Laser exhaust fiasco

[Kevin Groenke]

We've been going back and forth with a mechanical contractor to get a recent roof-mount laser exhaust fan relocation project resolved. This system is supposed to provide the specified 500CFM for two ULS X-660's. They've installed and re-worked the system twice already (at a cost of ~$18k) and the system still doesn't meet specs. They're current plan is to increase the size of the riser pipe from 10"-14" dia.

The lasers each have 2)4" outlets totaling ~50sqin of surface area - the 10" riser has 78sqin of surface area. Common sense tells me that making the riser pipe larger will not help and may in fact make matters worse - am I missing something?

I contend that the problem isn't the diameter of the riser, but the engineering of the manifold and a filter box transition adding ridiculous amounts of static pressure to the system.

http://lh4.ggpht.com/_rHdeE-SI5cc/TA0jKAD2MJI/AAAAAAAABzg/X0ijtNwv-Uo/manifold.jpg

http://lh6.ggpht.com/_rHdeE-SI5cc/TA0jKel_o5I/AAAAAAAABzk/9D2ceCWfasY/filter%20box.JPG

If increasing the size of the riser doesn't work, the contractor is talking about dropping 2)10" pipes out of the filter box - one to each laser (which I can only presume they'll stupidly put 2-tee's and a dead-end on).

Thoughts?

I appreciate any info or suggestions anybody can provide. If my reasoning is unsound I want to hear about it.

The text of an email outlining my concerns to the contractor (which they've decided to pretty much ignore) follows if you're interested.


The plan to replace the riser and laterals on the rooftop can proceed if --- feels that this will solve the problem.

As a layman,however, I have my doubts that this will be successful. I do not purport to be an engineer, but I have spent 15 years around industrial dust collection systems and as an avid whitewater kayaker I have a pretty good working knowledge of hydraulics and turbulence.

There are two specific elements of the exhaust system that seem problematic to me: the filter box inlet and the branch manifold.

The filter box inlet looks to me just like a dam: a big wall with a small hole in it. In my minds eye, I see a tremendous amount of air swirling around (an eddy) inside of that box.

The branch manifold at the end of the line seems equally troubling. Everything that I have ever heard or read about dust collection systems indicate that wye's should be used rather than tee's, that reducers should be long tapers rather than abrupt changes and that dead ends increase static pressure. The end of the system as it stands defy's all of these conventions. Again, as a layman, I see significant efficiency loss at that manifold.

Perhaps retrofitting a round to square transition at the filter box and/or a tapered branch manifold (http://www.spiralmfg.com/latsf.htm) at the end of the existing line would increase the airflow by improving the efficiency of the system (by reducing static pressure). Obviously making these changes would be much faster to make and a fraction of the cost. It seems to me that there is adequate space to enact both of these changes.
http://www.spiralmfg.com/images/multi-branch4.gif

It would be interesting to see what sort of airflow we got without the existing manifold. If --- wanted to remove the end of the system and take readings I would not be at all opposed. I actually have a manifold that could probably be installed to see how it affects performance.

Again, if ---, wants to go ahead and change the exterior ductwork and you're ok with that, then they have my blessing. If they are interested in pursuing any of the suggestions above, I prefer that interior work be done while I am around.

Thanks,

[Anthony Whitesell]

The 4 4" outlets have 12.5 sqin each for a total of 50 sqin. The 10" riser should be plenty.

My question to you, how much airflow area is in the "front" of the machine? You need to have equal entry area and exit areas. More on the entry if it is filtered. If you open the machine and run the system, can you make the CFM requirement?

I figured this out at SOHK (school of hard-knocks). I build a nice router table with three ports, each 4". When connected directly to the DC, I could not get ANY dust collection. I then tried two, then one port. Having 1 port fully open and the other port 1/2 open provided the best dust collection. It was at that point that the surface area theory hit me and now, of course, I see it mentioned everywhere.

[Mitchell Andrus]

There are ways to calculate this stuff. If they're guessing, let them do the trial and error on their dime. If they have calculations to back up proposal to you, sign off on it with a proviso that a 10% 'miss' costs them money.

Seems to me the experts should not get paid to fiddle around until it works.
.

.

[Dan Friedrichs]

I'd agree that the T's are hurting tremendously. Replace with wyes.

Also, the flex duct is a major restriction. Can it be replaced with solid pipe?

As Anthony said, make sure you have an equal area for intake air.

Making the riser pipe larger won't help (you have enough area already). It might help very slightly, but it's not the problem.

I doubt the filter box is hurting it very much, and dead ends do not increase static pressure.

I think your major problem is the T's and the flex duct.

[Greg Portland]

It's even worse that a T-connector... those look like pipes inserted into a 10" tube and bondo'd in place. A smooth 90 is bad enough but you have a rough edged pipe @ 90. I bet just replacing the manifold and flex duct would solve the problem.

This whole thing looks cobbled together with parts that they had laying around (which they are charging you full cost for).