|
|
Post by Richard OConnell on Mar 25, 2013 9:34:07 GMT -5
Looking good. Sorry to hear about the picture uploading problems, this seems to be a frequent problem here and I am going to work on figuring out the problem and hopefully resolve it soon.
|
|
|
|
Post by britishrocket on Mar 25, 2013 16:06:26 GMT -5
Thanks Richard, This is the error message I get when I try to upload an image file:- You must provide a valid auth token or dev key. see code.google.com/p/imageshackapi/Hope this helps. |
|
|
|
Post by Richard OConnell on Mar 26, 2013 8:08:21 GMT -5
Hmm.. You could try using either imageshack.us/ or www.photobucket.com/ directly and follow the instructions on uploading there. Just copy the BBcode or "Forum Code" generated on the last page and paste it in your message and it should work fine  I'll work on the problems with the uploader and hopefully I'll resolve some of these issues! |
|
|
|
Post by britishrocket on Apr 1, 2013 4:09:55 GMT -5
Hi Richard,
Many thanks for that, and thanks for your effort on behalf of all of us.
|
|
|
|
Post by britishrocket on May 14, 2013 6:33:39 GMT -5
Hello All,
Currently still working on the swirl gas centred injector prototype. Will post more on it once I have an assembled unit.
|
|
|
|
Post by Richard OConnell on May 14, 2013 17:26:42 GMT -5
Nice, always look forward to reading about your work |
|
|
|
Post by britishrocket on May 18, 2013 19:26:09 GMT -5
Manufactured and tested another swirl inducer today. This was done by cutting a two start 3mm metric trapezoidal thread on a brass round section. Also made a body to hold the swirler. I measured a flow rate of 0.067 kg/sec flow of water through the unit. The cone angle of the produced spray was very wide. I am starting to get some idea of the flow rates through these units and co-relating these to the calculated ones based on the trapezoidal thread geometry. That way I can get some idea of the discharge coefficient of these swirlers.
|
|
|
|
Post by britishrocket on May 22, 2013 2:35:58 GMT -5
Here is a quick snapshot of the swirler working. I took this to check the quality and operation of the spray. The unit can just be seen in the top centre of the picture. A thin swirling cone shaped film of liquid can be seen. At the bottom of the swirling cone, ring shaped sections are breaking away under the influence of centrifugal force. These then start to ligament and form droplets. This is classic swirl atomiser behaviour. I checked the size of these droplets using micam and they were in the range 160-180 micron. This indicates that the film thickness is in the region of 180 microns. The next stage is to drive the gaseous oxidiser through the centre of this cone, which will then further atomise it, hopefully leading to an order of magnitude decrease in drop size. Essentially this method uses the pressure of the liquid to effect atomisation first before employing the gas.  |
|
|
|
Post by britishrocket on May 22, 2013 5:17:31 GMT -5
The following images (taken from Rocco, Goncalves and Iha) show the classical swirl atomiser disintegration mechanism. The image with three shadowgraphs of a swirling cone shows the swirl angle increasing with inlet pressure and the break up length decreasing (which corresponds to a thinning of the swirling sheet) This is due to the greater tangential velocity produced by the higher pressure. The conical sheet exhibits longitudinal surface waves on it's surface. The second image (line drawing) Shows that ligamentation and droplet formation proceeds from the break off of a ring of liquid whose thickness is of the order of the film thickness and whose length is of the order of the wavelength of the surface waves on the sheet. Looking at the photograph of the swirl atomiser on test, the rings of liquid can be seen rupturing from the cone,ultimately rupturing further into droplets. The swirling cone with surface waves can also be seen.   |
|
|
|
Post by britishrocket on May 22, 2013 9:45:55 GMT -5
Just machined another swirler for test. This one is a triple helix version. This was produced by generating a 3 start 3 mm metric trapezoidal thread on a section of 12mm diameter round brass bar. A 3 start thread when the nominal pitch is 3mm would have to be machined with a pitch of 9mm, advancing the compound by 3mm for each helix. Since the Harrison M250 can cut a maximum pitch of 8mm, I had to advance the compound by 2.6 mm each time. This is of little importance in this application. The photographs below show the thread as machined, and with black, green and red ink to highlight the three helices.   |
|
|
|
Post by turbochris on May 23, 2013 9:06:38 GMT -5
this is priceless. excellent pics!
|
|
|
|
Post by britishrocket on May 23, 2013 14:53:39 GMT -5
|
|
|
|
Post by Johansson on May 26, 2013 2:03:09 GMT -5
Extremely interesting, thanks for sharing!
|
|
|
|
Post by britishrocket on May 26, 2013 18:14:32 GMT -5
No problem. The blog is starting to get up to date with the actual project progress now. I'm starting work on a gas centred swirl coaxial injector that is intended to overcome some of the shortcomings of the shear coaxial device. I will also post more on the results of the shear coaxial tests.
|
|
|
|
Post by britishrocket on May 26, 2013 18:18:08 GMT -5
What I also meant to say...and I really need to stop doing this, should just marshall my thoughts and put everything in one post...was that the injector research has been made a lot easier and more directed by having a reasonably accurate and repeatable method for assessing drop sizes.
|
|
I'll work on the problems with the uploader and hopefully I'll resolve some of these issues!
