In my recent post on my field testing of different filter technologies, I included a chart of the performance of the system with conventional filters on a typical day, which I have reproduced below.
What you are looking at is:
- Total flow (the fuzzy blue line) based on average velocity across the filters,
- Bag filter pressure drop (the red line) measured with a differential pressure sensor dedicated to the purpose, and
- Prefilter pressure drop (the green line), also measured with a dedicated differential pressure sensor.
The smaller inset graph, which I have enlarged and reproduced below, looks at the portion of the main grahp where the system flow rate has dropped of to the minimum level that is typically seen at low load.
The reason for focusing on this is that the bag filter pressure drop data shows an anomaly. Specifically, the bag filter pressure drop falls off with flow, just as you would expect, but then suddenly jumps up, even though the flow continues to drop off.
When I first saw this, I thought I had a problem with the data. But after thinking about it and looking at what is going on out in the field, I’m thinking that the jump may be real. The photo below illustrates why.
In contrast, the alternative technology we are using in the other unit, which is a rigid pocket filter looks like this under similar conditions.
So, while we haven’t proven it conclusively yet, we think that what happens is that the pressure drop across the bag filters takes a jump when the flow drops to the point that the bags collapse or un-inflate. This is not a new issue; in fact, it is something I was concerned with as a designer years ago and the NAFA guide I mentioned in the post on prefilters also includes a discussion of the topic in a paragraph in Chapter 4. What’s interesting to me is that I think we may actually be measuring the impact.
In addition to the pressure drop implications, I think the deflation of the flexible filters at low flow and the tendency for them to move around a bit as a result of flow variations in general may have some other interesting and important implications. Specifically, it makes me wonder if there is any tendency to shed particles, especially when the filters collapse.
Jeff and I are discussing ways to test this in the field involving measuring particle counts in clean side of the filter plenum for both systems while we cycle them from low flow to high flow and back. On the one hand, obtaining meaningful quantitative data with a particle meter in the field in a relatively dirty and uncontrolled environment with a lot of variables is probably difficult at best. But on the other hand, we may be able to develop a test that gives us some qualitative data and insights.
Bottom line is that we plan to look into the pressure drop issue and the potential to shed particles as the flexible filters move around in more detail as we dig into the data I have accumulated and move forward with our test. Stay tuned and I will share what we learn in future posts.
Meanwhile, if any of you have been looking at this in your facility and systems and have information you would like to share, let me know.
Senior Engineer – Facility Dynamics Engineering