Sharing information about model turbines
Turbine Guy | 14/10/2019 21:54:42 |
541 forum posts 578 photos | The following chart is updated to give the latest performance of my Turbine 3. The drawing showing the changes that gave me the best results will be added in the next post. I ran several tests with air after I got the maximum performance with all the changes I tried. The results stayed the same for both with and without the reversing chamber. The improvement in performance running on air was very small. The speed increased by 500 rpm without the reversing chamber and 600 rpm with the reversing chamber with the improvements. The turbine still made slightly more power without the velocity staging running on air. Apparently, the back pressure caused by the second stage was enough to keep the performance from getting better than with a single stage. |
Turbine Guy | 14/10/2019 22:07:45 |
541 forum posts 578 photos | The following drawing shows some of the important dimensions of my Turbine 3 after getting the best performance I was able to obtain with velocity staging. Making the outer edge of the reversing chamber as thin as possible made the biggest improvement. This opened up the space for the steam or air exiting the second stage. |
Turbine Guy | 16/10/2019 18:51:48 |
541 forum posts 578 photos | I ran Turbine 3 on air after Loctiting the stainless steel inlet tube back on the housing. The speed never got back to the speed the turbine was running with the reversing chamber installed but my airbrush compressor was shutting down periodically. This indicated the turbine nozzle was partially blocked. I cleaned out the nozzle and ran the turbine on air again. The speed of the turbine went back to the same maximum speed it had before with velocity staging. Running the turbine on air immediately after running on steam appears to keep the ball bearing oil from gumming up. I wasn’t sure if the blockage in the nozzle occurred before or after the run on steam so I ran the turbine on steam again. After cleaning out the nozzle, Turbine 3 reached a maximum speed of 32,500 rpm with velocity staging. The resulting power of 12.4 watts was about what I hoped I would be able to achieve with velocity staging. This is an increase in power of approximately 4.5 watts with velocity staging. This increase in power is a little misleading. It should be pointed out that if the turbine was turning a load that could be adjusted like the motor running as a generator used by Werner in his testing, the power without velocity staging of Turbine 3 would be around 11 watts at 32,500 rpm. The power added by the velocity staging would be approximately 1.5 watts compared with the power that could be achieved at this speed without velocity staging. Using a propeller for the load results in the speed being set by the torque required to turn the propeller. The speed was very consistent during the short run on steam. Also, the relief valve on the boiler was constantly releasing steam so the pressure of 50 psig (3.4 bar) was the maximum the boiler could produce. I ran Turbine 3 on air after the run on steam and the maximum speed obtained was 22,600 rpm so the friction of the ball bearings has not changed. I am very pleased with the performance of Turbine 3 running on steam with velocity staging. The following chart is updated to show this test. Edited By Turbine Guy on 16/10/2019 19:02:11 |
Turbine Guy | 07/12/2019 18:42:02 |
541 forum posts 578 photos | I tried running an APC 4x3.3EP propeller on my turbine 3 to see if I could confirm my estimates of power. The speed I obtained both with and without the reducing chamber was 4,250 rpm running on air from my airbrush compressor. The actual power of approximately 0.5 watts I obtained using the performance chart for that propeller was about half or what I estimated. I had to make a lot of assumptions in the estimated power for the GWS EP2508 propeller, so I searched the internet for some actual test results for this propeller. I found a report titled Reynolds Number Effects on the Performance of Small-Scale Propellers. In that report I found the charts shown below for the actual performance of the GWS EP2508 propeller running in the speed range of most of my testing. The static power coefficient Cpo for the speeds ranging from 10,000 to 25,000 rpm was approximately 0.03 from fig. 8 chart (b). With this value of power coefficient, the power at 14,500 rpm is approximately 0.58 watts. This power is about half of what I estimated for this speed. I compared the values for the power coefficient from this report with values from APC for one of the tested propellers and they were almost identical, so I think the APC charts are valid. The report indicates that my estimates for the power using the GWS EP2508 propeller is almost double the actual power. Because my testing indicated that the gains in performance with each change matched what Dr. Balje’s methods estimated for the change, I went back to his Study of High Energy Level, Low Power Output Turbines to see if I missed something. In this report he states that the profile losses and consequently the overall losses, is almost doubled by decreasing the Reynolds number by a factor of 4 from the assumed value of 2 x 10^5. I had assumed that the Reynolds number effects were included in the equation I used for the estimated efficiency. I found this equation doesn’t include the effects of Reynolds number. When I corrected my estimated performance for the effects of the Reynolds number, the estimated power was approximately what the APC chart and the report Reynolds Number Effects on the Performance of Small-Scale Propellers gave. This was true for all the tests except the tests with the GWS EP2508 propeller running on steam. Those tests had speeds above 28,000 rpm which was outside the range the actual power coefficients were given. It takes a combination of a very low power input and very small dimensions to get this low of a Reynolds number but all my testing on air was in that range I will add the updated chart showing the new lower values of power in my next post. |
Turbine Guy | 07/12/2019 18:48:28 |
541 forum posts 578 photos | The following chart shows the updated performance described in the preceding post The values for running with air should be good estimates. I cannot verify the powers shown for the testing on steam but believe they should be reasonably accurate. |
Turbine Guy | 21/12/2019 11:53:53 |
541 forum posts 578 photos | The testing of the very small propellers can be greatly affected by friction and flow blockage. The following picture shows the large portion of the propellers flow area that is blocked by the relatively large rotor housing. Since the propeller is only about 0.4in. (10mm) away from the housing, the effects of flow blockage might be substantial. I did more research on the effects of Reynolds number and found most other sources showed much less effect than I was calculating using Dr. Balje’s guidelines. Dr. Balje in his summary stated the Reynolds number effects should be minimal for most turbines, so I don’t think I am estimating them correctly. Because of these concerns, I tried another way to estimate the power that will be shown in the next post. |
Jens Eirik Skogstad | 21/12/2019 13:30:27 |
![]() 400 forum posts 22 photos | Although the horsepower is not high, the horsepower / torque will increase at the reduced speed via the gearbox.See at my steam turbine in this link.. Model steam turbine with 79 mm 4 blade propeller. **LINK** Model steam turbine with Graupner 65 mm 2 blade propeller (in fault rotation direction) **LINK** Edited By Jens Eirik Skogstad on 21/12/2019 13:32:01 |
Turbine Guy | 21/12/2019 13:59:47 |
541 forum posts 578 photos | The following picture shows the setup I used to find the stall torque of my turbine 3. I used a 3/8 oz. fishing jig as a weight. To confirm that the weight of the jig was correct, I measured the jig and used the lengths, diameters, and densities of the steel hook and lead to calculate the weight. The calculated weight was 0.38 oz. I held the propeller in the horizontal position, placed the jig on the propeller in the approximate position shown, and turned on my airbrush compressor. I then moved the jig to a position that it could keep the propeller from rotating when I removed my hand. Next, I moved the propeller by hand to where the left side was slightly above and then slightly below the horizontal position. The jig could hold the propeller from turning in all three positions with the full air pressure of 24 psig. The picture below was taken in the last position tried. The distance from the center of the turbine to the point of contact of the jig on the propeller was 0.54 in. in all three positions of the propeller. The stall torque is 0.54in. x 0.38 oz. = 0.21 in.- oz. The rotor velocity coefficient can be calculated from the stall torque with the following equation. ψR = [(2gTst/(wDciψN) – cosα]/cosβ ψR = The rotor velocity coefficient ψR= [2x32.2x 0.00109/(0.000483 x 0.102 x 1324 x 0.96) – 0.933]/0.906 = 0.21 |
Turbine Guy | 21/12/2019 14:53:08 |
541 forum posts 578 photos | Posted by Jens Eirik Skogstad on 21/12/2019 13:30:27:
Although the horsepower is not high, the horsepower / torque will increase at the reduced speed via the gearbox.See at my steam turbine in this link.. Model steam turbine with 79 mm 4 blade propeller. **LINK** Model steam turbine with Graupner 65 mm 2 blade propeller (in fault rotation direction) **LINK** Edited By Jens Eirik Skogstad on 21/12/2019 13:32:01 Hi Jens, I'm glad to see someone with another turbine. Please tell us more about your turbine and your testing. Any information on model turbines is greatly appreciated. |
Jens Eirik Skogstad | 21/12/2019 16:04:28 |
![]() 400 forum posts 22 photos | Hi Turbine Guy.. see at this link where i wrote about steam turbine. **LINK**
|
Turbine Guy | 21/12/2019 17:12:10 |
541 forum posts 578 photos | Hi Jens Erik, I followed your link and read all the posts. You and others in this link added a lot of useful information. Thanks for contributing to this thread. |
Turbine Guy | 21/12/2019 20:07:29 |
541 forum posts 578 photos | The rotor velocity coefficient of 0.21 found in the post of 21/12/2019 includes all the flow losses including the Reynolds number and empty blade losses. This velocity coefficient can be used to calculate the hydraulic power. The hydraulic torque is the maximum torque that can be obtained for a given rotor tip speed, mass flow, spouting velocity, rotor diameter, nozzle angle, and blade angle without any losses such as friction or windage. All of these parameters except the rotor tip speed were given in the 21/12/2019 post and are shown again below. ψR = The rotor velocity coefficient The equation for the hydraulic torque, Th, is Th= wD/2g[ψNCicosα(1+ψR) – u(1+ψR)] = 0.000483x0.102/(2x32.2)[0.96x1324x0.933(1+0.21) -u(1+.21)] = 0.000000765[1435-1.21u) ft-lbf x12in/ft x 16oz/lb =0.000147(1435-1.21u) in-oz u= the rotor tip speed, ft/sec u= πDN/60 = πx0.102xN/60 = 0.00534N N = Turbine Speed, rpm Th = 0.000147(1435-1.21x0.00534N) in-oz Edited By Turbine Guy on 21/12/2019 20:10:07 |
Turbine Guy | 21/12/2019 20:21:33 |
541 forum posts 578 photos | The hydraulic power, Ph, is given by the following equation. Ph =ThN/63000xlbf/16ozx746watts/HP=0.00074ThN watts Ph = 0.00074ThN, watts These are the equations for the hydraulic torque and power for my turbine 3 at the maximum output of my airbrush compressor. I will add a table showing the values of these at given turbine speeds and include the values of the power, P, from my turbine 3 spreadsheet that includes friction, windage, sonic losses.
Edited By Turbine Guy on 21/12/2019 20:30:51 Edited By Turbine Guy on 21/12/2019 20:36:53 Edited By Turbine Guy on 21/12/2019 21:07:33 |
Jens Eirik Skogstad | 21/12/2019 20:26:31 |
![]() 400 forum posts 22 photos | Deleted due double post. Edited By Jens Eirik Skogstad on 21/12/2019 20:33:31 |
Jens Eirik Skogstad | 21/12/2019 20:28:14 |
![]() 400 forum posts 22 photos | Mathematics related to steam turbine is not my strong thing. Only practical experimentation with failure and testing results in the creation of a fully usable steam turbine. With the reduction gear you get better use of steam turbine. Heavy turbine wheels + high revolution above 15 000 rpm = better torque due to stored energy in "flywheel". Small turbine wheel diameter is easy to get high revolution than a large diameter turbine wheel for same steam pressure/velocity.
Edited By Jens Eirik Skogstad on 21/12/2019 20:32:33 |
Turbine Guy | 22/12/2019 01:05:05 |
541 forum posts 578 photos | The following data is what I have been trying to post in my last two posts. Copying from a Word file didn't work well because the subscripts were enlarged and the table could not be added. I converted the Word file to a jpeg file and I think this will be easier to read. Edited By Turbine Guy on 22/12/2019 01:07:10 |
Turbine Guy | 31/01/2020 20:10:35 |
541 forum posts 578 photos | I had a discussion with Mike Tilby about whether the velocity can go supersonic with a converging only nozzle. My sources indicate that the velocity can go supersonic as discussed in the 19/05/2019 post. Because of the large Reynolds number effects of tiny nozzles, I decided to see if this works with miniature nozzles. My estimation for the nozzle maximum velocity running on air at 24 psig, and 1.74 lbm/hr from my airbrush compressor is 1,271 ft/sec. I estimate the sonic velocity to be 1,081 ft/sec. My estimations of the velocities included the pressure loss due to friction in the nozzle. The estimated impact force for these two velocities is 8.7 grams and 7.4 grams respectively. Any reaction force in this range will indicate my estimations are reasonably correct. If the force is above 7.4 grams it will indicate the velocity can go above sonic and will give the approximate amount. If the force is at or below 7.4 grams it will indicate the flow stays at or below sonic. The following photo shows the setup I used to measure the impact force. I supported a cover plate containing the nozzle I estimated the impact forces for in my lathe tailstock. The plate was angled to where the nozzle was perpendicular to the precision scale. As shown in the attached photo the impact force measured was 8.3 grams. This indicated that the velocity may go supersonic in an open space. The remaining question is the effect of the nozzle outlet being so close to the rotor blades or pockets. The pictures and diagrams shown in the post of 29/05/2019 indicate that some distance from the nozzle throat is needed to establish the supersonic velocities. The diverging section of a convergent divergent nozzle allows the gas to reach the maximum velocity before any contact. I doubt that the maximum supersonic velocity can be obtained without some space for the gas to reach full expansion before any contact. For this reason, I tried using the sonic velocity instead of the supersonic velocity in my analysis of the tests for my turbine 3 running on air. This assumes that the loss in velocity is due to the nozzle and the nozzle velocity coefficient is reduced accordingly. When I analyzed the test performance this way, the rotor velocity coefficient increased to approximately the values given in Dr. Balje’s Study of High Energy Level Low Power OutputTurbines. I forgot when I added this picture that it could not be rotated to the direction the picture was taken. Because of this I normally hold my camera horizontal when taking a photo I want to add to my album. If there is a way to add a photo that has the long dimension vertical into the album correctly or rotate the photo when viewing it, I would greatly appreciate someone telling me how to do it. |
Turbine Guy | 01/02/2020 15:38:41 |
541 forum posts 578 photos | I weighed the 3/8 oz. jig used to find the stall torque in the post of 21/12/2019 on my precision scale and the actual weight is 0.34 oz. The following sheet shows the update to my estimated performance using this weight and assuming the maximum velocity of the nozzle to be sonic as discussed in the last post. These changes brought the estimated performance very close to that found using the propeller power coefficient described in the post of 7/12/2019. The following is Revision A . |
Turbine Guy | 01/02/2020 15:44:03 |
541 forum posts 578 photos | The following picture is the one referred to in the last post. |
Turbine Guy | 07/03/2020 16:17:09 |
541 forum posts 578 photos | In the post of 31/01/2020 I stated that I thought some space was needed from the exit of a converging only nozzle to the first contact with the rotor for the flow to go supersonic. I couldn’t think of a way to provide this space with my tangential rotor, so I tried testing the axial impulse rotors Werner Jeggli sent me. Since the axial nozzle was in a cover plate and the axial rotor could be spaced from the cover plate by shims, I could test the effect of adding space. My testing indicated that the space was important and that the space needs to increase as the Mach number increases. Werner also ran tests that confirmed this. The results of both of our tests indicated that a significant increase in power could be obtained by optimizing this distance. Since my tests were with Werner’s rotors, I’ll leave it to him to provide test results for these rotors. For my tangential rotor, the clearance on the OD of the rotor is important to prevent flow from escaping so I can’t increase this clearance to provide space for expansion. Since running on air from my airbrush compressor results in the ideal flow only going slightly supersonic, I thought I would see the effect of enlarging the nozzle and getting the maximum flow closer to sonic. My calculations indicated that the pressure drop due to friction would decrease because of increased nozzle size and lower velocity. I gradually increased the size of my nozzle about 0.001 in at a time and measured the power. The power increased each time I made an increase in nozzle size. I stopped with the nozzle size of 0.031 in. since there is only 0.002 in. of material remaining on one side with this diameter. I made the following chart to show the significant tests running on air. I only show the maximum power for each of the shown parameters. Some of the results shown in the last chart I provided were caused by changes in friction of bearings after running on steam, a partially plugged nozzle, and changing the minimum thickness of the material around the nozzle. The results of the tests shown in the following chart have values that I get consistently with ball bearings that have only been used with air. I did not include the tests with velocity staging since it was not effective running on air. I will try to do something similar with steam using ball bearings with oil recommended for running on steam at high speeds. |
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
Sign up to our newsletter and get a free digital issue.
You can unsubscribe at anytime. View our privacy policy at www.mortons.co.uk/privacy
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.