Fume hood venturi.

[goldenchild]

The hatch to this hood doesn't lay perfectly flat. There are a few millimeters of space on the bottom and on the sides to about the half way mark. That’s where the air is coming in from.

I've already started working on a more traditional design though. I was doing some AR reactions today and caught a few whiffs. I guess since AR fumes are heavier that just NOx and the reaction was much more aggressive, the fume hood almost couldn't keep up. The solution would be a bigger blower to create more suction. But as it is, the blower is more than needed if used traditionally for this tiny hood. It's 465 cfm.

 

[eeTHr]

Hmm. That's even more than the leaf blower (with an air speed of over 200 MPH)!

 

[eeTHr]

Well, I got curious about the venturi design, and the possible amount of flow through the air inlet (pipe to the hood), as a percentage of the CFM from the blower. So I did some searching on the Internet.

The basic venturi has no air suction inlet, but merely produces a high velocity, low pressure area in the restricted portion (throat) of the device.

It looks like most of the designs for suction type venturis are for the purpose of merely creating a vacuum, rather than producing a significant flow volume.

Surely there must be a configuration to maximize the flow.


Here is one used to produce vacuum to operate aircraft gyros. It produces enough flow to spin a rotor, powering the gyro---

Nine Inch Suction Venturi

Here is shown design characteristics for suction venturis, but it's based on liquid or gas being forced through the device, thus sucking air into the output flow---

Modeling of Suction Venturi Flows

Below is shown the innards of an air-air model, but again, it seems to be for vacuum rather than flow---

Venturi Vacuum Generators

Here is a multi-stage air-air suction venturi, again emphasis is placed on vacuum creation---

Air Powered Vacuum Pump

There are even several types of venturi scrubber systems, which could prove to be interesting---

Venturi Scrubber


I think the data is somewhere in the above, to enable the design of a configuration for maximization of flow using suction venturis. But what the maximum percentage of blower CFM obtainable is, I don't know.


Below shows differing designs for the air inlet nozzle, and although it is for water-air devices, it may be somehow applicable. It also mentions adjusting the nozzle depth to alter the suction (and thus flow rate?) of the air.

Basic Venturi Construction for the DIY Crowd


The question is, "Does the venturi system offer benifits over the direct suction method, which requires a chemical resistant blower?"

 

[goldenchild]

The basic venturi has no air suction inlet, but merely produces a high velocity, low pressure area in the restricted portion (throat) of the device.

This makes perfect sense as to why the fast rising NOx fumes were easily exhausted but not the heavier AR fumes. Also the insense smoke when testing.

The question is, "Does the venturi system offer benifits over the direct suction method, which requires a chemical resistant blower?"

Besides not having to buy a very expensive chemical resistant blower, I would say no. I still have a trick or 2 up my sleeve to get around this but if it doesn't work I may have to break down and buy one of these distressfully expensive units

 

[eeTHr]

goldenchild---

I was thinking about the picture of the multi-stage air-air suction venturi that was in one of the links I posted. And it reminded me of the diagram you posted, with the multiple suction ports.

I wonder what would be the effect of having the same setup, but all of those ports suck from the same hood?

 

[goldenchild]

goldenchild---

I was thinking about the picture of the multi-stage air-air suction venturi that was in one of the links I posted. And it reminded me of the diagram you posted, with the multiple suction ports.

I wonder what would be the effect of having the same setup, but all of those ports suck from the same hood?

That's actually what I was going for in the picture. I made the picture for glondor's hood. I figured it would work somewhat like mine. I think a downfall of having multiple ports would be weakening suction the further away you got from the blower. Or maybe not. I don't know.

Sooo. I modified my fume hood. It now draws and exhausts directly through the blower. An "air foil" was cut out because it was imploding without it. Smoke draws in from about 5-6 inches from the opening. How did I get around the corrosion thing you say? Dupli-Color Bed Armor. I know it probably won't hold up forever but for a while at least. Then I will get serious and build a good sized hood out of wood and get a real corrosion resistant fan. I see this as an inexpensive learning experience 8)

 

[Geo]

im sorry i just now started to read this post.to create the venturi effect the air supply must be constricted to create a vacuum behind the constriction.your blower would have probably been fine if you had used a smaller inlet on your blower than your supply.in other words you should have used a 2" pipe on the blower for a 4" supply. also use a "Y" fitting and use flex pipe on your blower so you can insert the smaller pipe past the "Y". i don't have a diagram so bear with me.the "Y" has 2 sides on one end and a single on the other. use one side of the "Y" for the blower and the other side as the supply and the single end is the exhaust. this should create the vacuum you are looking for. good luck.

 

[Palladium]

I am currently building a fume hood. Here is what i'm using to coat the hood as well as my counter tops with. http://www.homedepot.com/h_d1/N-5yc1v/R-202588482/h_d2/ProductDisplay?langId=-1&storeId=10051&catalogId=10053

http://protective.sherwin-williams.com/tools/epoxy-coating-troubleshooting/epoxy-coating-comparison-chart/

I was in Commercial coating for years and this should do just fine for the application at hand. For the price of 2 gallons you can't beat it and you should have plenty to do with what is needed. No you don’t use the chips. lol
.

 

[eeTHr]

I did an experiment on the venturi concept today.

First of all, in all the Internet research I did, the actual "venturi" concept was only using a duct with a restricted zone, called the "throat." The throat is a section in the duct which has a smaller cross section than either the air input section or the output section.

The point being that the original "venturi" didn't have any suction port. Mr. Venturi's concept only proved that as air (or fluid) passed through the throat, the air speed increased and the air pressure decreased. I couldn't find when or who came up with the idea of including a suction port.

It could be assumed that the suction port was a natural addition, because the decrease in air pressure would result in a suction. However, the only decrease in pressure is that which is relative to the pressure of the input air. That is, the throat still has pressure above that of the ambient air pressure from where the suction is to be drawn. Since the only history of the "Venturi Effect" includes just the input duct, throat, and output duct, with no suction port, I'm not sure that the suction effect should properly refer to the "venturi."

Further, since any suction is not produced by the decreased air pressure in the throat, then the venturi configuration might not be necessary to the creation of suction.

Rather, it seems like a restriction in the flow, with it's associated resistive losses, might actually be lessening the suction efficiency of any suction system which uses air flow to create it.

So I did this simple experiment---

A shop vac, 5.5 HP, with a 2" blower outlet on it, was used. The PVC and fittings were 2". The outside of the PCV pipe needed to be shaved down a little to fit into the shop vac air outlet. This resulted in a strong air flow, similar to a leaf blower. Leaf blowers appear to come in two general power ranges, about 6 Amps and about 12 Amps, so this should approximate the smaller power versions.

A 2"x2"x2" "T" fitting was placed onto the PVC pipe coming out of the shop vac. This made the suction port enter the system at an equal cross section point, rather than using a throat section. A suction was created at the "T". The suction was tested by putting a piece of notebook paper over the opening, and the suction would hold it there. If the paper was slid off the opening about 1/3 way, the suction would pull the edge of it inward about a half inch, indicating there was also some significant flow created, as it maintained some suction even with the opening present (made by sliding the paper back, producing a "vacuum leak"). Since there is not any vacuum or flow measuring devices handy, this is the best indicator which was available.

Next, a longer length of the PVC was attached to the "T" air flow output end. The suction port then blew air outward, instead of sucking inward. This indicates that at least half of the Venturi configuration is necessary to create suction, and that being the larger cross sectioned output duct. Apparently, using just the "T" connector alone, approximated the larger output section of the Venturi configuration, by allowing the air flow to exit almost directly into the surrounding atmosphere. So it appears that larger exhaust ducting is necessary to make the suction configuration work.

Since reducing the cross section, in order to form the traditional throat of the original Venturi configuration only produces a reduction of air pressure relative to the input air pressure, yet still greater than the air pressure in the hood, the necessity of going to a smaller pipe to create suction is questionable. And since it appears that air pressure is not what is creating the suction, it may be considered that it is instead the air speed which is doing it.

At that point, the efficiency loss from the friction of the constriction of the throat, and the flow reduction of the suction caused by the constriction of the throat, should be factored in.

In other words, if the air speed is already sufficient, why constrict it by adding a smaller throat section? The combination of optimum air speed past the suction port, and size of the outflow duct from the "throat" area, in order to produce the maximum air flow would need to be determined. So far, all the "venturi" suction devices I've seen on the Internet are designed only to produce and hold a vacuum, rather than to produce a considerable air flow.

The multiple port configuration was also tested. Another "T" fitting was added directly to the existing one, without a piece of PVC (which would have constricted the flow slightly) in between. A suction was produced at the second fitting, approximately equal to the first. That's encouraging.

Also, several sizes of reducer fittings were plugged into the suction port of the single "T" fitting, when suction was being produced, but it did not increase the suction. It seem the only thing this did was reduce the suction air flow which would come from the hood. So it appears that the idea of using the "T" sections which have a smaller diameter pipe out the side, would not help the suction any.

It looks like the main factor, for producing suction, is, generally speaking, that the air duct outlet directly after the suction port have a larger cross section than the area prior to the port.

It also appears that multiple suction ports are the solution to maximum air suction flow.


Edit: Since the doubled up "T" connectors worked, it seems that the problem with adding the length of PVC to the single "T" connector was that the PVC inserts into the fitting, thus reducing the cross section at that point, and changing the dynamics of the duct at that point enough to cause air to flow out of the suction port, rather than inward. So a continuously increasing pipe diameter is probably not necessary, but a way of connecting fittings would need to be found which would not cause any minor air restrictions. Even angle fittings might cause enough disruption to throw a system like this off-balance.

Last edit: Spelling of "produce" in 11th paragraph.

 

[Barren Realms 007]

That is pretty good eeThr. Could you take a 3" tee or a 4" tee and do the same test and see what you come up with? Would it be asking too much for you to try it with a wye as well to see what happens?

It sounds like Glondor's problem then is just that his hood is too far from his tee and if he get's it closer it might solve his problem.

 

[eeTHr]

Barren---

One of Glondor's problems was that he was dividing up whatever suction might be created at the suction "T" three ways, resulting in only 1/3 of whatever could be developed, at each suction pipe in his hood. That's a parallel configuration. What is needed for multiple suction points is a series configuration for them.

The main problem, though, seems to be that he violated what appears to be the "equal or greater" rule for exhaust duct resistance. This is demonstrated by the above experiment where the same size PVC pipe was inserted into the output end of the "T." The PVC fittings have a larger inside diameter than the corresponding size of pipe, which allows the pipe to slide into the fittings (if they are the slip type of fittings). Since this small difference is significant to determining whether air flows in or out of the hood ports, then the length, and any change in direction, of the exhaust pipe is likely to affect that determination also, due to air resistance per foot from the straight ducting and also additional turbulence from the curves.

The basic rules of an airflow suction system would be the same for 3", or any size pipe, as it is for 2". I don't have a "Y" fitting here, but from what I've seen on the Internet, for various air and fluid flow-suction systems, it appears to me that the 90 degree suction port creates the most suction. I can pick one up on my next trip into town, and try it though.

goldenchild's post about Glondor's blower being on the wrong side of the run past his hood ports, and his multi-port diagram is what got me thinking about all this. And that's how I realized that a parallel hood port setup would produce less flow in each one. But there should be some suction produced, rather than air blowing into that hood. So that's where the exhaust duct being equal or greater in size comes into play, along with the equivalent of duct size reduction by the air resistance of the ducting length, and the turns made by the angle fittings.

So far, it seems that going one or two sizes larger in the pipe, after the "T" or a series of them, especially for a long exhaust duct run, or one with a few turns in it, would be necessary.

I'll post whatever the results are.

 

[eeTHr]

P.S. I would also think that the volume of air being sucked from the hood, being added to that already in the flow from the blower, would affect the speed/pressure computation in the following suction ports and exhaust ducting.

 

[eeTHr]

I thought of another test that I could do with the materials that I have on hand now.

Since adding the PVC pipe to the air output end of the "T" fitting caused the air flow to reverse in the suction port, and blow toward the hood instead of sucking inward and going out the exhaust, and since I realized that the PVC was actually smaller on the inside diameter than the same size "T" fitting, I thought it would be good to see if the problem was caused by the size of the exhaust pipe in relation to the size of the suction port, or in relation to the size of the air input PVC pipe from the blower.

I had already tried all of the reducer plugs, when the exhaust PVC pipe was attached to the output of the "T," with the result being that air still blew out of the suction port, rather than sucking in. But just as a confirmation, I tried putting the same size PVC pipe into the sucktion port, so now all the three openings of the "T" had the same size pipes.

In addition, I made both the exhaust PVC pipe piece and the suction port PVC Pipe, the same length, to eliminate any slight air resistance difference.

But air still blew out of the suction port, rather than sucking inward.

So right now the formula seems to be that the exhaust duct needs to be larger than the inflow duct from the blower. I take it that this is either because the speed/pressure differences of the larger cross section is what creates the suction, or that the added volume of air from the suction port adversely changes the dynamics of the exhaust ducting. Probably because the pipe from the blower is already allowing as much air flow as it can, for it's given pipe size and air pressure from the blower, and adding more air volume from the suction port merely creates a greater density of air in the exhaust which makes more back-pressure against the forward inflow from the blower, and thus stops any vacuum production. I guess it creates a state of equilibrium between the suction port and the exhaust, and the "T" becomes an ordinary forward-configured "Y," simply splitting the air flow to go out both of the open ends.

Incidentally, the "Y" fitting would be placed backwards from this hypothetical equivalent if it was used, instead of a "T," to try and create a suction.)

I would modify the suggested formula or rule, above, to say that the total air resistance needs to be less in the entire exhaust ducting system, than in the input ducting system from the blower, plus some amount inversely proportional to the added air volume from the suction port.

 

[Barren Realms 007]

Too many variables to say exactly. Best bet is to make sure your exhaust is larger than your incoming air port to create the suction. You did a good job on this. Thanks

 

[eeTHr]

Barren---

Yes, but too many bends in the exhaust ducting could drag it down also.

And the more hood ports, the larger it will need to be, too.


I needed to go into town, so I swung by the Ace Hardware for a 2" and a 3" "Y," and a 2" to 3" reducer. They don't have any "T" fittings larger than 2".

Replacing the 2" "T" with the 2" "Y" fitting, resulted in slightly less suction. The funny part was that it worked almost the same in either direction, with it "forward" or "backward." Backward, it had some vibration to the suction, and was a little weaker.

The inner diameter of the 2" "Y" was smaller than the inside of the 2" "T." The "T" is 2 5/16", while the "Y" is the same ID as the PVC pipe---exactly 2". Maybe the "Y" needs more strength because the the hole in the side is longer due to the entry angle. They are both by the same company. But the smaller inner diameter could be the reason the 2" "Y" didn't give as much suction as the "T".

So then I put the reducer and the 3" "Y" on there, with the suction part slanting forward (toward the exhaust) as it goes into the fitting, and there was about twice the suction as with the 2" "T". Moving the paper to give a 1/3 opening (vacuum leak) still gave suction, so the air flow volume was also increased.

I would envision the best chance of a workable system to be three or four "T" joints, close together, with slightly increasing connecting pipe size between them, and a larger size exhaust ducting system, as short and with as few bends as possible. And a 12 Amp blower for 115V, 15 Amp, service outlets.

 

[eeTHr]

P.S. There may also be increased performance by modifying the area where the suction port meets the split fitting. I know that certain flow disruptions in water lines with a high flow rate, will cause such a vacuum that air bubbles will spontaneously form, which can cause the pumps to become inefficient and over heat.

So it may be possible to create more suction than a simple joining of the suction port pipe, by fashioning some type of baffle in there.

 

[eeTHr]

P.S.S. I tried both sizes of "Y" fittings with the air blowing into the angled port of them, instead to into the end of the straight section. They both blew air out of what would then be the "suction" port. So that would be another problem with Glondor's original setup.

 

[Barren Realms 007]

Barren---

Yes, but too many bends in the exhaust ducting could drag it down also.

And the more hood ports, the larger it will need to be, too.


I needed to go into town, so I swung by the Ace Hardware for a 2" and a 3" "Y," and a 2" to 3" reducer. They don't have any "T" fittings larger than 2".

Replacing the 2" "T" with the 2" "Y" fitting, resulted in slightly less suction. The funny part was that it worked almost the same in either direction, with it "forward" or "backward." Backward, it had some vibration to the suction, and was a little weaker.

The inner diameter of the 2" "Y" was smaller than the inside of the 2" "T." The "T" is 2 5/16", while the "Y" is the same ID as the PVC pipe---exactly 2". Maybe the "Y" needs more strength because the the hole in the side is longer due to the entry angle. They are both by the same company. But the smaller inner diameter could be the reason the 2" "Y" didn't give as much suction as the "T".

So then I put the reducer and the 3" "Y" on there, with the suction part slanting forward (toward the exhaust) as it goes into the fitting, and there was about twice the suction as with the 2" "T". Moving the paper to give a 1/3 opening (vacuum leak) still gave suction, so the air flow volume was also increased.

I would envision the best chance of a workable system to be three or four "T" joints, close together, with slightly increasing connecting pipe size between them, and a larger size exhaust ducting system, as short and with as few bends as possible. And a 12 Amp blower for 115V, 15 Amp, service outlets.

Yes too many bends will cause problems, air turbulance that disrupts the flow.

That is interesting that the wye worked in either direction. I have never taken the time to measure the inside diameter of the fitting's and didn't know there was a difference. I have a bunch of older fittings I picked up years ago I will measure them when I get time and see if that is something that they have changed in the last few years. The wye is made stronger because of the stress it is put under for the applications it is used for.

If you use a 12 AMP blower I would suggest using a 20 AMP receptical because you will end up causing a 15 AMP receptical to over heat under continuous use conditions and possibly cause serious damage.

 

[Barren Realms 007]

P.S. There may also be increased performance by modifying the area where the suction port meets the split fitting. I know that certain flow disruptions in water lines with a high flow rate, will cause such a vacuum that air bubbles will spontaneously form, which can cause the pumps to become inefficient and over heat.

So it may be possible to create more suction than a simple joining of the suction port pipe, by fashioning some type of baffle in there.

When a pump does this is is called cavitation.

P.S.S. I tried both sizes of "Y" fittings with the air blowing into the angled port of them, instead to into the end of the straight section. They both blew air out of what would then be the "suction" port. So that would be another problem with Glondor's original setup.

That was why I suggested he change that part in the first place.

 

[eeTHr]

Barren---

I figured they limited the leaf blowers to 12 Amps just because 15 Amp breakers are so common.

That's the word for it---cavitation. It can also do that within the PVC run, itself, I think at the turns, if the flow is high enough. A partially open valve can do it also.

I know you suggested to Glondor that he turn the "Y" around and put the main air flow through the straight section. That's why I thought of confirming that for anyone interested.

It would be interesting to see how many CFMs it can be gotten up to with air suction techniques. If the cost of a regular blower plus PVC fittings turns out to be significantally less than the cost of an acid resistant blower, it would certainly encourage more beginners to build and use fume hoods.