check your local building codes. Many towns have adopted the international ruling on makeup air which states that above 400cfm must have MUA!
Good luck

The problem with assuming that some burners will be off and hence the total possible BTUs will be reduced and hence less cfm are needed is that we don't have hoods in which various portions over unused burners can be sealed off, saving all the air velocity for the zone above the operating burners.

Hoods do not suck smoke off of burners. A turbojet intake could probably do that, but a typical hood flow rate can barely pull in smoke skimming past the edge of the aperture. A velocity map of this [lack of] flow at a distance can be found in the ASHRAE Handbook: 2003 HVAC Applications. Hoods capture the rising smoke (if the hood's aperture is large enough) and then contain the smoke if the air velocity is high enough. (If not, the smoke curls out of the hood.) In between capture and containment, the baffles attempt to capture the larger grease particles.

While a single 10-inch duct is normally preferred for the higher flow rate applications, rectangular ducts to fit walls and even pairs of ducts are not out of the question.

Actual flow rate will be considerably less than rated flow of the fan/blower because fan cfm values are typically taken at zero static pressure, whereas the ducting, hood, baffles, MUA, etc. conspire to cause a pressure drop across the fan, not likely to be less than a tenth of an inch (of water column) with perfect MUA.

This interior pressure drop alone is three times greater than certain combustion appliances can work with correctly if they share the same air pressure as the kitchen. The resulting combustion backflow is dangerous. For more, review MUA related threads here first.

What goes out has to come in, or else the house would implode. So the negative pressure will increase until the fan is gagged to be equal to how well the house gags the MUA. Actual flow rate in that case may be only 300 cfm.

kas

"Hoods do not suck smoke off of burners."

Kas - I'm in the process of a kitchen remodel so I can't speak to my new range hood, but I'm replacing an old school wall fan which was about 3-4' away from the cooktop. That fan was filled with grease, and I could visibly see smoke being drawn towards the fan when searing meat. Certainly nowhere close to all of it, but the air movement was evident. Also, conventional wisdom states to start up your hood before cooking in order to stabalize the air flow before actually cooking. Just wondering how you quote above reconciles with this?

A good question that begs to be answered.

First, we should observe that a simple wall fan, without any ducting or hood or hood filters/baffles, has a very unrestricted path, and might move 300 cfm depending on design. If the kitchen were filled with greasy smoke it would eventually all be pulled out by the fan, although some grease would have precipitated on the walls. The velocity of air from this fan can be high close to the fan, unlike many pop-up hoods that pull lower cfm over a larger area.

Second, when I write about upward velocity of effluent of 3 ft/s as measured in cooking experiments (references on My Clippings page), it is just a number pulled from an infinity of numbers that would apply to rising effluent as a function of height, angle off vertical, and temperature of the cooking surface. The goal is to not have this effluent curl out of the hood due to momentum reflection in the hood, hence the hood flow (at least close to the baffles) should equal the rising velocity of the smoke. (See schlieren photography of hoods working and failing as published in various sources.)

When the smoke rises and meets the air flow going into the wall fan, the smoke velocity vector and the velocity vector of the air induced by the fan at that particular point in space relative to the fan are added vectorially. The new vector bends toward the fan, and if it gets close enough it will be further rotated until it is sucked in.

Similarly, a 1000 cfm through a 10-inch duct would do the same at a wall, as the velocity would be high enough that the great fall-off of velocity with distance would still leave enough velocity to turn some of the smoke. But when this duct velocity is reduced by hooking the duct to a large hood intended to capture all of the greasy effluent and extract some of the grease, the velocity at the hood is modest and at the cooktop negligible. (It is only 10% at 1.6 times the hood short dimension width below the hood, per the ASHRAE Handbook.)

To put it another way, the pressure at the cooktop is not significantly less than elsewhere in the room, so no sucking of the smoke at the surface is going on.

So, if greasy smoky effluent rises past a fan that is itself causing a significant local velocity, a portion of the effluent will be immediately pulled in; the rest eventually after mixing with the room air.

As far as starting the fan early, the house air has some inertia, and it has to be moving to support the air flow. I've never tried to analyze the time to accelerate the air to the steady-state condition. My guess is that it strongly depends on the amount of the house from which the MUA is being taken. An open nearby window would probably reduce the acceleration time to be nearly the same as the fan rotation acceleration time.

kas

Check with your building department if you wish. If there is a limit a large dimensioned capture fixture would help make use of a smaller than 1000cfm blower. Or fans are variable speed, so you control it.

So if a typical hood vent captures smoke so poorly, is a higher amount of CFM actually useful? There are some posts on various sites that state anything over 360 is overkill.

This was not a typical hood vent being compared; it was a fan in a wall blowing air from inside to outside. The one's I've seen are of the order of 8 to 10 inches in diameter (see example).

The useless answer is that there is an amount of cfm for a given hood aperture, baffle or mesh area, height above the cooktop, cooking temperature, and food that is enough, so more might not be useful for cooking ventilation, although it could aid household air replacement. Call it overkill if you like.

More than 360 cfm might be overkill for a tiny hood over a toaster oven. 360 actual cfm (likely requiring a 500 cfm rated fan) would probably do for mild cooking on a 30-inch or 36-inch cooktop. There are people here who are grilling on 48-inch gas cooktops. They will need a lot more than 360 cfm.

I have a 10 square foot aperture island hood 34 inches over a 36-inch induction cooktop and 3500W induction wok hob. I use a 1500 cfm rated fan. This seems adequate. Actual flow rate is roughly 900 cfm with sufficient MUA, and because it works with 90 cfm/sq.ft., 90 cfm/sq.ft. is the specific flow rate I recommend.

Why does this work, one might ask, when the effluent upward velocity can be 3 ft/s and 90 cfm/sq.ft. is only 1.5 ft/s. I believe the answer lies mainly in the design of the baffles. When the rising effluent reaches the baffles, it tends to be entrained in the roughly 3 ft/s velocity of the air flowing around them and into the upper part of the hood. In other words, there isn't much reflection (effluent curling downward) at the baffles, even though the baffles present an apparent "blockage" area of roughly half the aperture area. What is actually going on would require computational fluid dynamics to investigate.

Industry has evidently found baffle configurations that are relatively economical to manufacture and are aerodynamically effective without having to emulate NACA duct designs.

kas

Here is a link that might be useful: example wall fan

I love this forum. THe information is fascinating. Thank you.

If I go with the Imperial, should I do the two 8 inch runs or combine? Imperial claims a 25 percent drop in CFM if it it combined to one.

A bold statement by imperial. What size are they assuming you are combining into, using what type of transition?

For equal duct sectional area, a single duct must be the square root of 2 larger than one of the single ducts. For an eight-inch single, this would call for a twelve-inch duct (eleven-inch ducts are not going to be on the shelf). However, the frictional loss on the sides of two eight inch ducts is significantly larger than one twelve-inch duct, so a ten-inch duct may be sufficiently equivalent. This can be checked with on-line duct calculators that I don't have time to look up myself.

Significant pressure loss can occur in bends, and in the transition ducting getting from two eights to one ten. So the pressure losses of the entire paths of both configurations needs to be analyzed if there is a question.

Using a roof fan instead of internal fans will allow fan sizing to compensate for estimated pressure losses. Note that when doing that evaluation, the pressure losses across the baffles and through the transitions that the hood itself presents need to be included. The hood + baffles might be 0.1 inches of water by itself at your desired flow rate. A roof fan also allows an in-line silencer to be installed if there is room in an attic or other point in the pathway.

I should note that when a hood is very long, it has to also grow in height to assure that air is pulled evenly across the length. Using two ducts instead of one reduces that need for a lot of height. Dual exhausts also make the in-hood fans fit easier than one larger fan would.

kas