New performance data from my DC

[John Lanciani]

After I posted some incomplete test results recently Michael Clark pointed out that by relying on supplied CFM data I did not really know how my system was performing. I agreed, and decided to go further with this. I was lucky enough to find A Dwyer 471 hot wire anemometer on E-bay for a reasonable price so I bought it. Here’s what I came up with.

My dust collector is an Oneida V3000. It is a 3hp system with a 14.25” backward inclined cast aluminum impeller and a HEPA filter. The filter is well seasoned and was cleaned prior to the testing. Ductwork is 6” PVC.
I used the following test equipment; UEI EM151 digital manometer, and a Dwyer 471 hot wire in duct anemometer.
The Claimed CFM numbers are extracted from an Oneida published system curve for the V3000. SP measurements were taken just before the cyclone inlet. Air velocity measurements were taken from the longest straight piece of pipe in my system, following Dwyer’s guidelines as closely as practical. All FPM measurements were rounded down to the nearest 25, and I used a divisor of 5 to calculate CFM from FPM. Since the accuracy of the Dwyer 471 is +/- 4% I didn’t feel compelled to calculate to the exact decimal.

The results were as follows;

SP Claimed CFM Measured FPM Calculated CFM
12.4(max)<100----------375------------------- 75
12 -------200------------- 925------------------ 185
11------- 400------------ 2050------------------ 410
10------- 525------------ 2950------------------ 590
9 ---------650------------ 3700------------------ 700
8 ---------775-------------4050------------------ 810
7 ---------900------------ 4550------------------ 910
6-------- 1000------------ 5100----------------- 1020
5-------- 1100 ------------5450----------------- 1090
4 --------1200------------ 6100----------------- 1220
3.8(min) 1250 -----------6200----------------- 1240



Somewhat surprisingly, my results fell almost exactly on top of the published performance data. Now I can rely on the published curve and the readings from the manometer that I leave installed on the cyclone inlet to gauge CFM as I use and modify the DC.

[Michael W. Clark]

Awesome John,
Now, as you say, you can rely on your gauge for an approximation of flow. As you know, once the filter gets dirtier, the flows and SP may not match up to your table. While the filter is clean, record the filter pressure drop if you can. Then, when the filter gets dirty (higher pressure drop), take another flow reading at a couple of the different SPs since you now have a way to do so. It should be pretty close to adding the additional filter DP and finding it in the table above, but there may be some variance since the cyclone DP is flow dependant.

For example, if you are now running at 7" (900 CFM) and you add another 1" of filter DP when it gets dirty, you should read around 775-800 CFM (the volume for 8"wg) when you have an inlet SP of 7"wg and a dirty filter.

Nice work, great information!

Mike

[Ole Anderson]

Great information John, thanks for posting. A hot wire is definitely the way to go, that way you just put it in your main run and by opening and closing gates, you measure all sorts of combinations. Not so easy with an anemometer. Nice to know that Oneida's fan curve for the V3000 is on the mark. As I noted in another post, their web page selling the 2 hp Super Dust Gorilla has two curves which are different. The first one you would see is a direct link to a graph that shows curves both with and without a filter, that curve overstates the maximum suction by about 2 inches. The .PDF curve referenced under the specification tab, which only has one curve with a filter seems much closer to what I was measuring. Go figure.

Your minimum flow is 75 cfm, safe to presume that is your leakage? Not bad. Would you be so kind as to post flow rates for various tools? What kind of filter loss are you getting at a normal operating flow of about 1000 cfm? I am not seeing a fan curve when I go to Oneida's V-3000 sales page. I do see a tabulation on the third page of their Spec Sheet .PDF under their tech spec tab, and it varies a bit from your column of their claimed fan curve points.

[David Kumm]

Great stuff John. When you open the ports for each machine in use, what is the cfm- sp range you get? In your real world use what cfm through what size port are you getting at the jointer, planer, etc. I think that would help those wondering what range they can expect to operate in the system curve. Given each system will differ some. Dave

[John Lanciani]

3HP V System Published 8-26-08.pdf

Here is the Oneida System curve. I'm not 100% sure where I got if from but most likely the Oneida salesman sent it when I was shopping.

Michael - regarding the filter, wouldn't that be accounted for in the SP I'm measuring at the inlet? If the filter was dirty the blower would have less "suck" since it would be trying to overcome the exhaust restriction, no?

[Michael W. Clark]

Michael - regarding the filter, wouldn't that be accounted for in the SP I'm measuring at the inlet? If the filter was dirty the blower would have less "suck" since it would be trying to overcome the exhaust restriction, no?

John, I had to think about this for awhile. I don’t think it will be the same, you will read a lower SP on your gauge and that will indicate a lower flow instead of a higher flow. The increased filter loss will shift the performance curve down. Here is my thought process:

When you built your chart, you were changing the system curve or relationship between flow and SP by opening and closing a gate. This allowed you to generate a performance curve for the Oneida equipment. As the SP goes down (gate opened), the flow goes up and you are moving farther to the right on the performance curve. If you held the blast gate in one position and varied the flow (assume VFD on fan), the SP would go up as the flow (fan speed) goes up and vice versa.

Let’s say your system is in normal operation and you take a SP measurement of 7”wg and a flow reading of 900 CFM. At that same point in time, there is a pressure drop across the cyclone and the after filter. Let’s ignore the cyclone since it behaves like the system and assume the 7”wg is analogous to the inlet SP for the fan. Let’s also pick a number for the filter drop of 1”wg. So, if the fan is overcoming 8”wg of SP, it is delivering 900 CFM for this situation/example.

The question is, what happens if the filter gets dirtier, how would your SP gauge respond? Assume the filter gets dirtier such that the pressure drop at 900 CFM increases from 1”wg to 2”wg. Based on the flow and pressure readings, the system needs 7”wg of static pressure to flow 900 CFM in this condition. This relationship does not change. If your filter DP is now 2”wg, the fan has to overcome 9”wg instead of 8”wg to flow 900 CFM, providing a gauge reading of 7”wg. However, this higher pressure would cause you to move up the fan curve to a higher SP and lower flow, as your performance chart shows. As you reduce the flow, the system SP goes down by the square of the change in flow. You would read a SP less than 7”wg which, if you look at your chart, would indicate a higher CFM. But, it would actually be less than 900 CFM based on the system requirement/curve. For example, if you read 6”wg on your gauge, the system flow would be 833 CFM instead of 1020 CFM.

Oneida is rating this system in terms of inlet static pressure to their equipment based on certain losses through their equipment. If the pressure drop within their equipment increases beyond what they have accounted for, and they have no way of applying additional fan speed and HP, then the CFM exhausted will go down. This is very common on large systems with automatically cleaned filters that operate between 2”-6” wg DP. If you size the fan for the 2”wg DP, then you can’t pull the required CFM at 6”wg DP.

I hope this made sense...It took me a while to wrap my head around it and now it hurts.

Mike

[David Kumm]

I think you are on to something Michael. I've always wondered how the curves would be affected with different filters, both size and type. It is very difficult to compare the oneida systems as some curves are with filters and some without. The 3 hp V system has much less flow than the 3 hp Gorilla. I thought the fans were the same, but the cyclone inlets are different sizes. I'm assuming that restriction accounts for most of the difference since the cyclone design and filter differences would not affect the curve at the cyclone inlet. If John can take the filter off and run the same tests the results should prove your theory. The motor won't overamp with the BI blade. It would also be informative to know the amp draw at maximum cfm to see how much of the motor power is utilized. I suspect not all since the same motor is used in the Gorilla series producing higher cfm. Dave

[John Lanciani]

Thanks Michael, that makes perfect sense. I do have a 0-5" Magnehelic on the outlet to keep tabs on the filter condition so I always have that to reference as well.

David, I'll give it some thought but I'm not sure that I'm brave enough to take the filter off and run the collector since I have no way of venting the exhaust outside. The V systems do have much smaller filters than the Gorillas - I actually figured that when the time comes to replace my current filter that I'd get some Nano's and put together a manifold run more than one filter. Maybe even a box so that I can run them out side in instead of inside out.

One thing I've been thinking of is if people are interested I may be inclined to send the anemometer and manometer on road trips to visit other shops. My only concern is that the hot wire anemometer is fairly delicate and we'd have to come to an understanding on responsibility if something happened to it. If anyone is interested please let me know and we can work something out.

[David Kumm]

John, I hear you. Wasn't remembering the filters were seasoned. Dust everywhere. I use Ole's method to measure at the port and then correlate the velocity or cfm to the amp draw displayed on my vfd. If I notice the amps dropping- usually in 1/10s I know the filters need blowing. Dave

[Michael W. Clark]

John, I hear you. Wasn't remembering the filters were seasoned. Dust everywhere. I use Ole's method to measure at the port and then correlate the velocity or cfm to the amp draw displayed on my vfd. If I notice the amps dropping- usually in 1/10s I know the filters need blowing. Dave

This works well too, fan amps are very sensitive to flow changes.

John, since you have the SP gauge and anemometer. Pick an operating scenario, say exhausting from the tablesaw. Take a flow, SP, and filter DP measurement and write it on the duct or put it somewhere for reference. Some time down the road, set the gates the same way to exhaust from the tablesaw and re-read the SP and filter DP. Calculate
CFMnew = CFMorig x (SPnew/SPorig)^0.5 This will give you the new exhaust volume at the current static pressure reading, assuming no changes have been made to the duct system you are exhausting (its how I calculated the 833 CFM above). Now read the filter DP and you will start to get an idea of how the filter DP affects the exhaust volume. This will help you determine when is the correct filter DP to clean it. Excessive filter cleanings take up a lot of time and will shorten the filter life.

Mike

[Carl Beckett]

I might have said it a little differently ( which could be wrong)

Yes you would see a change in your SP reading. But, suction pressure increases have a different effect than outlet pressure increases. So you wouldn't see 1:1 change

On any inlet the most you could ever pull is 15psi (vacuum limit). Power requirements increase exponentialy as you approach this. The outlet can push higher values.

[Michael W. Clark]

I might have said it a little differently ( which could be wrong)

Yes you would see a change in your SP reading. But, suction pressure increases have a different effect than outlet pressure increases. So you wouldn't see 1:1 change

On any inlet the most you could ever pull is 15psi (vacuum limit). Power requirements increase exponentialy as you approach this. The outlet can push higher values.

I used the linear analogy as an approximation and it is likely within the accuracy of the measurement equipment. There are too many variables to calculate what the flow reduction would be as a function of filter DP. A calculation would still not be as good as measured values.

If I were designing this product, then I would state the performance with a filter drop of x"wg and size the fan accordingly. This is what we do on pulse jet collector systems unless the CFM is very large or the customer must maintain constant CFM in the process.