Chuckle, yah gotta love those functional test guides by PECI.
Originally Posted by freddy-b
The one you linked to I read quite some time ago and some of the cautions and problems they mentioned were no surprise to me.
Demand based reset of duct pressure SP is not new, by far.
And as I said back in 2006 in another thread that touched on the subject, it can work and work well.
But it does have issues and concerns that need to be addressed.
The problem with so called "rogue" boxes, for instance. Which may be VAV boxes that were poorly set up (programming/configuration wise) to start with, or poorly chosen, or poorly installed. Or which may be experiencing some sort of malfunction.
The possible causes and reasons, are almost endless.
Just in the past few months I've had to troubleshoot some example issues.
On one site, new construction, a couple or 3 VAV's experienced so much flow resistance before their inlets that one had to run an excessively high duct pressure just to overcome that problem. It was a combination problem of some darn poor duct work to the VAV's in question ... coupled with poor selection of where to physically locate them.
In another case, a problem with clogged pitots serving the VAV controllers flow sensors.
In another case, the lead balancer was good. As that particular individual usually is. However, his helper on that job .... sucked.
In yet another case, 4 of the VAV's serving cooling only purposes, were undersized for the real load to be handled and unable to maintain setpoint. They operated full open all the time.
Etc, and so on.
All of which can be identified and handled. That's no sweat. But it does take a bit more diligence, attention to detail, monitoring ALL of the system over a period of time, setting up trends and historical data ... and then actually ... really ... looking all that over to find the signs of trouble.
I didn't bookmark it, wish I had, but not long ago ... not more than a couple weeks ... I read a final report published by an energy auditing firm. They been hired to audit 17 buildings owned by the folks who hired em. (A government agency IIRC) All the buildings employed DDC controls and multiple demand based reset schemes, for air, water, etc.
Out of 17, they found 11 had quite significant problems that were causing them to save no energy or minimal energy from their reset schemes.
The firm didn't fault the reset schemes themselves. They faulted the implementation (programming ... in many cases they identified control loops that were too simplistic, which did not do simple checks to flag "rogues", etc.), and the follow up. By follow up they meant a continuing, regular maintenance program to identify and fix malfunctioning components, do regular calibration checks, etc.
Duct pressure reset CAN work, and work well. If done well. And if one periodically does what one needs to do to check over the system and find developing problems. And there WILL be developing problems.
Balancing the whole system out, fine tuning the control loops so one doesn't get excessive hunting of damper motors, valve actuators, etc can be a bit more demanding and time consuming.
Personally, I LIKE it when a system is so tuned that yah have to look close to even tell if a damper or valve actuator ever moves. Or the VFD display shows <= +/- 1 hz variations for long periods.
The overloading of the network with data? I'm not sure that's a big issue. How often are yah polling or making adjustments? Too often is not good if you're trying to obtain a smoothly operating system. Fewer swings in temp or air flow translates out to fewer complaints from the occupants.
I'd think that a lot of folks hanging around this site ought to read some of those PECI Functional testing documents. There are a ton of them. And they're handy. One might learn something, or be reminded of something yah forgot.
If you keep a maximum static setpoint that is at the current static setpoint before you implemented the VAV damper control all of the VAV faults mentioned will have no more detrimental effect then if you didnt have the VAV dampers controlling the VFD.
How do you do your supply air temp reset now; what do you reset against, zone temp, box load, Ambient, Return Temp?
Originally Posted by freddy-b
Depends on the control system. peer to peer load being preferred if capable. if not then it defers to zone temp with weighted min max proxies.
Originally Posted by azzad
Return temp if boxes are pneumatic, and AHU is DDC.
Which part of it touches your concerns? Did you read this part?
"The intent of these control strategies is to reduce air handler capacity in ways that improve energy efficiency"
and this part?
"Resetting duct static pressure keeps the duct static pressure only as high as is needed to satisfy the neediest zone, which saves fan energy. "
Or this part?
"Often, the controlling static pressure setpoint (duct or discharge) is automatically reset up or down based on zone loads to minimize supply fan energy and still ensure proper air flow through the VAV boxes."
Note the last part - "and still ensure proper air flow through the VAV boxes" - read "does not mess up the air balance"
I read the caution on not running a SAT reset schedule in parallel with a SAP reset schedule, and I agree with that.
Nikko- I never doubted it saved energy. There is just way more to it than a couple lines of code and 5 minutes.
Originally Posted by nikko
The balancing issue is beyond the unit it self....its the total building dynamic envelope.
Also, back to the document... I guess you missed the 11 out of 17 studies were jacked up?
Originally Posted by azzad
I'm glad yah told me because otherwise I probably would never have figured that out. Thanks.
Putting sarcasm aside, as such really needs to be put to rest since it benefits no one.
I believe you missed my point.
First off, I do believe I've stated numerous times in various posts that demand based duct pressure reset schemes CAN work, and work well. If properly implemented and executed, certain precautions are observed, and if one does the work to ensure everything is calibrated and operating properly.
I've never debated otherwise.
Now, I haven't a single clue how things go in your work world/area. Or what sort of customers you deal with.
But in mine, if I tell a customer that I've implemented such and such a control scheme ... they have this darn ridiculous, unreasonable attitude about such things.
And that is that it had better work, just EXACTLY like I said it would, all the time. Or they're gonna tend to think I'm an idiot, or a hack, or some fast talking used car salesman more interested in the greatest possible profit margin and getting that check cashed as fast as possible than I am in making sure my customer got what he or she considered a good value for the price paid.
Doesn't matter if that control scheme was spec'd or I gave it away as a freebie.
A demand based duct pressure reset scheme that's acting just like an constant SP scheme ... is worthless.
No better than if that used car salesman offered to sweeten the deal and make yah feel better about it by tossing in a CD player ... for free ... just for you to find out later that the darn CD player doesn't work.
Even if that car yah just bought works okay, you're gonna have a sour taste in your mouth and be wondering if maybe yah just been had every time you stare at that CD player in the dash.
As concerns the customers I deal with, if I say I've included such and such a control scheme, which will achieve such and such an end goal ... believe me, they WILL check it out with expectations that its working just as I said it would. PERIOD. Or they expect I'll make good on it and fix whatever to make it work as I said it would.
Guess what? I'll make it good. No matter what it takes. That's how things are done here. If one expects to stay in business for long.
Part of my entering this whole debate/thread centered around the fact that several folks making comments like:
"If it saves the customer some money great, if it doesn't then I wasted 15 minutes of my time."
"Seriously, this is 15 minutes of code - it's NOT a big deal. worst case is the setpoint stays constant and you've not gained anything."
"It takes 3-4 more lines in line programming or 6-7 more blocks in graphical languages, depends on controllers."
Hmmmm. Maybe in somebody else's world. Not mine.
Coding a demand based duct pressure reset routine, in and of itself, is NOT a big deal. In truth, I can't think of much in the BAS world, when it comes to control programs within controllers, that's just all that challenging. Programming skills-wise, its all pretty simple and basic stuff.
And once one has worked out and debugged a good subroutine or loop, thereafter whenever you come up with a need to do the same thing again, it is just a matter of cut and paste ... with maybe a little customization thrown in. No big deal.
Easy as can be. I know. If I can do it, most anyone can. I'm far from the brightest apple ... on any tree.
However, just tossing together a control loop that works is just the beginning.
The true indicator of a skilled programmer, and good code, is its ability to recognize and adequately handle exceptions, errors, and faults.
In the example you mention ... if your control scheme does not include some method to identify, flag, and call attention to a fault condition, and possibly auto-correct for it (not necessary but a nice touch) ... then IMHO your job isn't done.
Now, we could debate for a long time the best way to do such a thing, and everyone will have his/her own opinion. But it really doesn't matter. You could add code to auto-identify out of bounds or abnormal conditions and issue an alarm or warning. Or simply set up trends ad spend some time with the customer to teach him what to look for and how to interpret the info. Whatever. Take your pick or come up with another solution.
But besides the programming issue, before yah can tell the customer, "Okay, there yah have it ... its working and doing this or that." ... one needs to verify that indeed all is well and working as you just stated it to be.
In the case of a demand based duct static reset scheme, I'd take that to mean that not only does the program work correctly and as expected ... but that you've also done the extra work to ensure that you've at least identified (if not corrected) the pitfalls of such as system.
For instance, in a recent case, new construction. There are 3 cooling only VAV boxes that will NEVER go below 100% air flow. Not and maintain the spec'd space temp SP. They're simply too small for the load. I identified them, and the cause. Alerted the customer to the issue. Those 3 VAVs are excluded from the reset scheme. AND, I checked with the manufacturer, as well as the installing contractor, as concerns what might be a REAL maximum temperature that the special equipment in those rooms might be able to put up with long term (as versus what the building HVAC designer spec'd) and changed the temp SP's and alarming for those rooms to reflect the new, higher SP. In another couple cases bad duct layout and installation was the problem, caused those particular VAVs to need a higher than "should be" duct static pressure in order to meet design air flows and I was able to prove my point and customer got the mechanical guys to fix the issue.
End result ... customer was satisfied, as well as the third party inspector/commissioning agent that the control scheme actually, really ... no sh*t, was working as advertised. Current issues were identified, known, and compensated for. A means to identify new "problem" children was in place (a separate program running within the front end ... a Tridium box that collects trending data and flags suspected out of bounds/abnormal conditions. Specifically looking for VAV malfunctions)
Again, no big deal. But to do it right ... did that take extra work and time? You betcha.
Of course, I am almost certainly just a dummy. Personally I'm not bright enough to figure out how to do all of that in just 15 minutes, with just 3 or 4 more lines of code.
Maybe I could just toss in an extra, standard PID loop? But no, I've tried that, and the results sucked ... by my standards and expectations. Instead I use a variation of the "Trim and Respond" type control loop. Much easier to tweak and tune, faster recovery from a zone going above or below SP, and more adaptable to setting it up so that it'll make only gradual changes, slowly so as to keep the whole system stable and smoothly operating.
Maybe I'm anal, maybe not. I noted that someone, I forget whom, said he placed energy savings above occupant comfort in many if not most cases. Maybe he can get away with that where he is. I can not. I've got anal customers in that they expect to have their cake and eat it, too. They want energy efficiency, WITHOUT sacrificing comfort. And if yah try to tell em you can't do that ... they start wondering if maybe they should hire someone else ... who actually knows what he's doing.
To me this whole thing is like driving across town during rush hour too save a dollar on a Big Mac...when the McDonalds right next door it is full price.
When its all said and done is that dollar savings worth the time, gas, and aggravation?
I say no.
If you just look at my bank statement and nothing else...WTF I could of saved that dollar. What a idiot I must be?
Did you read this part, which comes BEFORE the parts you quoted?
Originally Posted by nikko
"To ensure the individual reset strategies function as intended; and "
"To minimize negative interaction between the two reset strategies"
Nikko, I seriously doubt that Freddie contests the fact that demand based duct pressure reset strategies CAN work and do what they're touted to do.
And I'm pretty sure one doesn't need to give him 1st year apprentice level lectures about how such a scheme might save energy, reduce system noise, etc.
I don't know Freddie from the man-in-the-moon, except for what he's posted to this site which I've bothered to read. And I don't read every post or thread, just the ones' whose subject lines seem to be about something I'm interested in. However, my general impression of Freddie is that while he can sometimes be excessively .... blunt ... and short of patience, he does seem to know quite a bit worth knowing. Far more than is the case with myself. And he generally has a valid point he's attempting to make.
So far, as concerns what I've read of his comments in this thread, his concerns appear to be more along the lines of:
(1) He looking for someone to quantify, in real terms ... not just assertions, not just results of computer modeling in labs, not just sales pitches, not just best case results under ideal conditions, etc what sort of energy savings one might expect from such a control scheme.
(2) I think he doubts that it's just a matter of an extra 15 minutes and 3-4 lines of code that some have asserted. I have the same doubts. I'd have to see it to believe it. I've implemented such schemes. Took me a darn lot more than 15 minutes and more than 3-4 lines of code ... to get it done RIGHT.
(3) Doesn't matter if one gives the code and implementation away free or charges for it ... if the customer finds something wrong later or is unhappy with the results for whatever reason ... yah gotta go back and get it all straightened out and make it right. Besides, who is fooling whom? You're NOT giving it away free. There are no freebies. Not if one intends to stay in business. If in the end of the project you realize a profit ... you gave away nothing for FREE.
Geez ... I've had that conversation with any number of car salesmen. If they don't understand my point, I walk away and find a more honest salesman who isn't gonna BS me.
When buying a car, the deal we make is for the WHOLE car, the final product. His profit is whatever we mutually agreed upon as fair ... for the whole car. So even if he SAYS he's tossing the whiz-bang, super-duper, it'll blow your mind away music system in that car for FREE ... if it doesn't work you can bet I'm bringing it back to have him fix it. If he didn't charge enough mark-up on the vehicle to cover such a situation ... shame on him. I'm betting he did, however.
(4) While everyone touts the PROS of doing demand based duct pressure reset schemes ... what are the CONS? What are the possible issues that might come back and bite one in the a**? What does one need to beware of, look out for, avoid, take into account and compensate for, etc?
Probably the reason he was interested in that Functional Testing Procedure. It tells him SOMETHING about what previous system testers have found that resulted in less than proper and adequate operation.
That's more than what I've read in many of the replies to him, so far. "Yeah, easy as pie, takes just minutes, no worries, no drawbacks or pitfalls, just DO IT or you're an ignorant fool who can't understand the simplest of concepts."
Or so it seems as if that's what folks are saying.
Now, just some observations on my part.
I've read some duct pressure SP's in this thread that seem excessively high to me. In my experience. Based upon where I live and work. People talking about 2", 1.5", even 1.3" as normal SP's for systems not utilizing reset. Huh? For the systems I see, retro-work or new, 1.3"WC would seem on the high side. 1.5"WC would be in the WTF is going on category and I'd be searching high and low for problems. Typically, without duct pressure reset, I see fixed SP's of .8 to 1.0"WC. What the heck are you guys doing wherever you're at? Or perhaps I should restate that, what the heck are the system designers, installers, and balancers doing? Is this right? Is it normal elsewhere other than where I live?
Then there is some guy who displayed data in a table, speaking about his reset scheme. I looked and the first thing that jumped out at me was "Whoa, look at all those deviations from SP for room temps ! Do they let him get away with that much where he lives and works?" Egads, around here, with the customers I have, I'd get hammered.
Now lets look at some of the references you provided. I did read them. Although I have read at least one of em before.
From the link you gave:
Now, okay, its supposedly from the CA government. Fine. Meaningless, but fine. Reads pretty much like any other sales brochure I've ever read. Somebody in the CA government getting a kickback under the table? If not, he or she should be. Pretty decent piece of sales marketing if yah ask me, the folks making the InCITE system are getting free advertising ... at taxpayer expense?
I see that some of their data is based upon short term "proof" with long term data extrapolated as versus real results shown. Okay, that's fine. But PROOF of nothing in particular except it is a system worth considering.
Yep, the number 30% is tossed out there as a claimed energy savings. I've no reason to doubt it might be realistic. Given THAT particular building, given the precise ambient weather conditions during the period of testing, etc.
Likewise, in a sidebar, they proudly tout a 42% savings for a one month period at some other building.
Hmmmm. Yah know, if I get to specify such a short period of testing, under average weather conditions I get to select ... I'd bet I could show that exact system in the same building producing a heck of a lot less savings.
I kinda think people like Freddie has heard and read all this sort of exaggerated hype before, and was looking for, hoping for ... some info that was a bit more concrete, unbiased, and realistic.
The next document you showed a link for. From the ever lovable and all knowing Taylor Engineering. How can one ever doubt ANYTHING these guys tout? Their name pops up anywhere and everywhere ... so they must be all knowing, right?
Hmmm. One of the first things I note is that they don't really suggest using PID loops in one's control strategy for demand based duct pressure reset. They say it CAN be done, but most times doesn't work all that well.
I'd have to agree with them there. That's my experience.
As one of their points, they mention that the "Trim and Respond" logic they suggest to be used, is able to ignore the first 2 pressure requests before responding. Thus eliminating or at least reducing the odds that a rogue VAV or two might defeat one's whole strategy. Commendable. Because that is in fact a problem and an issue one should address. Of course, they then state that this COULD cause comfort complaints. Again, true and reasonable.
No arguments from me. There are of course problems with this reset scheme. Which can be overcome. But one should be aware of them and plan accordingly.
They go on to say, quite reasonably and accurately IMHO,
"Tuning by trial and error after installation is almost always
required. Every system and application is different and few
will work well with the parameters listed in the example Trim & Respond logic above. As previously noted, the reset must be relatively slow or a cyclic instability results.
The primary way to test stability is to plot pressure setpoint and pressure requests over time.
Another way to test if the reset is working is to plot the actual static pressure versus static pressure setpoint, as shown in Figure 4. This plot is particularly useful in showing if rogue zones are limiting the range of the reset."
Wonderful ! And quite true.
But all that looks to me as if it might be just a tad bit more time consuming than 15 minutes. Which would be in line with my own real world experience.
They then go on to discuss SOME of the causes of Rogue zones (there are many more I've encountered) and possible ways to deal with them.
Okay, I've gotta admit, can't find much in that article to pick a nit with.
I doubt SERIOUSLY ... if you'd ever get an experienced engineer from Taylor to ever claim 15 minutes and 3-4 lines of code are all that's needed to implement such a scheme adequately and well.
The Trane article you pointed at:
First off, I have trouble determining if the info and data provided comes from REAL installations, or they're simply simulations. From the way they word things, I can't be sure.
But let's just go with the data they provide.
They are at least honest by stating this up front:
"System operating energy cost saving potential is of course, dependent on the building location, construction and operating profile, e.g., number of hours in cooling, heating, etc."
Bravo ... good for Trane.
But then they waffle a bit. First stating that possible fan power savings ranged from 5% in Miami to 27% in Seattle. Looking at their charts, it seems that in this case they're comparing it to a system set up to maintain a fixed duct static SP based on a sensor 2/3rds of the way down the longest duct/duct with most demand.
Which is previously acknowledged to be more efficient than is the case with the pressure sensor installed at the fan discharge.
Then they go on to use even bigger numbers, citing 19% savings for Miami, and 40% for Seattle. But that by comparing performance against systems with the pressure sensor installed at the fan discharge, already admitted to be the worst scheme.
Does anybody even install SA pressure controls using a sensor AT the fan discharge any more? I can't even remember the last time I saw that. But maybe its a common practice elsewhere? <Shrug> I don't know.
Okay, they waffle a little there.
Nicely they give figures for how much of the TOTAL HVAC systems energy those fan savings amount to, and cite 2% to 4%. For Miami as compared to Seattle? I guess.
Interestingly I note they didn't brag about performance in Minneapolis, one of the examples they show. For Minneapolis (I live in Minnesota) they indicate 18% possible fan energy savings, or what amounts to an estimated 1.2% of the total HVAC energy costs.
Okay, can't fault their findings. What they show is more or less in line with other studies I've seen done on REAL buildings, over the long term.
I wish I'd saved the bookmark, but didn't. But some time back I was reading a report by some engineers in Hong Kong. Who'd tested several buildings. With the tests running a year long. Keeping in mind they tested, tuned, and tweaked all first to make sure things worked as they should. Their results showed that fan energy savings varied quite a bit, depending on the building, climate for the time of year, etc. With months where they might only see a couple percentage points, to months where they would see maybe 30+%.
Their final evaluation was that they were seeing an average, over a year, of 15+%. Of fan energy savings. They did not quote any info about what percentage of total HVAC energy costs that might represent.
But I never argued that one could not save energy using the demand based duct static reset schemes.
I had a lot of doubts about savings numbers being tossed out, and about the amount of time and work involved in making sure the scheme was PROPERLY implemented.
Originally Posted by nikko
I've done mainly retrofits, design build and some plan spec. I always start out with 1.5" w.c on my high side and I bump down from there if I can. Yes it takes me longer than 15 minutes to work with both resets, but then again we typically don't do plan and spec and if we do we typically maintain some kind of maintenance contract after to pay for our time. Granted I am able to bump the static down typically, but on numerous occasions, at least my experience up north, the couple of weeks of 100 deg weather may need the 1.5" setpoint and now that I am south they definitely have needed it and actually have had to use 1.75" on the last one. I try to overshoot initially and then bump down until I see a problem. Let things settle out to find the problem zones, account for those and then start tweaking until there is a problem and then bump it back. Yes good thing for a maintenance contract otherwise I couldn't justify it, short of doing it for my own pleasure or having it sold in the contract initially, which hasn't been done, yet.
Yes both resets can have an adverse affect; however if you start them out around the extremes, like up north when it stays below 35 deg for a couple of months you can reset both your supply temp and static on the low end. Easy enough to tweak an economizer damper vs. having to worry about running a chiller/DX etc. I do see this strategy being a problem on large buildings with the core load being less likely to change on OAT conditions. For me yes it took alot of time and was it worth it? To me yes, I learned alot about PID's/bumps, system operation, and where various systems fell on their face. To some of the customers yes, and yet to others I am sure some have gone back and gotten rid of the reset schedule or set things up to minimize it's affects to account for rooms that have now become server rooms with no thought of having the hvac accounted for.
To each their own but I am willing to try things to save energy $ and hopefully make me a better programmer. Its bit me numerous times as well, but good thing for the internet and remote access! Now if I was working plan spec where it was get in, get done and get out would I implement it, probably not, short of high/low extreme settings only, but even that would be minimal. The tweaking time and/or potential for unaccounted zones and the drastic measures I've had to go to just to prove it was not the controls that was 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
OK - work with me on the balancing part, I'm not understanding how static reset can affect balance.
Originally Posted by freddy-b
True VAV is pressure independent and the individual zone airflow requirements are met on a per-box basis and are determined by a measured flow rate (calculated from the pitot cross or ??). The primary damper opens or closes to achieve that rate (regardless of the calculated rate request). So - if a box is requesting 500 cfm of primary air and the duct static pressure behind it is high, that box will have a damper position of X. But if the same box is requesting the same 500 cfm of primary air and the duct static pressure behind the box is lower, the damper position will simply stroke open beyond X until the requested rate is achieved.
Either way, the same amount of primary air is being introduced to the VAV box. How is air balancing adversely affected?
As for the "building envelope", I assume you're talking about building static pressure? If so - the static pressure reset scheme has absolutely nothing to do with building static pressure. Building static pressure is a result of the relationship between outdoor air being brought into the building, and total air exhausted from the building and this loop is completely independent of anything going on in the duct.
am I missing something?
The time it takes to properly test and verify the system is just not there on plan and spec jobs. I do mostly plan and spec where hours are so tight you barely have enough time to do what you need. To really get the benefit of this type of reset takes time to trend and adjust under various conditions something that we never have time to do.The time to actually write the code is not the true issue. It is all the time needed to make sure it works, if you just dump it and walk away then the time was wasted and the customer in most cases was better off without it. Typically in my experience we barely set foot on the job after the balancing is complete. How can you properly tune this type of system in the balancing is not complete? I would love it if we had the time to spend doing a complete system tune and check out but it is just not there on these jobs anymore. With the Tracer Summit system it is simple case of just set up the VAS but I find months later when we get a call that even that does not always work right and the owner more than often has us turn it off to make the building comfortable.
Originally Posted by osiyo
I agree with Freddy I know people who will drive all over town to save 2 cents on a gallon of gas just does not make sense.
Not everyone is blessed with automated building static control. Not everyone has a bad ass control system...not everyone has perfect engineering behind them.
Originally Posted by nikko
Anyway, I Still see many barometric dampers out there that are balanced/set manually. Do you see this as being a non issue?
Just so you know...I rarely get to job sites in a install/Commission capacity. I am more of a "mob fixer" type dealing with problems that get overlooked or just flat out were attempted to be hidden. Its not my first rodeo.