems and not the root cause of these issues.
This was a stimulating thread. Thanks.
RE: Bizarre VAV problem
RE: Bizarre VAV problem
Apparently 1-row coils were installed in every VAV box in the system (18 boxes in all). The submittals called for 2 and in some cases 3 row coils. That's the reason for the low delta-T. There's not enough coil surface area to get a good heat exchange and drop the coil water temperatures. I can't imagine low delta-T being a problem in a hot water system.
imok,
Yes, you did say that. That's one reason I got the manufacturer involved. Thanks for your input.
Once again I appreciate everyone's input. I've been in the business 10 years and have seen alot, but this was a new one.