The AHU should have had face & bypass with the HWV open 100% below 36* OAT and then modulate the F&B to keep discharge temp. We have this setup in NE Ohio and it works well.
But it sounds like you don't have that.
I think your problem is when the valve modulates towards closed, the outlet side of the coil is tripping the freezestat, (if it's a running trip).
Have you checked the water flow thru the coil? Is it at design?
Is the coil piped properly? (you'd be surprised at how many problem jobs we find that are piped backwards)
Is the airflow at design?
Adding a run-around pump may help, but it's expensive. If you can, I'd add a temp well sensor at the coil outlet and bring it into the control scheme as a low limit function.
Adding glycol may be a problem if the capacity is sized tight.
What is controlling the VFD?
Four wheel therapy, my 1968 Camaro is gone and will be missed
I assume your freeze stat is taking the unit down.
In a kitchen app, the colder the air the better is usually the norm for the cooks.
If you must run hot water to temper the air glycol is a must.
Another cheap way (from an energy perspective)is to install an air to air heat exchanger, get the oa @ 40 on those cold days and your golden.
Or Get one of those refracometer thingies and dump in the glycol.
Then take the freeze stat out.
That's the cheapest cost and fastest solution.
If you cant fix it right, try again.
I was just working on this exact problem today.
On a poorly designed system such as this, I will start the unit with a 120 deg supply air setpoint. Then sloooooly ramp the setpoint down to desired temp. I'm talking 0.1 deg every 10 seconds or so. This keeps the PID control from slamming the valve closed on start-up, and leaves a well tuned loop. (I fail the valve 100% open on a loss of airflow). This is easily accomplished with the numeric delay block in Tridium. One downside is the overheating at start-up, just call it "morning warm-up".
Of course, a properly designed system would have Face/bypass, and/or a coil pump, as well as glycol in the system.
Have you priced up glycol lately? 20 large for the last job I did. Still cheaper than a freeze-up though.
Originally Posted by toptech
Beer 30....right on
I agree with beer 30, we have similar issues running 100% outside air with no face and bypass. I also ramp the discharge air down slowly on start up. This causes your control valve to go wide open on start up as opposed to having those outside air dampers opening wide, making the end switch and starting the fan. Then you pull in freezing air that trips the freeze stat before the loop has a chance to respond to discharge air temp.
many engineers size units to "just to the job" so if you then factor a 10% loss of thermal eff. due to glycol and the unit wont make/meet design spec.
Originally Posted by hvacker
propper unit design with controls works fine..... If you can find the original design doc's I would bet you that the OA temp is falling out of the design range..... I see it all the time, if that is your case there is no Right fix. I suppose just the one that will make it work. If you go glycol your going to lose AT LEAST 10% heating/cooling capacity
Do you have glycol in the system?
As others have asked when is this tripping - during the off cycle/startup/other?
If the heating coil is undersized can you have electric pre-heaters installed? Enable them to operate when the OA gets below 30, but you should be able to lower that once you get you PID loop tuned correctly. gkstriper is right on with his comments. Leave the valve open for full flow through the coil and then slowly bring down your discharge air on startups.
I've controlled a very similar system - a fancy hotel kitchen - but the only difference was my heat was gas heat with duct heaters and the CHW coil had over 30% glycol, actually it was like 40% which was way overkill. I got rid of the mechanical freeze stat and programmed in my own freeze stat that will shut down the unit if the OAT and the DAT are below 35 and the supply fan proofed on after an 8 minute time delay. If it tripped someone would have to either go physically reset the unit with a reset button on the wall or log into the web server and reset it there. Also it will not reset if the OAT and the DAT is below 35 when the reset is pressed. Unless you have glycol in this system I would not recommend getting rid of the mechanical freeze stat. Too much liability unless you test and re-test your programming and know it'll work.
Another consideration would be to put some kind of baffles/air diverters around your sensing bulb/tube or on your OA intake. We've had poorly sealed dampers cause us nuisance trips during the off cycle and just diverting the airflow just a bit fixed the problem, or replacing the OA dampers with ones that sealed properly also fixed the problem.
"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/
Because of loss of heat transfer possibilities what we did in Northern Illinois was just take the glycol solution down to 5-10 deg above zero. If freezing conditions happened the solution would just get mushy and never a hard freeze. No damage.
Originally Posted by Vortech
Tracers work both ways.
If you don't have a 2-position preheat valve and face/bypass dampers, you need to tell the owner that they need to budget money to make it this way. The freezestat still needs to be downstream of the heating coil just in case you lose hot water. I'm surprised in that climate it wasn't done that way in the first place.
In northern Illinois we experenced a number of long power outages. Some all night long. No amount of controls will save a system from loss of power.
Tracers work both ways.
Now i Get it
Originally Posted by cwinches
This place Imho was over engineered.
Yes 100% make up for the hood only
If they balance the exhaust and make up correctly,
All you need is a couple(depending on size of kitchen) Of standard RTU's with CW\HW coils in them, and some economisers at Min pos to keep the space postive.
Sounds like they either didn't have room in the ceiling or didn't have the cash for return ductwork.
If you cant fix it right, try again.
Here is another thought, With square footage at a premium and all the value engineering these days, very, very often a proper mixing box and/or a push through fan setup (the ultimate mixing box) take up too much precious space and cost too much...so they often get axed ...so they can afford the pretty facades , and marble statues.