Booster Pump Followup
Subject: Booster Pump Followup
From: Steven Hoagland <[email protected]> Date: 11/17/2015, 1:16 PM
To: Marsha Atkinson, Rossi Snipper, Adam Tolou-Shams, Dee Jacobson, Mary Pynenburg
Dear Colleagues,
To begin with, my technical background includes tuning hundreds of PID controllers. This component was identified by Erick as one factor in the pressure fluctuation problem which a new booster pump created. So I'm at least somewhat familiar with the language and concept of such devices, which are used extensively in chemical engineering research at UC Berkeley, my former employer.
I called California Motor Controls (CMC) on Friday, the guys who manufactured the new booster pump controller, and asked for an opinion about our current situation. In particular I was interested in data logging for the new controller (Pump Vision PB570). Comments of their their technical support engineer (Riley) are summarized below [1].
It appears the controller is s8ll dropping out at 88%. One of the ques8ons I asked at the last board mee8ng (which wasn't answered) is where that number came from and why it matters. This wasn't intended to criticize anyone; I was simply trying to understand if spending another eight thousand dollars makes sense when something important may not be adjusted right. Rashid's subsequent personal attacks seem misguided.
In my opinion it's reasonable to have more than just simple proposals or the advice of self-interested vendors before acting if we want to ensure owner's money is well spent every time. Rob got us into this mess when another board picked a bid and didn't look deeper into needs, options, or potential consequences. Consulting a fully qualified expert at the start could've save everyone a lot of time, cash, and inconvenience.
Just because something can be done which might make things beLer doesn't necessarily mean it's the best solution, or even that it's required. A little due diligence can ensure all the easy fixes are exhausted first. Will controller adjustments alone (or in conjunction with pressure regulators) cure this problem? It's hard to say. But to me it makes sense to find out before a board starts throwing even more homeowner money at a problem which s8ll hasn't been explained.
Best Regards,
-Steven
[1]
- the VFD speed setting of 88% is probably the low speed limit. This is the slowest speed motors can go before they stop. It reflects all the losses of pressure regulators, pipes, gravity, etc. which must be overcome so, at a minimum load (one faucet) at the end of the longest loop (15th floor), an acceptable flow rate is possible. This so called "end pressure" is normally about 60 PSI. He thinks 88% may be high for a typical setup. With that setting the full range of pressure adjustment via VFD control is only 12% and the controller may not be dropping down. Any unnecessary pressure created by overpowered motors is normally absorbed by the pressure regulators (PRs) and represents money wasted on extra (unneeded) electricity.
- This might explain pressure spikes. Currently the controller is set up for "no flow stop", which means when the system detects no pressure changes as the VFD is ramped down (indicating no one is using water) it shuts off. At a high minimum speed there's a smaller control range and the controller can start operating in a so called "bang bang" mode - meaning it only goes on/off/on/off.., as opposed to proportional regulation. Which is the whole point of using a PID (proportional, integral, and derivative) controller [2].
- Determination of a low speed limit setting is based on design calculations which were done to size pumps for the entire system. Since new pressure regulators (PRs) came into the picture after the pump was already purchased in this case, it's not clear where the 88% came from or if the motors are properly specified. Riley thinks the factory default value is probably closer to 50% for a controller right out of the box.
- PR's also have their own settings, which may alter the optimum low speed setting if changed. PRs are commonly used with this controller so there shouldn't be any problem (significant pressure fluctuations) if all system settings are adjusted properly for our installa8on. There are additional settings for factors such as VFD start speed, which may also change how the pressure ramps up from a stop condition and if it overshoots or not, which may also create pressure issues.
- Pressure logging is available for this controller. However it requires a specially formatted SD card. Riley will get back on this but he thinks their office can supply that for free or with a small fee. Both sides (pressure and suction) can be logged and the resulting file should be viewable with a free downloadable program or perhaps using Excel.
- We should probably contact the company who sold the controller (not the installer). This information is available from the install schematic title block, at the bottom. Ninety percent of CMC controllers are sold by California Hydronics. They may charge us a setup fee but they're very good with boosters. It may be the case that the controller was simply adjusted at a bad load time (e.g. very light or heavy use).
- Riley approved these comments.