I need to take the pressure output on a pump and turn it into a GPM flow.
I have a good DP sensor, and a pump curve from the pump... Looking for the easiest way to code this to make the curve read nice and seamless and neat. I will be working in a Niagara wiresheet. I have a good grasp of the mechanical side, the issue is making a reading convert when the translation is not linear. So scaling with a reset block will not give a good result.
I usually spend a little time head vs wall before posting but I am in a hurry. Sorry if the answer is super simple.
Yes - I know the result is not as good as a properly mounted flow meter - i am working with what I have.
That is what I ended up doing... I was hoping to map the DP input and use a tabulated proxy extension like a custom sensor... but NRIO will not let me map the same point twice like Bacnet will...
And I needed the DP input on its own.
So reset blocks it is... not great but better than nothing
I would use chiller design pressure drop. The manufacturer knows what that should be. When doing TAB I compared chiller DP to the pump readings. Pump measurements need to include shutoff head readings to verify impeller size.
No man can be both ignorant and free.
Thomas Jefferson
Normally put the DP across the chiller barrel and control pumps off that vs the other way round. Much easier to tell the chiller rep, sorry your DPs are met vs the pump curve tells us so. Works with pumps on a common header multi machine as well.
Normally put the DP across the chiller barrel and control pumps off that vs the other way round. Much easier to tell the chiller rep, sorry your DPs are met vs the pump curve tells us so. Works with pumps on a common header multi machine as well.
When doing design that's the way I did it. The rep and chiller manufacturers loved it.
No man can be both ignorant and free.
Thomas Jefferson
In 20 years doing TAB I never had a chiller that didn't perform. You are correct though orion the chiller guy doesn't have any place to turn as long as you can verify the accuracy of your readings.
I did have to fight a couple of design engineers that didn't understand that chiller DP tracks flow a lot better than flow meters especially those with small turn down ratios.
No man can be both ignorant and free.
Thomas Jefferson
In 20 years doing TAB I never had a chiller that didn't perform.
To design or their equipment submittal?
Prob right. Had dumpster fire designs and delivered machines that equaled train wreck. Doesn't seem to be much pre-bid work to point out crap shows, its deal with it after the machine ships in most cases. No ship before payment. Funny enough support dies off with payment in most cases.
Guess the moral of the story is when the $hit hits the fan, the chiller rep is looking for DP or flow measurements on their machine. Minus that, its going to be a gong show deflecting the blame quickly without sucking endless time. Extra DPs are pretty cheap in a major chiller plant project end of the day.
Yes I know. Remember, fixed speed pump dedicated to a chiller barrel... the flow is whatever it is, nothing I can do about it, and it has worked great for years. It is not controllable.... unless I wanted to do something weird like stroke the ISOs a little to throttle the pump. This is a retrofit, replacing NAEs with JACEs.
The fun details I could not get to earlier.... the site actually has 2 flow meters, one on the CHW main, one on the CW main. Perfect! But the 1992 Johnson prints did not show it. And these prints were the basis of design. So there was discussion on installing new meters (since the site does not have any.... according to the prints) between the engineering company and the large corporate customer that barely included the onsite staff. The onsite staff only hears 'we wanna put in flow meters - so we will need to install x feet of straight pipe and we will need to shut down the chiller plant for a day or 5.'
'Woah woah, this is a data center, you gonna truck in a temp chiller for this?' Says the onsite staff. The question - 'do you have flow meters already?' never gets asked.
'Uh..... maybe we will just get it from the bacnet integration' says the engineering company and corporate customer, not wanting to spring the extra cash.
I come along 3 mos later, job is booked. I do what I usually do, mistrust the prints entirely, and go through the existing automation system and re-engineer everything so nothing is missed, and we can get some change orders. Maybe 20% of the site was missing or misengineered - as it was based on the 25 year old Johnson prints that sucked when they were new. I find flow meters in the chiller plant and the PM says 'they are being abandoned.... don't even pull wire to them'.
Well aright, whatever. I asked, he is sure - no explanation. Now 4 mos later job is done and a week before commissioning I get asked for the Bacnet integration for flows.
'Say what now? You will never get flows out of a chiller integration. The chiller only knows he has enough to run, they are not going to have the onboard stuff to calculate flows. Maybe you shuda asked me about that when I was redoing the prints.'
Now it is a fire that needs putting out. I finally get briefed on the the backstory. But we do not have time to pull wire to the flow meters, as-built the prints, submit it to the commissioning agent and get it back in time for commissioning in 4 days....
So I get to provide a reasonable number using the DP on the chiller for informational purposes. Fortunately there are already DPs on every chiller barrel.
The best part was when the PM, felling bad about the last minute mess he handed me asks 'you know the equation for calculating the flow? I can get it for you.' And he sends me a link from an add filled, residential construction website that has a sheet showing the pressure drop of water in PVC pipe.
Jeez, where to start....
I got it buddy, thanks for trying. At least he means well.
We had a retro-commissioning team who measured the dP across the chiller pump reported that the flow was at 'design'. When I measured the dP across the chiller evap barrel and used the factory performance data I got crap flow (using calibrated digital pressure gauges - don't use the analog gauges). Informed the AC staff to check the piping and strainers. Week later the flow was at design - they must of found something. I measured the pump dP before the AC staff checkout - using the pump curve it appeared that the flow was 'ok'. Tricky - need accurate pressure sensors. The pump is like a 'battery'. Trying to measure the pump flow (gpm) is like determining amps from the voltage drop of car battery's 'amp-voltage' curve. Very small voltage changes.
Better to use the pressure drop across the chiller evaporator barrel (IMHO). See attached PDF
There is no excuse for the mess you were handed numbawunfela.
Eh... I do agree. The nice thing is I been at places where the response is to conceal it, pretend it didn't happen, lie about it and shame you for knowing about it.
Here they own it, apologize, buy me lunch occasionally, and thank me generously and regularly when I save their bacon. For me, crap happens everywhere - so is there sanity to see it as such? Or is the twilight zone weirdness in the workplace? Here, there is sanity.... so the crap happens.
Me? I make good money and I get paid by the hour. So when that 8 hr job goes into OT and it is clear I am the hero who made completion even possible because someone up the food chain messed up... I have a fat paycheck, which is nice. The cost of that same paycheck serves as incentive to whomever to try better next time... or to someone else to get rid of that guy because he is costing more than he is making.
I tend to be pretty zen. Usually.
But I appreciate the sentiment.
Eh... I do agree. The nice thing is I been at places where the response is to conceal it, pretend it didn't happen, lie about it and shame you for knowing about it.
Here they own it, apologize, buy me lunch occasionally, and thank me generously and regularly when I save their bacon. For me, crap happens everywhere - so is there sanity to see it as such? Or is the twilight zone weirdness in the workplace? Here, there is sanity.... so the crap happens.
Me? I make good money and I get paid by the hour. So when that 8 hr job goes into OT and it is clear I am the hero who made completion even possible because someone up the food chain messed up... I have a fat paycheck, which is nice. The cost of that same paycheck serves as incentive to whomever to try better next time... or to someone else to get rid of that guy because he is costing more than he is making.
I tend to be pretty zen. Usually.
But I appreciate the sentiment.
You are so right. I was attacked many times simply because I knew things weren't right and insisted on everything set right. I never understood that and never will. Even the owners in some cases were pissed because I held up completion of the project.
No man can be both ignorant and free.
Thomas Jefferson
Reply
EFC Committee/Professional Member*
Professional Member
jhawk liked this post.
Originally Posted by orion242
