Member postings for Turbine Guy

I ran Axial Turbine 3N with everything the same as the last test shown on the sheet of the previous post. It took approximately three hours to heat my shop to 70 F. My shop was at 47 F when I decided to start testing and I made a run at that temperature. The maximum speed obtained at 47 F was 5,600 rpm. I made another run at 60 F and the maximum speed obtained was 5,900 rpm. I then made a test at 70 F and the maximum speed obtained was 6,000 rpm. Since the test with the 0.033” nozzle had the same performance with everything the same as in the last post, I opened up the nozzle to 0.035”. The maximum speed reached was 5,900 rpm. I did multiple tests at 70F with the 0.035” nozzle and with fresh oil added before each test. The maximum speed reached was 5,900 rpm in each test, so the results were repeatable. Since the maximum speed dropped with the larger nozzle, I will assume the 0.033” nozzle size is the optimum for running with my Master TC-96T airbrush compressor. I updated the following test sheet to show the results of the last test. I did confirm that it is important to have the temperature close to the assumed 70 F when running a test. I also found that these latest ball bearing would only run for a short time without adding oil before the turbine would start losing speed. The ball bearings I had been using would make multiple runs before the speed would start dropping. I also found that the exhaust port had to be at the bottom even when running on air. With an inlet temperature of 70 F, the air after it expands in the nozzle is cold enough to condense the moisture in the air if the humidity is very high.

Hi David,

I use the free version of Onshape and have been very pleased with it. I opened the public tab and at the top of the screen there is a search box for the public files. I typed in TAPBOX_LT and the solid model you are looking for came up plus some copies of it that others have made. See if this works for you.

Byron

I found the optimum gap between the rotor and cover plate for Axial Turbine 3N to be 0.031” when running on my Master TC-96T airbrush compressor, using the APC 4x3.3 EP propeller, and with a 0.032” nozzle. The maximum speed obtained was 5,900 rpm and the corresponding torque was 0.290 in-oz. I next increased the nozzle size to 0.033” and ran a test without changing anything else. The maximum speed increased to 6,000 rpm with a corresponding torque of 0.300 in-oz. I ran tests of my Tangential Turbine 3 SD Gap and Drag Turbine 4 for comparison and made the following test sheet. I found that I get a significant loss in performance when the tests are run at temperatures much lower than 70 F. My shop is uninsulated and takes approximately a half hour to heat up on these cold winter days. This is not a problem if I need to run several tests like I did to make this test sheet. The next test I want to make is enlarging the nozzle size of Axial Turbine 3N. This test will only take a few minutes to perform but will require me to heat my shop up to approximately 70 F and make another test with the existing nozzle to verify that the results are consistent. If I get approximately the same performance with my existing nozzle, I will increase the nozzle size in small increments until I get a loss in performance. The nozzle size I end up with will be slightly less efficient than the previously tested nozzle, so I want to be very careful that my tests are valid.

Hi Roger,

Thanks for your kind remarks. I was glad to see someone comment about the outstanding rotor Mike Tilby made. I am disappointed that there has not been more remarks.

The problem with the new ball bearings appears to have been caused by trying to run them at too high of a speed. I made several runs with the larger propeller and the results have been very consistent. The only thing I loose by testing at lower speeds is the amount of power produced. Since I only need to see if a change increased or decreased the performance, it doesn't matter what speed I do the testing at. I am going to start my testing again and do the tests with the larger propeller.

Thanks for your comments,

Byron

I noticed that after just a few tests of Axial Turbine 3N I could not reach the same maximum speed when comparing tests with everything the same. When I did tests to optimize my other turbines, I did not have this problem unless something damaged the ball bearings. I think that it is very important when testing a change, that the results of tests be close enough to confirm the results. The tests I started with Axial Turbine 3N were to find the optimum distance of the rotor from the face of the nozzle for a given nozzle size, propeller, and a given energy source. I add or remove shims and see if the maximum speed increases or degreases from the previous test. After finding the optimum spacing, I go back to the number of shims I used on the first test and see if the top speed is approximately the same as the first test. If the results of the first and last tests that had everything the same turn out to be very close, I feel the results are valid. I was able to do this many times with the tests of my other turbines that were checking something that I could change back to what was used in the original test. Not being able to do this with tests I started with Axial Turbine 3N caused me to try to find out what was different. The only thing that changed was the ball bearings. I could not get the ball bearings I had been using, so I tried the closest available ones. When I received the last ball bearings the quality did not appear to be as good as the previous ball bearings. I am going to try using a new pair of these last ball bearings and do my testing with a larger propeller and see if keeping the speed much lower results in the performance being more consistent. I just made my first run with everything the same as the test of Axial Turbine 3N shown in the last post except for using the APC 4x3.3 EP propeller. The maximum speed was 5,400 rpm and the resulting torque was 0.243 in-oz.. I will do several more tests with fresh oil before each test and everything else the same. If the maximum speed stays approximately the same, I will assume that the new ball bearings are working at the lower speed and repeat the tests for finding the optimum spacing.

I ran the first tests of Axial Turbine 3N and the results are shown in the following table. In the first test the speed went over 28,000 rpm with the GWS EP 2508 propeller I had been using, so I changed to the GWS EP 2510 propeller I recently purchased. The table shows the power required by each propeller and the source of the propeller test data. I was very happy to find the GWS EP 2510 propeller since it doesn’t require too much additional torque and the speed gets high enough to be a reasonable comparison of the test results with the other propeller. This is an update of the last test results I posted. I added the torque found from the tests since I feel this is what should be compared when tests are made with different loads. The results shown for Axial Turbine 3N are for the best distance of the face of the rotor from the face of the nozzle plate. I tried adding shims to move the rotor toward the cover plate until I found the position that gave the best performance. All my other turbines did not run over 28,000 rpm with freshly oiled bearings, the GWS EP 2508 propeller, and the maximum continuous output from my airbrush compressor except Drag Turbine 4. In the test of Drag Turbine 4 the pressure had to be lowered to the value shown to keep the speed below 28,000 rpm. I plan to run a test with Drag Turbine 4 to see what it can do with the full output of the airbrush compressor. The drawing Axial Turbine 3N R1 shown in album Axial Turbine 3N shows the position of the rotor in the housing for this test. This drawing was revised to show the actual dimensions of the rotor and nozzle size used in this test.

Edited By Turbine Guy on 29/01/2022 18:18:10

I received the rotor Mike Tilby made for me shown in the picture below. As you can see from the photo his machining is outstanding, well beyond what I am capable of. He spent weeks making this rotor and I can’t thank him enough. I mounted his rotor on the shaft and sleeve I machined and will call this assembly Axial Rotor 3 R3. The R3 indicates the drawing for this assembly was revised three times. The original drawing showed the dimensions of the solid model of this rotor based on my understanding of what Mike intended to make. The later revisions were made for adjustments we had to make for machining or to fit in my existing nylon housing. I had already created a folder called Axial Turbine 3 for this rotor and it was intended to use my existing aluminum housing. I created a new folder called Axial Turbine 3N for using this rotor in my existing nylon housing. The second photo below is a photo of Axial Turbine 3N. I am updating the folder Axial Turbine 3N to show the parts, photos, and drawings. I will do the same for Axial Turbine 3 when I decide to use the aluminum housing.

I decided to add a nozzle for Axial Rotor 3 on the cover used for Axial Turbine 2. I broke a couple more drills trying to do the nozzles the same way I have done before and decided I would try something different. I bored a 1/8" (3.2mm) hole all the way through the Cover R2 as shown in the following drawing. This hole did not drift and broke through the cover in the right place. The plan was to make an insert as shown in the following drawing and use Loctite 290 to hold it in the hole. If the drill breaks in the insert, only the insert will be ruined and not the hole in the cover. I tried drilling the nozzle hole through before cutting the bevel on the first insert I made. The drill broke at about 3/4 the way through. Every one of the drills that were damaged, broke after going quite a ways into the hole. I think that the torque on the drill due to long contact is what is breaking the drills. I machined the bevel first on the next insert, so the drill only had to go through about half the distance. I ran the HSS drill at the top speed of the lathe and it passed all the way through. I installed Insert 2 and the second inlet tube in Cover R2 and used Loctite 290 to hold them in place. I added photos in my Axial Turbine 3 album showing the finished insert and the insert installed in Cover R2. After the Loctite cured over night, I tested the new nozzle for leakage and found the maximum pressure my air brush compressor could maintain with the 0.031" (0.79mm) nozzle size. The maximum pressure was 36 psi (2.4 bar) that matched the pressure I found when using my test nozzle, so the nozzle is working well.

I decided to go ahead and machine Nylon Housing 2 to where it could be used for axial rotors since the nozzle holes for the tangential rotor were not useable. It required approximately 0.008” removed from the faces for cleanup. I enlarged the bore for the rotor to approximately 1.250” to give a total clearance of 0.005” (0.0025” per end) with the new 1.245” OD rotor I plan to use in this housing. This diameter gives enough clearance for the shrouded axial rotors given to me by Werner Jeggli as shown in the following photo. I used his printed rotor since I was able to center it the best on the shaft and I wanted to see if the clearance was about the same all around. I also wanted to show how little material needed to be removed to balance the rotor. My photo of his cast rotor shown in my Axial Turbine 2 album required a lot of extra material be removed for balancing since it was off center a few thousandths. Even being off center it out performed the printed rotor due to the much smoother surfaces of the blades. With the extra wall thickness of this housing there was no slippage in the 4 jaw chuck during all the machining and the reamed hole for the ball bearings was a very light press fit on both ends. All the machining took only a few hours to finish the housing and the bores remained concentric and perpendicular to the faces. This housing should work well for testing the axial rotors.

I think I will be able to run the new axial rotor using Nylon Cover 2 and Nylon Cover 3, but I plan to make a cover plate similar to what I used on Axial Turbine 2 to compare the performance with the type of nozzle I have used on all my other turbines. The two covers type of nozzle could be very useful for closely spaced multiple nozzles if the nozzle efficiency is high enough.

The following photo is of Nylon Cover 2 tilted at an angle where the angled nozzle holes are a little easier to see. The reflection of the nylon makes it difficult for me to show how much of the nozzle holes was printed. Even though the smaller holes didn't get to the correct size or go completely through, they were partially made and in approximately the right places. I think if the design takes into account the limits of the printing process, the printed parts can be very useful.

These are the photos of the printed nylon parts.

The following drawing shows the design dimensions with the actual as printed dimension below in parentheses for the parts of the Nylon Turbine. Only a few dimensions on each part are shown to illustrate the accuracy of the printed nylon parts from Shapeways. None of the holes with a design diameter of less than 0.040” went all the way through. Housing 3 SD Nylon had a nozzle hole with an design diameter of .024” that had an actual printed dimension of 0.022” diameter that went all the way through. The description and drawing of Housing 3 SD Nylon is shown on the 21/09/2021 Post. The overall accuracy of these printed parts was quite a bit lower than on Housing 3 SD Nylon. Even a needle with a sharp point would not enter either accessible end of the smaller holes. The accuracy given by Shapeways for PA 11 printed nylon is +/- 0.3mm, +/- 0.012", so I expected way too much for the small diameters and closely spaced holes. I will have to decide what I can do with these parts. Without the holes going through all the way, I can’t open them up while using them to guide the drill as planned. Also the placement of the nozzles will require more accuracy than Shapeways tolerance will allow. The increased wall thickness of the housing will allow me to machine it without it deflecting too much but the thin walls of the inlet and outlet tubes do not look strong enough. The next post will show photos of these printed parts.

Edited By Turbine Guy on 26/10/2021 19:36:31

Edited By Turbine Guy on 26/10/2021 19:38:10

I ordered the parts shown in the following drawing made of printed nylon from Shapeways. The designs met Shapeways requirements and the parts have been printed and shipped. These parts will give me a little more experience in printed nylon and allow me to try several things. As shown on the drawing there are inlet ports for two different axial impulse rotors and one tangential impulse rotor. The inlet ports shown as Axial 1 will be used for running my existing Axial Rotor 2 shown in the Axial Turbine 2 Album. The inlet ports shown as Axial 2 will be used with an axial rotor that will be made later. The inlet ports labeled tangential will be used with my existing Rotor 3 SD shown in my Tangential Turbine 3 album. I plan to compare the performance of the single nozzles made of nylon with those made of aluminum. I will also compare the performance of the single nozzles that will run supersonic to the multiple nozzles running subsonic. The new axial rotor will use a single nozzle large enough to run subsonic so I will be able to compare the single large nozzle with multiple smaller nozzles.

Because I have tried so many different turbines and combinations of parts, it is difficult to tell from a brief description what the configuration includes in the turbine test results. To try to make this a little clearer, I have started to make individual albums for each configuration. If you are curious what the parts look like and/or want to see the drawings you can open the applicable album. This is a work in progress and I have only got the axial and tangential turbines started. I will also add albums for the drag turbines and any new turbines. Hopefully this will make it a little clearer what I am testing.

Edited By Turbine Guy on 12/10/2021 14:35:53

I finished all the machining of Housing 3 SD Nylon. Everything was fine until I tried opening the hole the inlet tube slides into. My setup for opening the hole apparently was not rigid enough and the drill wandered enough to break through. Because of this, I can’t use the nozzle. I thought that I would try using this housing with Axial Rotor 2 instead of the aluminum housing. When I opened the bore enough that Axial Rotor 2 had enough clearance the wall thickness became too thin. The slightest pressure on the OD of the housing caused contact with the rotor. I believe this makes the housing not rigid enough to try mounting the Axial Nozzle plate and running as an axial impulse turbine. I still believe that if the wall thickness is enough, the nylon housing will work fine. I am in the process of designing a new housing and nozzle plates made of nylon that will be stiff enough to work with Axial Rotor 2 and Tangential Rotor 3 SD. These parts will be designed based on what I have learned from machining Housing 3 SD Nylon.

I started the machining of Housing 3 SD Nylon by centering it in the 4 jaw chuck and reaming the bore for the ball bearings as shown in the following photo. I used the smallest change in drill sizes progressing up to just a little smaller than the reamer size. The jump from a K (0.281) to an L (0.290) drill size caused the housing to move in the 4 jaw chuck. I centered the housing again and tightened the jaws a little tighter and had no further problems. I hope that I didn’t lose much accuracy after needing to re-center the housing. The ball bearing took a light push fit to enter the exposed end. A dowel pin the same OD as the ball bearing had a tight slip fit all the way through the bore. I kept the dowel pin in the bore and tightened the 4 jaw chuck a little more before machining the rotor bore and housing faces. The machining of the rotor bore and housing faces was very easy and the finish quite smooth. The problem of gripping the extension of the housing could be helped by making the OD larger. When the ball bearing bore is machined to full size, the wall thickness of the extension is too small for nylon. I made the housing with minimum wall thickness for casting in bronze to reduce the cost. Extra thickness doesn’t add much cost when printing nylon and would be an improvement. 

Edited By Turbine Guy on 23/09/2021 14:19:26

I received the printed housings as shown in the following photo. I measured the housings, and the following drawing shows the design dimensions with the actual dimensions below them in parenthesis. The housings appeared to be stiff enough and strong enough to work running on air. The details of the velocity staging in the VS Housing Nylon looked very good and would be very difficult to machine. The printing process leaves a rough finish, and I am hoping the machining needed to make the critical dimensions will leave a smooth surface. I will start the machining on Housing 3 SD Nylon first and give updates as the progress continues.

One of the reports I read mentioned that they made their prototypes from plastic until they found their optimum designs. This cuts the cost down considerably and probably the amount of machining time. I haven’t machined any plastics for a while, so I am only guessing they are easier to machine. Shapeways uses a laser to sinter nylon powder into plastic layer by layer that uses SLS technology to print the PA11 nylon. Since the PA11 nylon is unaffected by the Kryox GPL 102 and 105 oils I use for my turbines, I ordered a couple of housings shown in the following drawing. I decided to try housings for my first plastic parts since they are the most expensive bronze castings. The cost of Housing 3 SD for nylon is approximately $20 and for bronze is approximately $180. The cost of the VS Housing for nylon is approximately $25 and for bronze is approximately $300. The Housing 3 SD is designed to be used with all the other parts of Tangential Turbine 3 SD. The VS Housing has 4 blades that circulate the flow back into and out of the rotor like used by Terry turbines. This is something that I have wanted to try and is beyond my machining capabilities. I picked the PA11 nylon since the minimum thickness was even smaller than for bronze (0.5 mm vs 0.6 mm). If Shapeways printed housings hold the tolerances tight enough and the plastic is strong enough to use these housings for my turbines, it will allow me to try many more options. The dimensions on the following drawing were intended to be used to check if Shapeways had the step files scaled correctly. Shapeways did not mention any shrinkage for the printing so I will check the accuracy of the parts by showing the actual dimensions in parenthesis next to design dimensions after I receive the parts.

I ran Drag Turbine 4 and Drag Turbine 5 after adding the sleeve to the inlet tube of Cover 5. The larger diameter and extra length allowed the hose clamp to fully tighten and the pressure to reach it’s maximum without leakage. The following table was updated to show the latest test results. Even with the bearings just oiled, Drag Turbine 4 was able to spin the propeller at a speed of 28,000 rpm with a 20 psig pressure. The speed of the propeller went over 30,000 rpm before the 24 psig the airbrush compressor is capable of continuously was reached. With the right load this turbine could produce over 5 watts with the TC-96T airbrush compressor. Drag Turbine 4 is capable of producing the most power but Drag Turbine 3 is more efficient and better suited to energy sources capable of high mass flows at low pressures.

After seeing the improvement of Drag Turbine 3, I ran Drag Turbine 4 that only the cover is different. After reoiling the bearings, it reached a speed of 28,000 rpm turning the GWS EP 2508 propeller. After several short runs the maximum speed started to go over 30,000 rpm so I reoiled the bearings and the maximum speed was 28,000 rpm. Drag Turbine 4 gave me the best performance of all my turbines running on the TC-96T airbrush compressor. Cover 5 added the nozzle directed in the flow direction that I thought would help, but in previous tests did not work as well as Cover 3. I tried to run Drag Turbine 5 to see if the breaking in helped it. I couldn’t keep the airline from blowing off. The hose clamp was tightened until it bottomed out and the hose would still blow off before a pressure of 20 psig could be reached. I cut off the end of the airline where it had been clamped and the hose clamp still bottomed out and the hose blew off. The inlet tube is too short and has a smaller OD than needed. I used Loctite to add a longer and larger OD extension as shown in the picture below. I will try running the drag turbines again tomorrow after the Loctite has fully cured.

Edited By Turbine Guy on 03/09/2021 21:01:01