Ducted Fan, effect of tip clearance vs thrust 1

[vonsteimel]

I was wondering if anyone knows of any studies concerning tip clearance of a ducted fan and it's effect on the performance/thrust? Any information will be greatly appriciated.

We manufacture amphibious vehicles powered by a ducted fan. In our case, we have elected to keep the fan diameter as small as possible in order to increase aesthetic appeal. Some competitors have a fan/prop nearly double the size of our 30" diameter fans, but pay for it with hideous looks. Regardless, we are and will continue to use our small diameter fans.

Since there is a performance penalty in choosing to use a small diameter fan, we must keep our tip clearances at a minimum in order to maximize our thrust. However, lately our fan manufacturer has been having trouble meeting our tolerance specification on the fans. My question is; at what tip clearance does the thrust/performance start dropping off significantly? If at all possible, I want to increase the fan tolerance in order to reduce price and assembly times but how much tip clearance is too much?
Thanks,
CS

 

[amorrison]

I have read the gap should be less than 1% of the diameter.

 

[berkshire]

Here is a paper I found with my first hit on Google.
Which suggests that you do not need to worry about it, if your tip clearances do not exceed 5%. Maybe you should try Google too.
B.E.

http://www.personal.psu.edu/faculty/c/x/cxc11/papers/GT2011_46356_DUCTED_FAN_1.pdf

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[berkshire]

Some years ago I used to work as a contractor for a hovercraft company, we also were using a 30" fan. One thing we discovered was that flow straighteners in the duct behind the fan, almost doubled the available thrust.
The downside to this was increased noise.
So there you have it more thrust more noise, less thrust less noise.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[rb1957]

rather than what the gap should be, i think the most interesting word in the post is "lately" ... ie manufacturing were able to keep the tolerances required, but now can't.

why not run some tests with larger gaps ?

stators, "flow straighteners", will extract the swirl velocity component from the slipstream; funny, but understandable, that they'd make it noiser.

 

[btrueblood]

Um, Berkshire, the paper you cited was cool, but your conclusion seems a bit off. From the conclusion of the paper:

"Decreasing the tip clearance from 3.04 % to 1.71 % also increased hover efficieny of the system by 17.85 % at higher rotor speeds. When the tip clearance increased from 3.04 % to 5.17 %, up
to 18.1 % drop in hover efficiency was observed . "

Roughly 40% decrease in thrust coefficient going from 1.7% to 5.2% clearance.

Regarding the OP, "hideous looks" vs. better performance. Hmmm, I know which way I'd choose, and I'd bet the majority of red blooded males would too. Gimme bigger fans, any day.

 

[rb1957]

what if you had zero clearance ? if the blade tips were attached to a (very) short cyclinder ... then you'd need to seal this against the side of the cowl (in a recess)

 

[btrueblood]

Adds a lot of tip mass for your fan...maybe you want some roller bearings on the o.d. to help support it?

 

[rb1957]

it would certainly add alot to the inertia of the disc.

and i imagine there'd be some aerodynamic problem with a portion of the cowl inner surface rotating and the remainder being stationary

 

[GregLocock]

If tip plates on a rotor aided performance they'd be used already, and I think your rotating duct is just a big tip plate.

However my mind is now exploding because I can't see why it wouldn't work.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[btrueblood]

Greg, it would add a lot of windage loss, even if shrouded.

 

[vonsteimel]

I’d like to thank everyone for their contributions. The report proved interesting. I tried searching before starting this thread but I just didn’t use the right key words.

Talking with 2 different fan manufacturers, the industry has undergone a switch from using a router to trim fan diameter to using a band saw to save time & money. As a result, the tolerance has more than doubled. So this is at the source of our problems “lately” and it seems this will not improve unless they change their band saw method. However they have told me they can hold a +0.000 / -0.040 tolerance on the diameter even though our latest batch suggests otherwise. I know they have been able to hit the sweet spot using their “band saw” method in the past, but only after we returned the shipment several times.

As many reports suggest, 1% seems to be the sweet spot. Since our duct is 28.00 (R14.00) 1% = 0.14 which tells me the +0 / - 0.040 would be more than sufficient to stay within the “Sweet spot”. For some reason I was thinking we needed tip clearances in the 1/32 to 1/16 range (on a 28.00 dia duct) to avoid sub-optimal performance but the reports I have read suggest otherwise. I still have a hard time believing a 0.14 tip clearance would avoid significant performance loss, is there something I’m misunderstanding?

And as for the “looks vs best performance”, this is not the case. I said we have to accept a performance penalty but must keep out tip clearances tight to maximize thrust. Our products out-perform any other product in its class. And performance is only a relative metric, so if the individual has never used a similar product before, how would they distinguish poor performance from good performance? They cannot; it either works or it doesn’t. Until they go get experience with a variety of similar products, they have nothing relative to rate the performance against. Therefore looks become a major selling point when marketing an unknown product like ours. This is reinforced by repeated attempts to counterfeit our product all around the globe. These big ugly machines don’t have to worry about counterfeiters. Why? Because our product is sexy and sex sells…

 

[MikeHalloran]

Perhaps the fan manufacturers could accept a compromise between speed and finish, and use a belt sander instead of a bandsaw.



Mike Halloran
Pembroke Pines, FL, USA

 

[vonsteimel]

I doubt that would be feasible...? In our case, the fan is originally 42.75" in diameter & is trimmed down to around 30.00". In some applications they will trim up to 12.00 inches off of each blade!

It would be very time consuming and complicated to this accurately with a belt sander... Not to mention the smell.

Perhaps if it were just a 1/4 inch or so, then it could work.
Thanks for your input though!

 

[berkshire]

vonsteimel
Have you thought about buying the fans untrimmed, or rough cut but oversized and setting up a trim operation in house. That way you control the quality. I realize that you would like to have them right the first time. But unless you are a big enough customer to the fan supplier, you may not have that clout.
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[MikeHalloran]

Right; have them trimmed a little oversize by the bandsaw boys, then finish them to size yourself with a belt sander.



Mike Halloran
Pembroke Pines, FL, USA

 

[berkshire]

If you set up a rotating fixture, you can mount the fan to. You can use a belt sander ,drum sander or router to trim the ends to a perfect radius, plus or minus a few thousanths of an inch or tenths of a millimeter.
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[KENAT]

Out of interest, what do you currently specify as the tolerance on your drawings?

Are your supplier shipping fans that don't meat specified tolerance?

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​