Testing Models

I ran the Tyne engine on steam with my Stuart Turner Twin Drum Boiler. This is my smallest boiler and was intended to be used with engines like the MSM Tyne or Stuart Turner ST oscillating cylinder engines. I thought I fixed the relief valve, but it leaked in this test preventing the pressure from getting to the 25 psig I planned to run at. After running long enough for the speed to stabilize at the lower pressure, I blocked the leakage out of the relief valve just long enough to get the pressure up to 25 psig and run at that pressure for a short time. The maximum speed obtained with the APC 15x10 propeller was approximately 700 rpm. The speed remained almost constant for all the time I had the 25 psig pressure. I updated the following chart to include this test. The Tyne was very easy to get running. It passed the slugs of water you get when first starting with steam and started running at a constant speed very quickly. The total time of the run was 8 minutes and 40 secs. It ran very smoothly for the entire length of the run. The displacement lubricator and steam oil provided by MSM appeared to give the piston and O-ring all the lubrication they needed. I oiled all the external joints needing lubrication with a light machine oil. The vibration of the Tyne engine was quite a bit less than the ST even though neither of these oscillators show any attempt to control vibration.

Deleted Post

Edited By Turbine Guy on 02/07/2020 12:21:31

The following table shows the maximum performance of my three steam engines running on air. It also includes the percent of the total available energy lost due to leakage for each of these engines. I wanted to point this out since it is the biggest loss for the simple oscillating cylinder engines and is significant for the more complex engines. I determined the amount of mass flow lost due to leakage by subtracting the mass flow required to fill the cylinder from the total mass flow going to the engine. The energy lost due to leakage is proportional to the amount of mass flow lost due to leakage divided by the total mass flow going to the engine.

Energy Loss due to leakage = (Total mass flow – Mass flow to fill cylinder)/ Total mass flow, %

The Tyne steam engine comes with almost all the improvements I made to the Stuart ST steam engine, plus a sealing gland on the piston rod. The one improvement I did better on the modified Stuart ST engine was the very tight clearance of the cylinder support shaft. The Tyne steam engine has tighter clearance on the cylinder support shaft than the original Stuart ST steam engine but not as tight as the modified Stuart ST steam engine. You can see from the table how important that clearance is.

To try to confirm that the most leakage of the oscillating steam engines was caused by the tilting of the cylinder, I measured the separation of the cylinder face from the standard. With the air pressure on, I rotated the propeller by hand and measured the maximum distance from the face of the cylinder to the face of the standard. The cylinder tilted away from the standard on the top side when the air was pushing on the bottom of the piston. The cylinder tilted away from the standard on the bottom side when the air was pushing on the top of the piston. I could see the tilting by shining a light on the side of the cylinder. I could only measure the separation on the top side because it was accessible to a feeler gauge. The maximum separation of the Tyne steam engine was 0.005” and stayed at that distance most of the time the port feeding the bottom side of the piston was open. I made a calculation for the leakage from the perimeter of the port assuming the average separation was 0.003”. The calculated mass flow was close to the loss found from testing. A loss of 65% of the available energy is so large that I tried to find an example of a way to eliminate the tilting. The following drawing was copied from E. T. Westbury’s article ‘The MUNCASTER steam-engine models’. This model supports both sides of the cylinder which could greatly reduce if not eliminate the tilting. This model is very sophisticated with many additional benefits but eliminates the simplicity that is the main virtue of the oscillating cylinder engines.

I measured the maximum separation of the cylinder face from the standard on my Stuart ST steam engine like I did with the Tyne steam engine. I couldn’t get my smallest feeler gauge into the gap, so the maximum separation is less than 0.002”. I could see that the faces were separating by shining a flashlight on the opposite side, but the gap was much smaller on the Stuart ST. Tightening up the clearance of the cylinder support rod has almost eliminated the tilting. This should have reduced the leakage mass flow much more than I found in the tests. I rotated the flywheel by hand with an air pressure of 22 psig and noticed the leakage from the gap between cylinder and standard faces was much lower than what was coming out of the exhaust. When I rocked the flywheel back and forth, the piston O-ring would seal again and the leakage through the exhaust port was reduced to a very small amount. When the leakage through the O-ring was almost stopped, the total leakage was very small. The floating O-ring concept that I am trying gets it’s name from the movement of the O-ring from one side of the O-ring groove to the other side each time the direction of the pressure changes. When working correctly, this movement reduces the friction and extends the life of the O-ring. I did these tests without adding any oil, so I added oil and did them again. With oil added, the O-ring moved across the groove and sealed the gap each time the direction of pressure on the piston changed. The floating O-ring worked when there was enough oil but not when most of the oil was gone. The oil is blow out of the cylinder fairly quickly when the engine is running, so the floating concept probably is not working with intermittent oiling.

I tried the floating O-ring concept when I added the O-ring to my Stuart ST oscillating cylinder engine because I was concerned that the friction added by the conventional O-ring would reduce the performance at low pressures. To compare the friction of the floating O-ring used on the Stuart ST to the conventional O-ring used on the MSM Tyne I needed to estimate the indicated power. The indicated power is the power produced by the steam engine without any mechanical losses due to friction or binding. It includes the gas pressure losses in filling and exhausting the cylinder. It also includes the losses due to compressing the gas if the exhaust port is closed before the end of travel of the piston. I need to have most of the engines dimensions to estimate the indicated power. I have a Stuart ST drawing giving me the dimensions for that engine. I don’t have a drawing for the dimensions of the MSM Tyne and have not taken this engine apart and measured the components. Since the two engines are almost identical, I assumed the clearance volumes, average port openings, and other key parameters would be close to the same value. The estimated indicated power shown in the following table is based on using the actual dimensions of the Stuart ST and Chiltern engines and assuming the key dimensions of the Tyne engine are the same as the Stuart ST. The mechanical efficiency of a steam engine is equal to the actual power output divided by the indicated power. Comparing the performance found from my testing of the MSM Tyne steam engine with the Stuart St engine seems to confirm that the mechanical efficiency is higher with the floating O-ring than with a conventional O-ring. My tests with packing indicated that the packing, after bedding in, performed about as well as the floating O-ring. I recently tried replacing the floating O-ring with Teflon packing. It showed promise of having even better performance than the floating O-ring, but I dropped the Stuart ST engine and damaged it before I could confirm this.

I put the piston packed with Teflon into my second Stuart ST steam engine. This engine did not have the standard and cylinder faces machined at the same time which was one of the biggest improvements on the other Stuart ST steam engine that was damaged. After several five minute long runs to bed in the packing, the top speed with the APC 15x10 propeller was a little over 850 rpm. The performance even without the machined faces was about the same as the MSM Tyne steam engine. Both of these engines have considerably lower efficiency than the Chiltern steam engine even though it costs less than either of them. The following picture shows the Stuart ST on the left, the Chiltern in the center, and the MSM Tyne on the right. You can see from the picture how much larger the Chiltern engine is. The advantage of the simple oscillating cylinder engines is that they can be made very small which is helpful for fitting into small model boats. The Chiltern engine would need to be used in a much larger model boat. The only non-oscillating steam engine in the size range of the Stuart ST and MSM Tyne engines I found is a Saito T-1. The Saito T-1 has a piston valve, is single acting, and uses close fits for sealing the piston and piston valve. There is no packing, piston rings, or O-rings used in this engine. I ordered a Saito T-1 engine and will see how it compares in performance to my other steam engines.

I received the Saito T-1 steam engine and it is about the same size as the Stuart ST and MSM Tyne Steam engines as can be seen by the following photo. The instructions are written in Japanese, so I couldn’t get any information from them. I did find on the internet that the pressure relief valve on the Saito boiler recommended for this engine is set for 1.2 bar. There is also an instruction manual for the Saito S3R that is written in English available from the Saito website. The Saito S3R is a three cylinder version of the Saito T-1 engine. The instruction manual for the Saito S3R steam engine gives all the information needed for operating and should be applicable for the Saito T-1. I'll give the test results for the Saito T-1 steam engine in the next post.

My first test runs of the Saito T-1 were with the APC 8x6 propeller that I thought would be best for this engine. I ran the Saito T-1 at different pressures up to 30 psig. I oiled the engine before each run with a new pressure and ran it long enough to obtain a maximum speed.  I then ran a series of tests with the APC 15x10 propeller doing the same thing. In all the first tests the airbrush compressor was cycling on and off. The Saito T-1 is designed to run at pressures around 1 bar or approximately 15 psig. Running the engine at that speed had my airbrush compressor cycling with the motor constantly turning off and on. Since the performance was best with the APC 8x6 propeller, I increased the pressure until the airbrush compressor was running continuously. The maximum pressure the airbrush compressor was able to reach with this engine was 35 psig. My tests with nozzles showed the mass flow was approximately 1.3 lbm/hr with the airbrush compressor running continuously at 35 psig. The following table shows the results of these tests. Running the pressure much higher than 15 psig will shorten the life and is not recommended. The performance at the pressures shown in the table illustrates what the engine is capable of. The Saito T-1 and the matching Saito boilers have been used in several small model boats as shown in videos on the internet. I will add the maximum performance of the Saito T-1 in the table showing the performance of my other steam engines in the next post.

Edited By Turbine Guy on 29/07/2020 12:11:48

Since the Saito T-1 steam engine uses most of the same parts as the Saito S3R steam engine, I checked on the Saito website what boiler was recommended for the S3R. The recommended boiler was the Saito BT-1L. The pressure relief valve for this boiler is set at approximately 2 kg/cm^2 (28.5 psig). This indicates that the Saito T-1 could be run at a pressure of 25 psig which is the typical design pressure of the oscillating cylinder steam engines. I looked at the exploded view from the Saito S3R instructions which showed the parts. The exploded view shown below is a copy of the portion showing the piston, cylinder, valve, and connecting rods. When I saw that the flow had to go through a hole in the piston valve, I was surprised that the Saito S3R could run as fast as it did in the testing. When I measured the parts necessary to estimate the indicated power, I found that the average opening area of the valve was approximately 0.0018 in^2. The average opening area of the Stuart ST steam engine is approximately 0.00084 in^2. Even though the Saito T-1 engine doesn’t take advantage of the larger average opening area that a slot in the piston valve would provide, it still has over twice the average opening area of the Stuart ST. This is because the hole in the piston is fully uncovered at one point of the piston valve travel. The maximum overlap of the ports in the Stuart ST steam engine is way below fully open as can be seen in the drawing shown on the post of 29/11/2019 in this thread. A significant advantage of using a hole in the piston instead of a slot, is the leakage area is much smaller.

I ran my Saito T-1 on steam from my twin drum boiler. Because it is single acting, it took several turns of the APC 8x6 propeller before it would run. After running a short time, the speed increased until a top speed of 2,400 rpm was reached. It held approximately this speed for the remainder of the test. The relief valve was leaking a little at the start of the test so I couldn’t measure the mass flow to the engine. The relief valve stopped leaking and most of the test was with all the steam going to the engine. The test with my turbine 2 with this boiler and at approximately the same pressure of 25 psig had a mass flow of 1.3 lbm/hr. The displacement lubricator supplied with the engine appeared to work well. The lubricator only has a removable top and no removable part on the bottom to drain the water. The instructions only say to add oil and there was space for oil after the first test. I ran the Saito T-1 on air after the test with steam and it reached a speed of 2,300 rpm with a pressure of 25 psig. I added the results from these tests to those of the oscillating steam engines as shown below. Unlike the MSM Tyne engine, the Saito T-1 had slightly more power with steam than with air at the same 25 psig pressure.

Edited By Turbine Guy on 07/08/2020 12:04:41

The following is a copy of a post I just made in my Model Turbines thread.

Posted by SillyOldDuffer on 06/09/2020 12:10:14:

One comment to show I've been paying attention relates to using the rpm of a propeller as a Dynamometer. They work most accurately at high rpm, but become less sensitive at low speeds. It shows on the graph as a long tail where power doesn't increase much over a wide range of low speeds, and then curves up nicely at the high end. There's a high-end limit too due to propeller tips going super-sonic but I don't think it's a problem. But power measured by a propeller at ordinary steam engine speeds won't be accurate in the way it does a good job at turbine rpm.

I thought about the comments you made copied above and went back and looked at some of my charts for the steam engines (especially in my Testing Models thread). I should not have extrapolated the performance based on the APC propellers to speeds below 1000 rpm. Even the APC chart for the 22" diameter propeller I used testing my largest steam engine does not show performance below speeds of 1000 rpm. I am going to copy these comments and add them in a post on my Testing Models thread.

Thanks for pointing this out,

Byron

This is why most engine dynamometers are hydraulic. You could get a higher turndown ratio with a propeller if it was shrouded, i.e. a ducted fan design, but there will still be a lower limit. The closer you can get to positive displacement, the wider the range will be.

The problem with models is finding a pump small enough with a pump curve that fits the application. You can pretty easily build your own calibration curve using a variable speed motor and current monitor to establish the speed vs. hp of your load device. You'd have to make a few assumptions such as electrical efficiency of the motor, but you can get pretty close. There are other operational factors involved such as discharge head and fluid temperature, but they can be compensated for or controlled. That's also true of a propeller - air temp, humidity and barometric pressure changes will affect the measurement accuracy & repeatability.

Hi Tim,

It's great to hear from you again. I miss all the information you add to this thread. I hope things are going well for you and that you are still planning on making a turbine.

Byron

Edited By Turbine Guy on 06/09/2020 23:59:00

Hello Byron,

I would use a brushlee dc motor as a test setup. This motors can be used as an alternator if you use the three phases with ballast resistors. These motors have an high effiency and a low no load torque.They start nearly from sero with the load torque.

Werner

Good idea. Brushless motors can be had in various powers starting low, so it should be possible to get a good match to model engines of different small outputs. And the range of each alternator can be tuned by altering the resistive load.

Don't know if it's still done, but car engine makers once tested motors off the production line by coupling them to a generator. Motor had to deliver rated power, easily measured in volts and amps, at various speeds.

Used as a dynamometer there's a problem with alternators. Their exact efficiency is unknown and it varies with speed. At rated speed it will be between 70% and 80%, so the measured power output will be 20% to 30% low, with error about ±10%. Much better than a propeller for indicating the output of a low speed motor but still imperfect. For Byron's comparative experiments the inaccuracy and error may not matter, but if necessary it could be reduced by calibrating the brushless-motor/alternator against a conventional Dynamometer and applying corrections.

By a conventional Dynamometer I mean one that measures work directly by lifting a weight or tensioning a spring, rather than indirectly as an alternator does. An alternator is simpler to set up and use.

Dave

Hi Dave and Werner,

Werner Jeggli uses the type of setup you are both describing. His post on 18/06/2019 in the Model Turbines thread https://www.model-engineer.co.uk/members/public_profile.asp?c=76819 shows his setup.

Thanks for the comments,

Byron

I made a few more tests with my turbines that are fully described in the following link.

Model Turbines

I thought it would be interesting to compare the performance of the model steam engines with the performance of the model turbines. The chart in the 07/08/2020 post above shows the performance of the steam engines I have tested. The following chart from the Model Turbines thread shows the performance of the turbines running on air. I like to compare the performance running on air since it eliminates all the issues of knowing how wet or superheated the steam is. When comparing these charts, please note the the torque for the steam engines is in-lbs and the torque given for the turbines is in-oz, so 16 times smaller.

Edited By Turbine Guy on 10/10/2020 18:25:23

Although I strongly feel that evaluating the effects of a given change are best done using air, for modelers planning to run on steam, the effects of using steam need to be considered. Also, since I have tried to concentrate on the simplest turbines and lowest cost steam engines, I want to show what my tests indicate you can expect running on simple boilers. The test I showed in the 29/05/2020 post of Model Turbines is an example of running a relatively easy to make turbine with a simple boiler. I think that a comparison of the results of this test with the Saito T-1 results shown in the 07/08/2020 post of this thread might be useful to modelers debating whether to buy or build a steam engine or steam turbine. The Saito T-1 was the least expensive and best performing of the small steam engines I have purchased and tested. It was designed to be run at the low pressures of small simple boilers like the Stuart Twin Drum boiler used in these tests. The size of these steam engines, steam turbines, and boilers would fit in model boats as show in the following photo. Examples of the Saito T-1 powering model boats are shown on the internet.  I will add a chart showing the test results and some additional commentary in the next post.

Edited By Turbine Guy on 12/10/2020 12:44:48

The following chart shows the performance of the turbine, steam engine, and boiler described in the last post. Both Turbine 2 and the Saito T-1 engine are affected by the way I tested them with propellers. The propellers blew cold air over the outside of them preventing the turbine housing and steam engine cylinder from getting very hot. The large mass of the turbine housing made it even worse by taking a relatively large amount of the available energy to change it’s temperature. The steam turbines have an additional disadvantage. In addition to the energy lost by condensation of the steam, the moisture in the steam reduces the efficiency by a factor of approximately 1.3 times the percent of moisture. Turbine 2 should have the unnecessary material removed from it’s housing and be insulated. The Saito T-1 engine has a relatively small cylinder wall thickness which reduces the mass needed to be heated to a minimum. It also has a very small heat transfer area to the standard, so the heat conducted to the standard and base is minimal. Also the thin Saito T-1 cylinder reduces the heat transfer area to the air so the radiant and convective heat losses are minimized. The video showing the Saito T-1 engine propelling a boat I mentioned in the last post can be seen at this Saito T-1 Video Link.