| Post Info | TOPIC: Jellyfish 2 |
|---|
sjh7132
Guru
Posts: 1252
Date: Jan 27, 2010
| RE: Jellyfish 2 |
|
| GoVertical wrote:
Greetings, would the efficiency improve if the turbine was mounted on a pipe with less flex?
Probably, this is why I have a bunch of mechanical questions in another thread.
Steve
|
|
GoVertical
Veteran Member
Posts: 73
Date: Jan 27, 2010
| RE: Jellyfish 2 |
|
| Greetings, would the efficiency improve if the turbine was mounted on a pipe with less flex? |
|
sjh7132
Guru
Posts: 1252
Date: Jan 24, 2010
| RE: Jellyfish 2 |
|
| electrondady1 wrote:
stieve
i am so glad you built this milll
i had become convinced you were no longer interested in real world machines
and had become obsessed with simulations.
Oh I'm still obsessed with simulations, I just needed something real to compare the simulations to. I guess it's time to work on turbulence models. I may also consider a wind tunnel.
Steve
|
|
electrondady1
Senior Member
Posts: 430
Date: Jan 24, 2010
| RE: Jellyfish 2 |
|
|
stieve
i am so glad you built this milll
i had become convinced you were no longer interested in real world machines
and had become obsessed with simulations.
|
|
sjh7132
Guru
Posts: 1252
Date: Jan 21, 2010
| Jellyfish 2 |
|
| Bernd wrote:
Steve why are the values of tsr so much different during measurement ? (from 0,1 to 0,7 )
To be honest i'm glad to see that the tsr is there at you in reallity also much lower than in simulation, like i measured in reallity too.
Or should the horizontal axis labeled with "RPM" and not tsr ?
No that really is TSR. I had an oscilloscope on one phase of my alternator and could measure the frequency, and calculate the RPMs. I think the values are very wild because the wind changed and my turbine has a lot of mass. The meter that gave me current was bouncing all over and I had to try to take an average over a few seconds. My measurements are very crude. I made the measurements by making the turbine generate power into a voltage source. I measured the current into that source. power = current * voltage. So I was able to change the voltage and change the operating TSR of the turbine. I also accounted for power lost in the alternator. When I looked at the data, I though it was faster and more powerful after it was cleaned, but I guess you can look at it another way and see no change. I used a Teflon based car wax. Try it on your C-Rotor. -- Edited by sjh7132 on Thursday 21st of January 2010 11:46:27 PM
Steve
|
|
Bernd
Senior Member
Posts: 233
Date: Jan 21, 2010
| Jellyfish 2 |
|
|
Steve why are the values of tsr so much different during measurement ? (from 0,1 to 0,7 )
To be honest i'm glad to see that the tsr is there at you in reallity also much lower than in simulation, like i measured in reallity too.
Or should the horizontal axis labeled with "RPM" and not tsr ?
In know that it is very difficult to get usefull measurements under "garden" conditions.
The windspeed always goes up and down, sometimes in one second.
I realized that my Anemometer was more sluggish than my c-rotor prototyp, which made it even much more difficult to get correct results.. :)
Comparing to my teststand it was very hard to get results.
Actually it only worked well when i tryed to build averages.
Did you use your ironless Generator for measurements ?
How did your measurement system works ? Did it make a measurement every minute or something like that ?
The statement that the polish makes any different should be verified some times more and comparing for a longer time and same windspeed
because i didn't see a point above windspeed 1 in the first graph and under it it looks nearly to the waxed graph under 1. :)
(I still think it doesnt make any significant different) :-))
Bernd
-- Edited by Bernd on Thursday 21st of January 2010 11:40:42 PM
|
|
sjh7132
Guru
Posts: 1252
Date: Jan 21, 2010
| Jellyfish 2 |
|
| We had some wind today and I got some measurements. The wind (average) was between 10.8 and 12.0 MPH. But still my measurements were all over the place. First I tested my turbine dirty, then I cleaned and waxed it. The results are disappointing. The max TSR it will spin at is < 0.8 even though simulation said it would go faster. The max efficiency was only about 12% - 15% so I wouldn't recommend building this one. The one good piece of data I got was that waxing the surfaces did increase the speed and the power. See the attached graphs. I think this means my simulation needs some work. I'm starting to see the need for a wind tunnel.  -- Edited by sjh7132 on Thursday 21st of January 2010 10:52:46 PM
Steve
|
|
Bernd
Senior Member
Posts: 233
Date: Jan 21, 2010
| Jellyfish 2 |
|
|
Some month ago i measured the slats from the c-rotor only, without the deflector blades.
That seems to be comparable to half tubes. The slats where mounted with not angle of attack, just tangential.
The result was a slower rpm (tsr) an much less of power.
Thats the reason while i looking forward what will be happen at your rotor.
Bernd
-- Edited by Bernd on Thursday 21st of January 2010 04:21:29 PM
|
|
sjh7132
Guru
Posts: 1252
Date: Jan 21, 2010
| Jellyfish 2 |
|
| Bernd wrote:
Into the simulation the "Jellyfish" has a tail and seemed closed at the back and not open like the half tubes now.
Don't you think it is really important for the function ?
May be thats only a first test and you will build the Jellyfish later.
I also simulated the half tube design and it was pretty good. Since it's a lot simpler I thought I would try that first. 3 days and no wind. Apparently I'm helping design turbines for other people, not me. :-) By the way, a while ago I put my genetic algorithms on a normal half tube design with smaller tubes. It was allowed to vary the angle and number of tubes. It optimized it to 13 tubes and a slight offset angle, but the efficiency was only about 12 - 14%. So my simulations don't always show lots of power. -- Edited by sjh7132 on Thursday 21st of January 2010 04:01:39 PM
Steve
|
|
Bernd
Senior Member
Posts: 233
Date: Jan 21, 2010
| Jellyfish 2 |
|
|
Hello Steve and thanks for the drawing and the pictures.
Until now i can not imagine how a "anemometer design" (with very low effiecient) can change to a good windturbine design,
just modify the angle of attack of the half cylinder blades a little bit ?
Into the simulation the "Jellyfish" has a tail and seemed closed at the back and not open like the half tubes now.
Don't you think it is really important for the function ?
May be thats only a first test and you will build the Jellyfish later.
I'm really looking forward to your power results.
Bernd
-- Edited by Bernd on Thursday 21st of January 2010 12:37:43 PM
|
|
sjh7132
Guru
Posts: 1252
Date: Jan 20, 2010
| RE: Jellyfish 2 |
|
| Caleb wrote:
Do you have a domain yet? I've set up a web site on Weebly. It's free and easy to use (I was up and running in 20 minutes), but they ask you if you want to buy a domain name every time you publish your page.
Thanks. I used to host web pages for a side business, so I have several domains at my disposal. My main two are pinerypointe.net and infiniteeyes.com. I still have one server up too.
Steve
|
|
sjh7132
Guru
Posts: 1252
Date: Jan 20, 2010
| RE: Jellyfish 2 |
|
| Caleb wrote:
I was thinking of modifying my VAWT design to accomodate a jellyfish. But if the half pipes work, all the better. Do you think the power would drop a lot if the shaft through the middle was about 1.50" in diameter?
The shaft probably would not make much difference, but I suggest holding off. All I know about this thing right now is that it spins in the wind. Give me a few days and I'll know if it makes any power.
Steve
|
|
Caleb
Senior Member
Posts: 473
Date: Jan 20, 2010
| RE: Jellyfish 2 |
|
|
I was thinking of modifying my VAWT design to accomodate a jellyfish. But if the half pipes work, all the better. Do you think the power would drop a lot if the shaft through the middle was about 1.50" in diameter?
__________________
- Bryan
Mechanical Engineer turned stay-at-home dad. |
|
Caleb
Senior Member
Posts: 473
Date: Jan 20, 2010
| Jellyfish 2 |
|
| sjh7132 wrote:
Project wrote:
Hey Steve,
Is this your design or do you have a link to originator??
This is my design, if I can take credit for what evolved in my program. I just took the results of the evolution and used the round head part. It's a pretty simple design so I would not be too surprised if it's been done before, but I didn't copy anyone.
I'm working on an Openfoam basics web page. I'd put it in this forum, but not being able to link files that aren't pictures is a big limitation.
Do you have a domain yet? I've set up a web site on Weebly. It's free and easy to use (I was up and running in 20 minutes), but they ask you if you want to buy a domain name every time you publish your page. http://calebeng.weebly.com/ -- Edited by Caleb on Wednesday 20th of January 2010 01:03:50 AM
- Bryan
Mechanical Engineer turned stay-at-home dad. |
|
sjh7132
Guru
Posts: 1252
Date: Jan 20, 2010
| Jellyfish 2 |
|
| Caleb wrote:
Nice work Steve. I had a few questions: Did you set the pipe at the same angle as came out of the simulation or did you round it off a bit? What anemometer are you using and does it give instantaneous wind readings? Did the power output drop at all after removing the "tail?"
The jellyfish design was just a table of the radius every 4 degrees. It was within one sample of being a half cylinder so I just rounded up to a half cylinder. (the last sample was about half the normal radius.) I believe the angle should be very close (4 degrees or less) to what the original Jellyfish was. Maybe you are being mislead by the power point sketch. That sketch is not really to scale and the angles are off. The numbers are about the only thing that's correct. My anemometer is a Honeywell weather station. It does not read instantaneous values which makes testing difficult. I've only done crude simulations, but if the power dropped from the missing tail, it didn't drop by much. The solidity seems pretty high. I'm thinking of simulating to see if I could move those 'blades' outward a little and still have it work as well. (if it works well that is.) We got some wind this afternoon and it produced some power, but because of my crude test setup, I couldn't really tell how much. I need to improve how I measure power. -- Edited by sjh7132 on Wednesday 20th of January 2010 12:34:10 AM
Steve
|
|
sjh7132
Guru
Posts: 1252
Date: Jan 20, 2010
| RE: Jellyfish 2 |
|
| Project wrote:
Hey Steve,
Is this your design or do you have a link to originator??
This is my design, if I can take credit for what evolved in my program. I just took the results of the evolution and used the round head part. It's a pretty simple design so I would not be too surprised if it's been done before, but I didn't copy anyone. I'm working on an Openfoam basics web page. I'd put it in this forum, but not being able to link files that aren't pictures is a big limitation.
Steve
|
|
Caleb
Senior Member
Posts: 473
Date: Jan 19, 2010
| RE: Jellyfish 2 |
|
|
Nice work Steve. I had a few questions: Did you set the pipe at the same angle as came out of the simulation or did you round it off a bit? What anemometer are you using and does it give instantaneous wind readings? Did the power output drop at all after removing the "tail?"
__________________
- Bryan
Mechanical Engineer turned stay-at-home dad. |
|
Project
Veteran Member
Posts: 97
Date: Jan 19, 2010
| RE: Jellyfish 2 |
|
|
Hey Steve,
Is this your design or do you have a link to originator??
I am interested in the open foam as well.
Thanks,
Jerry
|
|
sjh7132
Guru
Posts: 1252
Date: Jan 19, 2010
| Jellyfish 2 |
|
|
This turbine is based on the Jellyfish that was the result of my genetic algorithms. I removed the tail and only used the round head and unusual angle. In simulation this is nearly as good as a Lenz2. But who knows if we can trust simulation?
One interesting feature about the Jelly2, is it seems to scale down with very little change in efficiency. So I've built this small model to test.
The radius is 8" (20.3 cm). The other measurements are in the diagram.
So far I've only had 0.5 MPH winds, and the turbine hasn't even turned. (The PVC is very heavy.)
I deliberately left the PVC surfaces dirty, and scuffed up in places. My plan is to get some power measurements as it is. Then take it down, polish it and put a good coat of car wax on it to make it smooth. Then take more measurements to see how important surface properties are.
__________________
Steve
|
|
|
