static pressure in a common supply duct
Got a system that has two 4-ton cooling systems that have separate ducted returns and which have a common supply.
The controllers are networked together currently, although they can be made to operate as stand-alone. The problem has been an extended history of intermittent low pressure alarms. The networking can manage up to six total units...N +1, with the last unit in the series being the backup unit. The master controller can also use stages in other units for multi-unit sequencing and/or capacity assist.
In this case, the "backup" unit is always operating in the capacity assist/sequencing mode because there is a full load of heat.
In the interest of assuring the best solution, I am going to test the static pressures and clock the blowers today, since it is my opinion that the heat load in the room is nearly maxxed.
Meanwhile, I made a change or two in the programming--I made both deadbands to be 2.0 degrees and I eliminated the changeover to backup units when certain alarms occur(specifically the Low Pressure condition). That way, the other circuit in the problem machine should continue to operate and of course, the blower whereas they were being shutdown.
When I test the static pressures, should I test individual supply ducts and then also take a reading after the two supply ducts join? Which of those two readings will best reflect the design pressure? I'm planning on contacting the manufacturer to see if they even are aware of the supply duct as built.
Hint--the issue is not a lack of refrigerant in the circuit. It could still be an intermittent wiring problem, but no amount of control wire-wiggling or wire-nut wrestling has caused a blip yet... And just as an aside, I disabled the pumpdown feature. It can bang on against a full load of refrigerant for all I care (yes the CCHs are working).
The air testing I'm doing is just to firm up the design/as built harmony, as I'm also looking ahead to the possibility of actual backup (sale$) cooling. But it would be nice if first I could solve this problem...
Question : Is duct work sized for the 2 units (8ton total) if not you could get this LP alarm
Would a ductilator help answer? Not being a duct guru, I'll have to put a tape on the duct to give a better answer. ..but I think it has 8 or 10
Originally Posted by jtricor
8" takeoffs roughly.
Some times the unit runs for several months with no problems. Other times it alarms every other day for a week. The deadbands are now both at 1.0 C. Normally I only do that in surgical centers, but this brand name has a minimum time off bottomed out at 3 minutes anyway.
so just to get this straight, you have one duct system with two airhandlers that sometimes run simultaneously and sometimes individually?
I would have to assume that the common supply duct is sized for both units operating simultaneously. That being the case I would test airflow with both units running as if there is an issue with the duct it would most likely show up under this condition.
An easy way to get a close total air volume would be to take a velometer to each of the return grilles. Multiple readings, average the velocity and look at the grille data.
The most accurate way would be to traverse the supply duct with a pitot tube and manometer. Again you take several readings inside the duct, average, do some math to arrive at total air volume. You want to find a straight section of duct before any takeoffs and away from turns so the velocity profile is fairly uniform.
You say the refrigerant charge looks good with both units operating? Normally if you have an issue with low airflow the txv’s will just close down to compensate resulting in low suction.
Typically air flow is not an intermittent thing. The separate return ducts is a good strategy on twinned units as it helps to eliminate air from one air handler back feeding through the other. But unless both fans are operating at all times then there will be some amount of back feeding, depends on how the supply ducts are tied together. You may want to consider check valves in the outlet of each air handler to address this.
I think you may be on the right track; twinning units together can be tricky. I hope the blowers aren’t ECM motors?
The cfm checkout would be a great start.
What about TD's? What is happening to your return temp? Do you have supplies blowing right back into the return?
Since you only said 4 tons each I doubt it but does this system have an economizer adding to the mix? If so what is your mixed air temps?
Since your changing the programming possibly you have or can add trends to get a better idea of RAT and SAT?
Is there anyway you could have an indoor fan starting backwards? Is there barometric dampers or means to prevent this from happening.
Also I would recommend re-enabling the pump down feature, it's added compressor protection that is worth leaving in-line.
"How it can be considered "Open" is beyond me. Calling it "voyeur-ed" would be more accurate." pka LeroyMac, SkyIsBlue, fka Freddy-B, Mongo, IndyBlue
BIG Government = More Dependents
"Any 'standard' would be great if it didn't get bastardised by corporate self interest." MatrixTransform
- not my website.
Versatile Hunting Dog Federation - www.vhdf.org/
The units always run simultaneously, unless one goes into alarm. Then that one would shut down entirely. I'm presuming that is because of the N+1 approach that the master controller uses when there is more than one unit, and when capacity assist/sequencing is selected as a programmable option.
The system operated this way for 3 years until 2 days ago, when I eliminated the "changeover alarm" of low pressure circuit 1.
The tech support could not explain to me why one low pressure alarm on one circuit of a two circuit unit would cause it to shut down entirely UNTIL he was reminded about "changeover alarms". He also didn't know that the literature on the controllers describe the "one time pump down" as being built into the circuitry/software.
So I eliminated the changeover alarm hoping that the blower and the 2nd circuit will stay online, the alarm dummy light will cause the user to call us, and we have a minute to get out there
when they reset the alarms, this particular manufacturer has provided a "clear" option and the alarm condition is gone, history, never to be seen again. The time and date is there, and you could then tell that SOMETHING happened, but not the specific alarm.
I mean, am I asking a lot for a less volatile memory than that??? What am I supposed to do, police how the end user resets alarms?!?
Anyway, they were both set up to run always.
The superheats in 3 of the 4 TXVs went down to 0-4 degrees at any given vapor pressure, and no adjustment stuck for more than a day or two. Yes, we had expensive icemakers. The rebuilt txvs were set to 8-12, although the manufacturer ELSEWHERE in their literature also called for 12-20.
Wouldn't that run even hotter evaporator temp?
Now it came to my attention that they have very recently had additional 4plug electrical outlets installed in this lab.
Doesn't that strike you as a class one indicator of additional heat load?
While this does nothing for the low pressure problem, which is probably a simple mechanical issue with wiring or a component, it tells me that the dynamics in the room may be changing.
We are also humidifying most of the time in the winter, so there's another bit of heat to the mix.
There is no economising at all. It's in a sealed laboratory.
The TD is better than 15--shooting 80 at the return grill and 64 at the supply.
I'd like to get some numbers from the ducts as far as pressures, and specific TDs, as well as provide some current superheats. Also, I intend to ask ATS (I'd say oops, but at this point I don't care) what kind of subcooling they want. When they were controlling the outdoor fan with a two-wire switch, they used to say "if the sight glass is clear, it's overcharged."
This system is operating with a p66, and I reduced the charge by 15-20 percent and still have clear sight glasses, with liquid line temps of 85-95 depending on which way Saturn aligns with the light fantastic.
Stranded in ATS territory...