Blower design

[bruce]

Next pic in the project.

I cut the spacers and drilled the holes, probably can't see it in the pic but the spacers have the ends machined down to fit into the holes. This gives a much tighter connection (actually probably me just being anal).

The spacers will have a band running inside that forms the scroll, this band will be welded to the spacers in place. This will allow the front panel or the back panel to be removed.

I still have to figureout what hole sizes I need to cut for the airflow. The air inlet hole will have some type of adaptor to attach to the flowbench within this adaptor I would like to be able to install reducers I can use for blower R&D to improve this design.



Edited By bruce on 1185311972

[Tony]

I guess the exit flow area of the scroll has now pretty much been determined by the radial spread of the holes in the end plates, and the length of the threaded spacers.

The flow area where the air enters the eye of the rotor needs to have at least a similar flow area, so as not to create a restriction point. Because the rotor width is only half an inch, the hole diameter will need to be made quite large, unless the rotor cover is flared outwards towards the entry hole.

While the back of the rotor is an important structural part, the rotor front cover only has to block airflow and seal the rotor blades, it could be made of very light gauge material.

[bruce]

The front cover is exactly the same as the rear, I used 11ga (.119) for both.

I can machine an inlet for the front cover to what ever dia or shape for R&D. So different inlet shapes can be tried. Well I should say the max dia would be the size of the hole in the front cover.

You are correct the outlet is pretty much a given now with my setup, it could be made smaller though. Final size should be ~1.5" wide by 7.5.

Basically as we discussed in emails I'm going with your calcs and building something that might or might not work? Not alot of $$$'s invested compared to the cost of trying to buy something that might or might not work. The time spent machining when added up is not all that much.

With luck it works the first time or with a little redesign

[ThomasVaught]

Great Project Bruce

Tom V.

[bruce]

Couple more pics . . .

[bruce]

Another . . .

[bruce]

Last one for today . . . they should all be self-explanatory

[Tony]

The exit area is ideal, 1.5 x 7.5 = 11.25 square inches, about the same area as a 3.75" round pipe. That is pretty much what I am using, 90mm on the Aerotech blower (3.5") and on my drum housing 4.0". I expect your new housing should work extremely well.

Flaring the rotor, is optional. But with no flare the rotor intake area may also needs to be fairly close to eleven to twelve square inches. That requires a hole circumference of 22" to 24" which is large, but not impossibly large. Would having a seven and a half, to eight inch diameter entry hole be feasible ?

[115-1172523331]

Tony, Does the rotor cover flare you are talking about look like the one in your yellow rotor from earlier posts? Following with interset! -- Doug

[Tony]

Yes Doug. The biggest design problem is getting the air into the middle of the rotor without creating an excessive restriction and pressure drop at that point. By flaring the rotor outwards at the hub, the entry diameter can be made less, without reducing the available flow area.

What actually happens is that the blower will generate some fixed static air pressure at zero flow, this is very predictable from rotor tip velocity. As mass airflow increases, the developed pressure always falls off. That drop off in performance is entirely due to pressure drops within the blower itself. Any internal pressure drop is not going to help blower efficiency either. So for most air, with least horsepower, internal flow restrictions should be minimised.

So the idea is to make all the internal flow paths approximately the same flow area, at least for a blower like this where the pressure ratio is fairly low. Say 14.7psi absolute in, 15.7psi absolute out.

Some sort of exponential cone (horn) shape will always be optimum at the intake, but it could be built into either the rotor, or the housing, or even partly in both. A slightly flared rotor also has the advantage that it is less likely to scrape the housing at the outside periphery where there may otherwise be very minimal clearance.

One of the many advantages of a blower like this is that the output pressure will stay fairly constant over a wide flow range, but only if it is built with well proportioned internal flow areas.

[ThomasVaught]

If you made an air inlet adaptor piece that could be attached to the front plate following the NACA air inlet/ "Flow Nozzle" dimensions it might work very well with what you have already. Basic geometry is for a inlet (when the inlet is laying flat on a table is: "Up 3 and out 5" so if your plate is 1"
thick, take 1.000" and divide by 3 and you have .333" which you then multiply by 5 so you get 1.665" at the outside diameter of the flare. On a 4" opening the curve would start at a 4" dimension at the plate and would be 5.665 at the widest edge of the flare.

Confusing?

Tom V.

[bruce]

[MMack]

It may not be an issue, but doesn't the hole size of the fan inlet somewhat restrict the total flow/and/or the total depression? Going larger than needed is a waste, but too small and you may bottleneck the whole thing. It should be a little bigger than the orifice for the highest flow you can generate, shouldn't it? I know that the inlet can flow more than the orifice as it can have a greater pressure drop across it, but the more you burn there, the less you have in the box.

Mike

[bruce]

I got something drawn up last evening that is to my liking and I should get it machined up this weekend. The adapter I'm going to machine will be able to accept a sleeve insert so I can play with different sizes of inlet bores. But it will not be larger in area than my outlet.

[Tony]

Mike, in theory, the flow geometry we are dealing with here could be thought of as being fairly similar to that of a poppet valve. For a constant flow area, valve "lift" needs to be one quarter of the incoming throat diameter.

A four inch entry pipe could flow into a rotor that was one inch wide at the entry point, and the flow "curtain area" would then equal the pipe flow area. That is the basis.

The other aspect is that the air in the rotor is accelerated, and it can do that more easily if the flow area tapers down smaller as the air moves outwards tangentially. That is a bit of a tall order for a home made rotor, but it can be approximated by flaring or dishing the hub area outwards between where the blades start, and the entry hole. Fortunately the blade area can have parallel walls, and the blades themselves need not be made very long.

We can get away with all this reasonably well because the pressure ratio is quite low.

Thinking about this a bit more, my approach to building a rotor cover would be to have a four inch entry hole, and that provides a flow area of 12.5 inches.

If my basic rotor design width was half an inch, the rotor flow area starts to fall to less than 12.5 square inches inside a pitch circle with a circumference of 25 inches, maybe eight inches pitch circle diameter. So the flare would be an inch wide at four inch diameter entry, and taper down to half an inch wide at eight inches diameter.

That would also place the half inch wide rotor about dead center in an inch and a half wide housing.