working principles of a freepower stage

[daniel]

Hello!

My name is Daniel and I'm curious if you experts could enlighten me about the facts, that a freepower stage uses to create it's Rpm and torque. I've read already that it's all about gas temperatures and velocities and pressure ratios.
I was wondering if that principle could be used with ambient gas temperatures at all and for a study I've designed and manufactured a turbine wheel unit that is powered by an electric jet fan that puts out 3.5 kp thrust at 250 km/h velocity. Wich in turn spins the free turbine up to 25.000 Rpm
Besides that it's equally noisy to an Rc turbine, the torque output is rather sluggish. Let's say the input of 2000 watts equals an output of 100 watts.
I knew that it would be nothing near economic numbers but that far off I haven't expected

[racket]

Hi Daniel

Firstly we need a pressure drop to create a velocity increase in the nozzle guide vane ( NGV) stator in front of the freepower wheel .

The NGV turns the gases from an axial direction into a tangential direction so as to impact the freepower wheel and produce torque .

Once the gases are in the wheel there is a deflection of the gases in a direction somewhat opposite to the approach direction .

Its the total gas deflection that produces the torque , when we then add in the wheel blade speed that torque has a distance over time component that then produces horsepower/kilowatts .

If you use an electric ducted fan "engine" as your "gas producer" , there will be losses at the fan to start with , then there'll be losses in the freepower stage , but you should be able to get a 50% recovery of energy imput to the freepower .

Freepower wheels turn out their maximum torque then stalled ( stopped) with the gas producer at full power, the torque can be roughly twice the torque when the freepower is at full rpm and producing its max horsepower , the torque "curve" is more of a "downhill" straight line , checkout this dyno chart from Andrews turbine bike postimg.org/image/4g1ghuvct/

If you have 250 kph air being supplied to a freepower wheel then you would need to use an impulse turbine wheel with the air entry at near right angles to the freepower shafting as you can't get any velocity increase from any pressure drop.

Checkout this contribution about freepowers jetandturbineowners.proboards.com/thread/680/diy-turbines.

If you can supply a few more details about your particular setup I might be able to make some suggestions to improve power production/conversion

Cheers
John

[racket]

Hi Daniel

If your "gas" velocity is 250kph that equates to ~225 ft/sec , if using an impulse bladed freepower the blade speed needs to around half that , so lets use 100 ft/sec for the mean (mid length)
blade speed , assuming your turbine mean blade diameter is ~100mm - 4 inches , then the blade travels ~1 foot per rev , so you need a turbine rpm of ~100 rev per second or 6,000 rpm , any faster and power will drop .

The freepower turbine wheel will probably need to be considerably larger than 4" dia due to the low "gas" speeds involved, this in turn will mean lower rpm .

If your freepower turbine is "undersized" and unable to willingly swallow the gases entering it then your gas producer will have extra backpressure placed upon it and it can surge , this may increase the static pressure going into your freepower but the total pressure ( and mass flow) will be reduced due to the drop in dynamic ( velocity) pressure , we need total pressure not just static pressure .

Its absolutely imperative that you don't put "backpressure" on your gas producer , its outflow must be allowed to flow relatively unhindered through the freepower because you don't have any static pressure available to "force" it through .

Cheers
John

[daniel]

Hi!



First of all I did not expect such a detailed reply by the master himself, thanks a lot in advance! I'll try and share some additional and well needed info along this.

My "gas generator" in this case is a 90mm ducted fan with 12 blades spinning at approx. 30-35.000 rpm creating the above written numbers.



It's "outlet area ring" measures 5100 square/mm.

Now here comes the data of the turbine stage that first have been designed by "thought and feel" so I am definately off of what it should supposedly look like:



NGV

Inlet area of the blade section (14 blades) 5012 square/mm

Outlet area of the blade section tapers down to 2576 square/mm, my thoughts were to increase velocity there



Turbine numbers look somewhat the same because the NGV is kinda just a mirrored design of it.



The Diameter of the turbine stage is 100mm.



This is what I experienced during testing. One of these units choked the fan unit by a fair amount, the NGV alone w/o turbine already did. The fan went into surge state. Less NGV blades helped there but we got torque reduction.

My design uses a quite large (in my opinion) blade surface area to accomodate the lower velocities (I thought) if I had to design a hot stage I would have probably used half of their width.

I'm also familiar that a larger turbine diameter should aid in a better torque caused by the "leverage principle"



Because I was anxious to try that thing anyways, I helped myself out using two turbine units to double the outlet area and reducing the choking wich the fan seemed happy about. I read that you guys tend to defy a 50% larger freepower inlet area to gas producer outlet area minimum to prevent choking, well I came across that rule of thumb to late.

As much as I would fancy a short, direct way out from the fan outlet into the turbine stage, my current "design" has to split the outlet up into two outlets in a Y-fashion wich is not very efficient, of course. From there I catch the rpm with a belt drive and into a 90 degree angular gear.

My next design, IF there is any chance to make that thing run, and this is why I did not feel afraid to ask here, would be a larger diameter turbine stage with longer, less width blades and at least 50% larger area than the fan outlet, and of course attached as close to the fan itself as possible.



The goal was to power an electric rc helicopter with this, in a scale way, just like the turbine guys do, just with the ease of an electric drive. Of course nowhere near the efficiency of a regular electric direct drive, trade all that off for the scale feel.
It's already installed in a helicopter and made some test runs. But the lack of torque kept it from lifting off. It reached a headspeed of 650 rpm with dropped down to 500 under load. 


It was just an idea and I wanted to prove it workable (adding power to move a gas and extracting it again, shortly after), but as of now I am close to banning that Idea.

Edit: The freepower turbines of that size for rc aircraft putting out SICK numbers, 5-8 Kilowatts is the normal range of what you can reach. I find those drive concepts very fascinating when I imagine, a small, full size helicopter turbine puts out ~750 HP with their compressor wheel alone consuming ~ 1000 hp and all that with an ETA of the whole system of like what, 10% ?  Loads of energy stored in that liquid fuel ... 

[daniel]

And I figured I need my turbine blades more curved, with the leading edge facing directly into the incoming airflow ? 

[racket]

Hi Daniel

You're using a reaction turbine stage rather than an impulse stage , an impulse stage has all the pressure drop in the NGV and none thru the turbine whereas a reaction stage has a 50-50 split of pressure drop , but as you don't have any static pressure available to drop , your impulse stage needs to have the same outlet flow area between the vanes as the flow area into the NGV , the NGV is simply a turning device with no pressure drop required , to achieve equal flow areas the outlet of the NGV vanes need to be radially longer than the inlets , I used this method with my turboprop PJ experiment.

www.youtube.com/watch?v=uqMn0A2aUdo

If you have a 4" dia inlet to the NGV then you'll be needing probably a >6" dia outlet from them , so a >6" dia impulse turbine wheel. .

The turbine wheel needs blading with consistent flow area between inlet and outlet , an impulse design , do a Google for images .

Heres a few pics of the PJ's "NGV" and turb wheel after its "destruction testing" at >1700 F and 22,000 rpm , those mild steel bits didn't cope all that well especially when combined with the massive pulsations from the 4" dia PJ exhaust







With a >6" diameter freepower your maximum rpm will be ~5,000 , possibly best power at ~4,000 rpm .

The outlet flow area between the NGV vanes will need to have at least your 5,000 sq mms .

The current setup using a Y piece will produce losses in the "diffusion" of the flow to a lower speed , then more losses in the 2 X NGVs and because the turbine wheels have reaction blading there will be a "blockage" from not having any more pressure drop available, its perfectly understandable why it didn't work .

Ideally you want the flow annulus ring from the EDF to match the NGV inlet annulus ring without there being any steps so as to maintain the gas velocity , the gases are then turned in the NGV though consistent cross sectionally area'ed passageways to their outlets at a tangential angle of ~20 degrees to the side of the turbine wheel , the turb blades will need to have inlet and outlet angles at ~40 degrees , or whatever angle the velocity triangles produce.

Have a think on this , then get back to me

Cheers
John

[daniel]

John,

I'm currently designing the new wheel, but 6" is getting ginormous and 5" is the most I could do. I got the flow area right anyways, quite a pain to get the geometry right! you think that would be doable that way?

Dan

[racket]

Hi Dan

Yep , 5" should do the job , I just did a rough drawing and some calcs , with a ~ 40mm hub diameter going through from the EDF to the freepower wheel , the actual freepower wheel diameter will depend on how "shallow" the approach angle is to the turb , the larger the wheel the shallower the angle and the greater the energy conversion from the greater deflection , but at 5 inches you should be able to get that angle down to ~20 degrees , theres generally not a lot to be gained by going to a lower angle as the losses creep up .

As long as the flow area stays at or above your 5100 sq mms throughout the freepower stage it'll work , ideally you probably need a slight increase in area the further through the stage you go to compensate for frictional losses lowering the air speeds, what we mustn't have is any restriction of flow area as this will only put backpressure on the EDF with possible surge eventuating , we need the EDF to be flowing at its optimal flow with the motor converting as much power as efficiently possible to the airflow , its the total energy in the outflow from it thats important .

I'll look forward to checking out your new design :-)

Cheers
John

[daniel]

Hi

Here you can see the current state of the latest 5" one, and the previous 6" one wich is huge compared to the old one.



Btw this is a really great forum with even better people. Could read through all this collected stuff all day and night!
And you haven't even considered calling my idea stupid! I'm confident that it is quite strange though, but even then you help me out as good as you can. Thanks

[daniel]

Oh and your project as well as the PJ as one of them are incredible


Dan

[daniel]

Latest version, had to recalc the available area. Flow path isn't quite constant but it is the closest I could reasonably get. I'll probably hollow out the blades though, as the final part will be quite heavy. 700 grams it tells me ! Plastic part.

[racket]

Hi Dan

LOL..............on this Site we don't consider anything is impossible or stupid, even when proven it is ;-)

New wheel is looking good

Cheers
John

[daniel]

Hi,

Well that's pretty much the finished unit. Now I need to get them made and wait for em to show up!
Then we'll see what it's about! Thanks for your help John









Daniel

[racket]

Hi Dan

Looking good

Cheers
John