Dust Collector Performance and measurement

[ian maybury]

Hi guys. Seems to me that there's a lot of ways of measuring aspects of the performance of a dust system - and that the choice depends a lot on what the objective in taking the measurement is.

Adequate CFM delivered at the point of collection in a hood that does a decent job of capturing flying dust is as David's method targets clearly the ultimate objective/the single factor that determines dust collection performance - all of the other stuff about duct and fan sizing and equipment pressure drops are only variables that play a role in determining this.

If you're selling/providing a system one major complication has got to be the likelihood that depending on which branch of the system and which machine is connected the resistance/pressure drops may be very different. e.g. it may be hooked up to a very long and bendy line of ducting, or the machine at the end has highly restrictive hooding - or not. Which means that by a CFM measure the system may end up looking good or bad depending on which line/machine it's hooked up to.

You can measure flow/CFM wherever you like in the system (below roughly 24in WG) and get reasonably accurate numbers, but the above is always going to kick in to mean that the figures may or may not be comparable with e.g. system specs.

Measuring pressure drop (across them) is the classic way of measuring how dirty your filters are. You probably can get a measure even when they exhaust to a room by reading the (positive pressure) at the filter inlet versus the zero in the space it exhausts into, but unless you can ensure a known air flow/CFM every time the numbers are not going to mean much. Meaning that the above question of which line and machine etc is connected is again a potential spoiler.

Motor amps (effectively the power drawn) can be an indication of the CFM and can hence can infer whether or not e.g. a filter is blocking. The problem again though is that unless the measurement is always made in the same situation (same line, no blocking - absolutely nothing changes except the state of the filter = same restriction/pressure drop except for the filter) it's not going to mean much. The related issue is that since there are three variables (amps/HP, pressure drop and CFM) involved you need one of the three plus the fan curve to extract an absolute value/measurement for the third. (calculation requires two of the three) Any of these be useful indicators if compared to each other - see below.

All this is a long winded way of saying that there is a reason for the the use stock test set ups for fan testing - specifically a given length of inlet duct of a given diameter with a given configuration of (variable) inlet. These deliver a known system curve and other consistencies, and hence comparable test results - normally fan curves that are reasonably good predictions of fan performance (CFM and pressure capability) when matched to system curves when designing systems.

It doesn't take a lot of thought to see that a fan maker putting out unreliable/inaccurate fan curves isn't going to be in business for very long - at least in industry anyway where engineering people use well established system design methods which use this information. Where hot air claims are quickly going to be quickly seen for what they are, leading rapidly to a court and claims for consequential losses.

Maintenance tends to be a different matter. Presuming no changes to a system, maintenance procedures normally entail taking pressure and amp readings in a known configuration (e.g. connected to a particular machine, and required that there have been no mods/leaks etc to the ductwork or machine hoods etc), and checking to ensure that they have not changed except within permitted limits. The point here of course is that these numbers are relative only to an original target set calculated or taken during commissioning - they are not taken to mean anything particular in terms of system capability.

Which brings us back to your issue Chris. Hooking a Clear Vue fan and cyclone up to a given line on somebody else's system should show up whether or not it's doing better (delivering more CFM) than the previous equipment, but the numbers are not in isolation typically going to be very comparable to specification data unless you have a good curve for their system as well to use in interpreting them.

If the requirement is tracking the state of filters then the route is probably to select a given line and machine hood, and establish a pressure drop (and maybe fan amps) figure for clean filters. Set the user up so that they know the figures will no longer work if they change hoods or ducting, and arrange that they repeat the reading at intervals in exactly the same way. (best to permanently install a Magnahelic meter or a U tube or similar if possible) That they change the filter if the numbers exceeds 'X'. (which number should be available from the manufacturer of the filters, but if not you it can be determined over time with reference to other measurements)

David's anemometer has to be a pretty good option for this purpose too, bearing in mind that there will be a bit of technique involved in using it.

Just don't expect the readings to be in any way very directly comparable with the fan specification unless as above you have a fan curve and good system information....

On woodworkers and not wanting to be bothered with the details. It's dead simple really I think. They either put a little time in to understand the basics - which actually are not complicated. If they can't be bothered the result is inevitably that they have to as above depend on others to tell them what to do. Which is (unfortunately) why the manufacturer's of hobby/DIY dust systems run riot with misleading performance claims and the like, and partly why even commercial woodworking has a pretty chequered history so far as the topic is concerned..

It's in the end an issue of motivation, and many people given the chance on issues that don't deliver immediate consequences will kick the can down the road rather than bite the bullet and solve the problem.....

ian

[Chris Parks]

Ian, I think it is more complicated than that. In a lot of cases the market is older men as a rule who have had by dint of the age they grew up in a limited education and find the whole thing totally baffling. I myself fall into the limited education category and I freely admit it but I have been looking at this stuff for years as I used to play with flow and measurement on a limited scale in a job when I was younger. If their education is not such an impediment then the "I want it now factor with no hassles" comes into it. If I did not attempt to simplify it then there would be a lot less cyclones sold in this country as those people would simply walk away. I have a few ideas which I am thinking about that might make the process far more enjoyable as it would give answers instead of confusion. Don't let us kid ourselves, until this issue can be addressed then the uptake of efficient dust extraction will be a lot less than it has the potential to be. I had a potential customer ring me today and he had just come back from the doctors who had told him that his lungs were suffering from wood dust ingestion, we need to get to these people before the doctor.

I don't see why it would be a huge issue to build a design app that lets you pictorially build the duct system and computes the performance through that ducting as you add or take away the bits, in other words a pictorial duct design. I know there are apps that do this but they are not easy to find and some are not easy to use mainly being aimed at engineers. We know the losses incurred by all the bits so given the computer skills why can't that be done? I know there would still be questions regarding hoods and the general stuff that hangs off the ducting but the person could do his own design and play with it to suit him as long as the air flow falls within acceptable parameters. Any computer experts out there?

[Michael W. Clark]

Chris,
I use a spredsheet for SP calcs based on ACGIH, but I'm sure an app could be developed if it is specific to WW. You could select a tool (telling the program the CFM and hood SP required), then enter the length/size of duct and number of fittings. It would spit out the SP required and CFM. I can build a spreadsheet to do this, so I would assume an app would also be possible. Two questions: 1) How would the user of the app relate this information to the DC system they need to buy? (minimal manufacturer's data avaialble today) and 2) If the customers are older or don't want to be "techy", why do you think they would use an app to get this information? If you do the application for them (which is what I do in my day job) then you are liable if the product does not perform as stated. For example, Oneida designing the duct system, if it doesn't "work", the buyer isn't responsible.

Back to your previous topic, I think an OEM could make a gauge that takes a DP and reads out in CFM. This would have to be correlated, as mentioned previously. I think you could do it with the cyclone inlet SP (see Grizzly's techinal documents for an example) reliably if you are not using a filter. If you are using a filter, then you could do this with cyclone DP. An electronic gauge would be slick and probably less expensive to mass produce.

Until users understand the importance of CFM at the tool, HP required to exhaust that CFM, and the need for good after filters (if returning air), we are going to continue to hear about health problems. Even with the best DC design, it is ultimately up the the owner to use/apply it properly. I have seen this many times in industry where the operator will not use a hood because it is inconvenient. It has to be convenient or made where there is not a choice to use it. I'm sure we have all been guilty of "its just one cut, I don't need to turn on the DC for that". I know I am.

Mike

[ian maybury]

I wasn't making a social comment, and I wasn't saying don't simplify it Chris. What I was trying to communicate though is that there are certain inescapable realities (i.e. the technical underpinnings) to the performance of dust systems - and that these have to be borne in mind when taking and using measurements. Otherwise they can be highly misleading, and lead to trouble for sellers or buyers.

This leaves only two choices - either we take somebody's word for it, or we choose to engage in the technicalities.

It's not even that it's all that complicated - woodworkers in general handle much more complex stuff all the time.

Given that many will not engage in the engineering the seller has no choice but to simplify the message (usually by responding to the specific concerns of the buyer), but the risk is always that this information will be used in situations beyond that to which the simplified version applied. The setting up of a maintenance regime as above whereby readings (which won't be all that useful for other purposes) are taken in clearly defined conditions to decide whether or not e.g. a filter change is needed is a good example of this approach.

It's technically very feasible (and i'm sure already done) to put together system design software - chase up a professional air systems design outfit providing services to industry and you will no doubt find something similar in use. Probably a lot more sophisticated than we need, but a cut down version would be very possible. (a nice project for a third level technical college)

There's a few issues that quickly follow however:

1. The published performance data for most hobby/diy dust equipment (which amounts to key input information) is incomplete and not accurate. Independent testing costs money.
2. The user needs to understand system design, or they will risk making errors.
3. The likelihood is that it'd move the user into a situation of responsibility for eventual system performance that most equipment makers have so far avoided - (a) because most sell only parts of an eventual system, and consequently can't precisely know the eventual circumstances of use, and (b) because many seem to happy to use the resulting 'space' to run riot with conflicting and OTT claims regarding the performance of their kit. (i'm reluctant to be more specific on this on a public forum)

The garbage in, garbage out saying comes to mind. Not to mention that there's still the need to simplify the message regarding the basis of the resulting solution for the eventual customer - and to get them to understand/accept that it has limits.

I agree 100% regarding the health issue. The trouble is though that the very guy that presents at 60 with breathing difficulties after a lifetime of woodwork is very likely the same guy who held for all those years that 'a bit of wood dust never did anybody any harm'. It's hard to know what the real risk is, but let's put it this way - there's very few of the kids that take up smoking that expect to end up with cancer.

This isn't an argument for inaction - the opposite in fact - but there's very little any of us can do (except our best to help raise awareness) to alter (a) the reality that we each make our own futures by our beliefs, and that (b) it's damn hard to shift those beliefs... i.e. stuff ends up where it does for good reason.

ian

[Ole Anderson]

Chris,

Your original post implied that you were looking for a solution after making a physical change in the system, now you are looking for an answer in the design stage and that changes the response to your question immensely. Any computer solution available to the end user will, even if greatly simplified, be overwhelming to the average non-techincal user. I have used AutoCad for 15 years, but now that I am retired, frankly, I am not interested in returning to that level of computer use given all of the other activities I have in retirement. (How did I ever get everything done and work 40-60 hours per week?) Yes I ran a few iterations of Bill Pentz's static calculator when designing my system, and I started a AutoCad drawing of my system, but that was about it.

Getting back to your original question which seemed to be "Is there a simple solution for my customers to know the performance of their system as they use it and make changes", I go back to giving them a SP gage, a system performance curve (which you could calibrate for them) and a CFM rate needed for each machine posted on a wall, all in one location. I can't imagine anything simpler for them.

[David Kumm]

Michael, for even your example to work we must have faith that the CFM requirement from the machine manufacturer is correct. They always seem low to me . Since in theory most ports seem designed based on 4000 fpm, it makes sense in my world to design for a higher number- say 5000 fpm minimum and hope for higher. I've gone to the extreme but run old machines. It might cost an extra $500 to over engineer but way more to replace. I've done that enough to know. Dave

[Chris Parks]

Ove, during this discussion I have thought about it even more and I have crystalised a bit more in my thick head what I think we need and make no apology for that. When I say an app I don't automatically think numbers I think pictures or objects. Put an object in place and that object has a value tied to it and so on. Think Sketchup compared to Autocad? maybe my view is to simplistic but then before sketchup we had complicated CAD programs but nearly everyone can use SU. Fairyland stuff? maybe....

[Michael W. Clark]

Michael, for even your example to work we must have faith that the CFM requirement from the machine manufacturer is correct. They always seem low to me . Since in theory most ports seem designed based on 4000 fpm, it makes sense in my world to design for a higher number- say 5000 fpm minimum and hope for higher. I've gone to the extreme but run old machines. It might cost an extra $500 to over engineer but way more to replace. I've done that enough to know. Dave

David, I agree. I use the ACGIH recommendations. Take a tablesaw for example, they recommend 550 CFM on the cabinet and 350 CFM on the blade guard. This would be a 5" on the cabinet and 4" on the guard, which would be about 4500 FPM in a 6" before branching to the base and guard. However, for a dado blade (which most of us use from time to time) they recommend 800 CFM on the base and keep the 350 CFM on the guard. This would best be done with a 6" base connection, 4" guard connection, and 7" connection before the split. I think you could cheat a little on the guard and keep it all 6" with the guard being 3", but in a large industrial system, it is very rare that I would go below a 4" and never below a 3" when designing a system for dust collection. It can easily be more energy efficient to add 200 CFM and go with a 4" duct instead of a 3" duct that boosts your system SP. In a home shop, the system is not as large and you probably already have the SP available because you are only running one tool at a time.

Mke

[Michael W. Clark]

Ove, during this discussion I have thought about it even more and I have crystalised a bit more in my thick head what I think we need and make no apology for that. When I say an app I don't automatically think numbers I think pictures or objects. Put an object in place and that object has a value tied to it and so on. Think Sketchup compared to Autocad? maybe my view is to simplistic but then before sketchup we had complicated CAD programs but nearly everyone can use SU. Fairyland stuff? maybe....

There is a program called "Heavent", and it is more graphical. I would not consider it easier to use, it is very quirky. There is a lot of information that has to be manually entered and would require someone to be familiar with loss factors, hood losses, etc.. The program is setup for "balance by design" where you add flow and fixed restrictions (more ductwork and elbows) to balance the branches, (the author doesn't believe in blast gates ) It may be fine for a single run, but it is not free. I think it is $300 or so. I only mention it because you might check it out and get some ideas. If an OEM came out with a "Duct Planner" like Grizzly's shop planner, it might go over very well. Like David said, the flows and SP requirement at each machine would be critical, but this could be behind the scenes in the program.

Mike

Mike

[Chris Parks]

If an OEM came out with a "Duct Planner" like Grizzly's shop planner, it might go over very well. Like David said, the flows and SP requirement at each machine would be critical, but this could be behind the scenes in the program.

Mike

Mike

That is the sort of thing I envision.