Novel flying machines / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Novel flying machines

first previous 26 27 28 29 next last

Move Thread

LAN_403

Ash Small

Mon Jul 07 2014, 11:13PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Patrick wrote ...

are you meaning your graph? or my 14 inch prop?

Neither really, just pointing out you need a prop with maybe 10 times the area in order to achieve the flight times you want from the same batteries. I don't have enough data yet to say exactly how big, and I don't know how much manouverability will suffer.

BigBad

Tue Jul 08 2014, 12:34AM

Registered Member #2529 Joined: Thu Dec 10 2009, 02:43AM
Location:
Posts: 600

Uspring wrote ...

According to the tables Patrick posted, the Î¶ for the 10" prop is 0.41, for the 14" it is 0.40. Both are quite less than sensational values. Note that the lower value for the 14" prop does not imply, that it gives less lift per power than the 10" one.

I expect that the props are optimised for some particular disc loading (to the extent that they are optimised).

Running them at lower disc loading may well not give optimal performance, although may give better performance anyway.

Patrick

Tue Jul 08 2014, 01:24AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

im looking at geometry and airfoil cross sections for now.
here:

and

we'll need at 16" to reduce the RPM, or motor load problems, whiich is why we start seeing those pancake motors.

involute gear link :

Dr. Slack

Tue Jul 08 2014, 07:46AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

I'm planning to re-do the plot, and see if I get the same thing, or something that looks like a straight line or a curve.

But I'd appreciate any other input

EDIT: It does look fairly straight, but maybe I don't have enough plots to tell if it turns upwards at the left hand end?

I think the point is that, compared to the original graph, which looks definitely curved, like it's asymptotic to both axes, the log-log looks straighter. Given that it's a collection of results from differently controlled experiments over several orders of magnitude, I'd call it 'ruler-flat', much the same way that an engineer would consider two quantities of the same order of magnitude 'equal'.

The next thing to consider, if you're building your own blades, is the distribution of the lift along the blade length. There are several arguments. One is to control the blade chord and angle of attack so that the airspeed is uniform across the disc. This means high pitch in the middle, low pitch at the tip. This would apparently give the lowest mean speed per momentum imparted, and so best efficiency. However, all sorts of sources from the elliptical spitfire wing, to tip-fences on airliners, suggest that a rapid change of velocity and pressure at the stream edge causes excessive induced drag, so reducing total efficiency. This would suggest a faster fall-off of pitch at the end of the blade. Just an idle thought, I wonder if a tip-fence is sort of half a duct? Now, a blade is not a fixed wing, the latter moves at constant speed across its width, the former is obviously fastest at the tip. This starts suggesting to me that the tip of the blade is 'expensive' in terms of viscous/wetted surface drag, and so should be made to lift a bit harder than it would be made to lift on a fixed wing. The blade has to end somewhere, so it is a problem that must be met somewhere, or ignored at the cost of best efficiency. Given that the blade tip is expensive for drag, I think this is why winglets or tip-fences may appear on fixed wings, but should be kept off rotor blades. So rather like the work-week, which should not really have a Monday, rotors should not have tips, they're just generally a bad idea. I guess this is where ducts appear from, allows the drag-expensive tip of the blade to generate a reasonable amount of lift, while mitigating the induced drag. Anyhow, figure out what distribution of lift you want along the blade, then this will drive what blade profile you end up with, not should it look like the LIPPS. What's the distribution of lift on commercial props you can buy, I'm not sure they're all idiots?

Patrick

Tue Jul 08 2014, 07:50AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Dr. Slack wrote ...

Anyhow, figure out what distribution of lift you want along the blade, then this will drive what blade profile you end up with, not should it look like the LIPPS. What's the distribution of lift on commercial props you can buy, I'm not sure they're all idiots?

fair enough, but the Paul lipps prop tapers to an unconventional tip for the reasons you state.
as for the pitch and lift, some of the APC MR's look like they have steep pitch at the root.

Uspring

Tue Jul 08 2014, 09:19AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

Dr. Slack wrote:

Anyhow, figure out what distribution of lift you want along the blade, then this will drive what blade profile you end up with, not should it look like the LIPPS. What's the distribution of lift on commercial props you can buy, I'm not sure they're all idiots?

I'd just make a survey on all the thrust and power consumption tables you can get your hands on with the biggest props you can accomodate. Look at the entries which correspond to the weight you need to lift. The shapes of the best props might give you an idea what to look for. Wrt to the weight of your craft, I'd choose battery weight about equal to the rest of the copter. Adding more batteries won't increase flight time significantly. Lowering the bare weight (and battery weight correspondingly) will, though.

Ash Small

Tue Jul 08 2014, 08:02PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

I'd consider leaving out the 'blade' near the root, or have less pitch here than the maths suggests, as this would reduce drag considerably without too much detriment to lift, as there is not much lift at the root anyway. Pitch will be least at the tip anyway, and if you go for high RPM, you can get away with less pitch across the whole blade.

We've not done 'prop drag' in detail yet, and I'm aware that zero pitch props still have drag, but I suspect there may be advantages to reducing pitch and increasing RPM, rather than increasing pitch and reducing RPM. For one thing the blades presumably won't be as wide. either. I'm not certain about this, but I think it's worth considering.

What I think is required here is long thin high RPM blades, with not a lot of pitch, but we'll see

EDIT: If gearing down is still required, I'd go for a reduction belt system, as it's much easier to adjust the gearing if required, for one thing.

EDIT: I suggest designing for the thinnest section that has sufficient strength. More pitch means you need a stronger prop, I think.

EDIT: The props you are currently using are all designed for a higher 'disc loading' than what we want.

Ash Small

Wed Jul 09 2014, 04:48PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Dr. Slack wrote ...

I think the point is that, compared to the original graph, which looks definitely curved, like it's asymptotic to both axes, the log-log looks straighter. Given that it's a collection of results from differently controlled experiments over several orders of magnitude, I'd call it 'ruler-flat', much the same way that an engineer would consider two quantities of the same order of magnitude 'equal'.

I did notice that the point I plotted at the extreme left hand side was to high (still getting used to log/log graphs), and I've drawn what I consider to be the best straight line.

I will give this some more thought, although at the moment I think it should be more curved.

For example, what can we deduce from the points where it crosses the X and Y axes?

Any suggestions welcome.

I do think we need to plot from 'measured values' in order to perform any meaningful extrapolation, although we can probably learn something from what we already have here.

EDIT: The units for this graph are the same as in the Wikipedia 'disc loading' graph.

EDIT: I'm expecting the point of intersection with the Y axis to be higher once we have some more relevant data to plot.

Dr. Slack

Wed Jul 09 2014, 05:01PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

if you have a relationship like
y = k * x^b
and you take logs, you get
log(y) = log(k) + b*log(x)
if you now plot a log-log graph, the slope is the power b to which x is raised, so it's handy for identifying power laws. I'll let you work out what the intercepts means from that log equation

Ash Small

Wed Jul 09 2014, 05:18PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Dr. Slack wrote ...

if you have a relationship like
y = k * x^b
and you take logs, you get
log(y) = log(k) + b*log(x)
if you now plot a log-log graph, the slope is the power b to which x is raised, so it's handy for identifying power laws. I'll let you work out what the intercepts means from that log equation

Ok, Thanks

In the meantime, though, the 'original' Wikipedia graph obviously cross the Y axis at ~12Lbs/Hp (maybe I didn't read the log/log graph correctly), and I'm now assuming that the reason 'our' measurement on that graph (the Wiki one) that I plotted earlier in the thread is so far from the other figures is that we are using a prop designed for higher disc loading than we really need, therefore there are extra drag losses.

Now, back to looking at that 'log(y) = log(k) + b*log(x)' stuff (I did look at this stuff briefly the other day, but it's a long time since I used it as a student or whatnot.)

EDIT: Ok, I've got as far as F(x)=constant.x^m, I'll look at it again later.

first previous 26 27 28 29 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob