Small Scale CHP Steam Engine Project

[buenijo]

I fired the steam generator again today. It held around 700F at around 800 psig for about an hour. Took about 20 minutes to get it dialed in. I took the temperature over 800F for a while. I also took pressure to 1200 psig for a while as well. No leaks in the steam and condensate systems. No leaks in combustion chamber plug. No leaks in the burner.

The main problem is LOW burner output. I previously considered this is caused by the combustion chamber being too small. But I am now convinced it is caused by insufficient air. I am forcing all air through the burner head. But I realized these burners are designed to burn in open air where a rich flame can mix with plenty of excess air. I know what I have to do to fix this - and it's gonna be a pain. Basically, I need to reconfigure the blower to force air into the combustion chamber bypassing the burner head. Yeah, I have no doubt I can do this. But I need to remove the steam generator.

Good news is it seems to work well (albeit at half the output that I would like).

ADDENDUM: I figured out a clever way to fix the burner problem. Should be a bit easier. I might not even have to remove the steam generator. We'll see.

 

[buenijo]

I made the necessary changes. I removed the steam generator and cut a hole in the base next to where the burner head mounts. I then replaced the steam generator. I did not have to disconnect the feed water or steam fittings because the feed water line is a flex hose, and the steam line connects to the condenser coil inside the pot still. I then adhered a PVC fitting to the underside on which the blower fan will connect.

I removed the steam generator lid for inspection, and sure enough there is a lot of soot on the tubing. This is a consequence of a rich propane mixture (i.e. insufficient air). During the next run, I intend to run it very lean and hot to burn off the soot. Then I will see how high the burn rate will go. I need 5 gallons of condensate vaporized each hour.

ADDENDUM: The adhesive I used didn't work out. So, I just put a coupling nut on the bolt that secures the burner head to the base of the roll cage, then used a second bolt to secure the PVC fitting along with gasket material. It's not going anywhere now.

 

[buenijo]

A TALE OF TWO BOILERS

The first video depicts a large monotube steam generator with all safety systems disabled and low (or NO) water flow. The burner was allowed to continue firing until the tube ruptured. The second video shows a firetube boiler that exploded due to operator error.

The lesson here is the monotube steam generator is VASTLY safer than traditional boilers. Indeed, I argue a monotube steam generator with all pressure contained within small diameter tubing does NOT present a material explosion hazard. The tube can rupture, but only at a localized point. The pressurized contents of the tube must move through the long tube before escaping. So there is no explosion. Furthermore, the mass of saturated water in a monotube steam generator is VASTLY lower than a traditional firetube boiler of the same power rating. So the quantity of energy contained in a monotube steam generator is a small fraction of the energy contained in a firetube boiler of the same power rating.

A firetube boiler heats a large mass of water contained within a large pressure vessel. During a catastrophic failure, the pressure vessel shell rips apart to expose the large mass of saturated water to atmospheric pressure. Much of the water flashes to steam almost instantly. BOOM!

Personally, I would NEVER build a steam engine system that uses a large pressure vessel to heat water for steam generation. This is why I designed my engine to operate at a constant output. A large mass of saturated water is very useful for meeting transients in engine output. This is very difficult for a monotube steam generator. So I just sidestepped the problem. The alternative I will try is using the engine-driven alternator to maintain charge on a small battery system, then either operate the engine ON/OFF or HIGH/LOW to meet variable energy demands.

 

[buenijo]

I fired the steam generator today. The burner was not working well at all. I shut down after a short period and ordered a second blower fan. Previously, when I had the blower on the burner head to force primary air, the burner worked well at LOW outputs. Increasing the output would cause soot to be produced. I could literally see some black soot blowing out of the exhaust port. I also verified soot accumulation on the steam generator tubing. With the blower moved to the underside to provide only secondary air (bypassing the burner head), what I witnessed was a white haze in the exhaust and strong smell of propane. What I believe is happening is the increased pressure in the combustion chamber due to the new blower position is preventing the burner venturi from providing sufficient primary air. So, propane gas is getting forced UP into the combustion chamber from the burner head with the vast majority not mixing with primary air. So, much of the propane gas is just moving through the unit, getting blown out the exhaust ports, and condensing in air to cause the white haze. Makes sense. So, I need to provide a blower for BOTH primary air (on the burner head as before) AND secondary air (on the underside bypassing the burner head). This seems reasonable. Good news is I don't have to do additional fabrication. I just need to keep the blower on the underside, then add a second blower to the burner head as before (I already have the attachment).

(02-02-25) The fan arrived this evening. I'll fire it up when I get a chance - hopefully this weekend. I will also enlarge the exhaust port as removing the steam generator and re-installing is not difficult.

(02-04-25) Nothing material to share. Just thoughts. So far, I can see a final prototype as not terribly difficult to assemble for small production runs. Yeah, of course this assumes I can get the damned thing to even work! So far so good. But the deck is stacked against me. I have no illusions. I'll just keep plugging away slowly and see what unfolds. That noted, I'm liking the basic design right now.

(02/07/25) Not gonna fire the unit today as it's raining. I removed the steam generator yesterday and added an additional exhaust port. I also placed some ceramic fiber on the base of the combustion chamber, then reinforced the seal for the secondary blower (the one on the underside). Everything appears to be ready for the next test.

(02/08/25) Gonna wait till next week for next test. No time this weekend.

(02/12/25) Raining for the next couple days. Hopefully, it clears up by this weekend. I have both blower fans installed and ready to go. Air flow is obviously boosted significantly! We'll see what happens. If I get a clean burn, then I'll try to get it REALLY hot and see if I can burn of the soot.

 

[buenijo]

I came upon a video of someone testing the same alternator that I selected for my steam engine project. This is the 48v model. There is another unit with different windings suitable for a 24v battery. Based on comments, the amp rating is likely generous. He suggests no more than 1500 watts continuous and likely 1000 watts to ensure no overheating. Not terribly surprised. China is very bad about inflating specs. Comment from the person who made the video: "yes they are made in China. And like most things from China they are over rated somewhat. I would expect to get 1500w max from this turbine and it will get HOT. At 2kw it might catch fire. 1KW would be a conservative output and will last longer."


NOTE: I have other alternator options. But I'm going with this one because it's about half the cost of the alternatives. The model I'm looking at uses 10 awg wire with 125C insulation. The rated current of this wire is listed as 70 amps. This implies 1700 watts is possible at 24v. However, the seller claims 2000 watts. Yeah, definitely inflated the specs there. Maybe the unit could handle 1500 watts (in part because a 24v battery is normally at 26v or higher). This would imply 58 amps at 26v. With good cooling it would probably pull it off. That noted, I wouldn't press it. I can always limit the output to 1 KWe which would take the current under 40 amps. It would also be a lot more efficient at the lower amperage.

 

[buenijo]

My electrical experience is very limited. I poked around many different sources to find a clear answer as to how the DC voltage changes when three phase AC is rectified to DC. I got many answers. The previous video and the video attached here provided a clear answer. The specs for the alternator I want indicates "24 VAC at 2000 RPM". But "VAC" is vague. Do they mean Root Mean Square (RMS), peak, line to line, line to neutral, etc.? Turns out they must have meant RMS line to line. Well, the DC voltage is about 1.8X the VAC in this case. NOTE: Of course, this is just an empirical estimate.

Example 1) The previous video show 53vdc at 1520 rpm. The alternator specs for that unit show "48 VAC at 2500 RPM". So, at 2500 RPM, the DC voltage will be about 1.8(48) = 86.4v. Therefore, 53v will correspond to about (53/86.4)(2500) = 1530 RPM. Actual speed was 1520 RPM. That checks out.

Example 2) The attached video shows about 14v RMS VAC. So, the expected DC voltage from the rectifier at the same RPM should be about 1.8(14) = 25.2. It checks out as well.


So, if the alternator provides 24v VAC at 2000 RPM, then the DC voltage would be about 1.8(24) = 43.2v. The RPM for 28v (which is about right for the charging conditions I want) would be about (28/43.2)(2000) = 1300 RPM. If I were to use the alternator ME1603 shown in the previous video, then charging a 48v battery at 56vdc would require about 1620 RPM. So, it seems my engine would operate at 1300 RPM for the 24v version, and about 1600 RPM for the 48v version. Again, assuming the damned thing ever works at all.

ALTERNATOR DISCUSSION: Just sharing a recent experience. I'm engaged in a discussion with someone who sells alternators designed primarily for wind turbines. He sells what are essentially automotive alternators with the rotors replaced with neodymium magnets. These have three phase windings, and the rectifier is either internal or external depending on the model. I'm sharing the details of the discussion because the details might help someone to understand more about how these alternators work. Anyway, the specs available on these alternators are limited. So, I contacted the seller asking about the stator resistance and the DC voltage output as a function of RPM. However, rather than provide the specs, he replied with questions about my application. This was frustrating because I don't need anyone to do the engineering. I just need the specs. The seller argued that knowing the voltage as a function of RPM is not useful because the actual RPM while battery charging will be A LOT higher than the cut in voltage.
MY REPLY: This is precisely why I want to know the stator resistance! The effect he describes is a consequence of HIGH stator resistance which is typical with automotive alternators. These alternators show LOW efficiency typically 50-60%. The primary loss is in the stator windings (especially at high amperage because the loss is equal to the square of the amperage X the resistance in Ohms - I^2R). Other losses include friction, windage, and rectification losses (which are higher at lower voltage). But the seller argues on his website that his alternators are highly efficient because using a permanent magnet rotor prevents having to use some of the electricity to energize the rotor, and there are no brushes which introduce a loss. Yeah, this avoids some losses. But the fact remains that the lion's share of the losses are from the stator windings which (I suspect) are unchanged in his units (and why I asked). So, naturally I'm a little skeptical of his "high efficiency" claims. MORE IMPORTANTLY, the effect that he describes where his alternators have to rotate at significantly higher RPM after cut in speed for charging is a consequence of VOLTAGE DROP ACROSS THE STATOR AS AMPERAGE INCREASES (i.e. low efficiency). Automotive alternators are NOT designed to be highly efficient. They are designed to deliver high amperage in a compact and low-cost package. Automotive alternators with higher efficiency are available. But notice they are LARGER because getting higher efficiency requires reducing the resistance of the stator windings (this means MORE copper!).

 

[buenijo]

Stopped raining today. Perfect timing. I fired the steam generator. The burner seemed to work well. The heat output was OBVIOUSLY much higher as it steamed up in less than 5 minutes! No propane odor. No evidence of soot in the exhaust.

I varied the burner output, and I operated at varying steam temperature and pressure for about a half hour. I took the steam line temperature to 1000F (pressure was 800 psig) for a couple minutes just to see how it behaved. No problems. After about a half hour of variable output, I set the burner to a relatively LOW OUTPUT (to provide EXCESS AIR to try and burn the soot off the tubing). I set the pressure to 500-600 psig, set the steam line temperature to 750-800F, and I held the conditions for exactly one hour. During this period, the system was VERY stable. I merely observed and made no interventions, and yet the pressure and temperature held within the ranges provided. Well, not exactly. I'm using a PWM controller for the feed pump motor that indicates 0% to 100% output in 1% increments. I operated the feed pump motor at 27% and 28%. At 27%, the temperature stabilized to about 800F. At 28%, the temperature stabilized to about 730F. So, I would just move the controller back and forth between the two settings every few minutes. Fascinating how the temperature change was so predictable even though the change in the feed pump RPM was so slight that it wasn't even discernable. I measured the condensate flow rate during this period as well. Based on this measurement, I estimate the burner heat rate was approximately 29,000 BTU/hour(*). Again, I purposefully set the burner at a relatively LOW rate to provide excess air. So, I am confident I can take the output much higher(**)

This was a good test. It seems to have worked PERFECTLY.

(*)The heat that transferred to the feed water to make steam amounted to 23,400 BTU/hour. I assumed 80% efficiency when estimating the burner output. This is a conservative estimate for a steam generator of this design.
(**)The alternator I intend to use should probably be limited to 1 KWe to prevent overheating and to optimize efficiency. So, assuming the steam engine works as designed, then I should not have to exceed about 35,000 BTU/hour on the burner. Next time I fire it, I will see how high I can go while still keeping a clean burn.

 

[buenijo]

I inspected the steam generator tubing coil. MOST of the soot was burned off! I might be able to get the rest burned off later. However, my steam generator design allows for removing the lid and literally pulling out the combustion chamber to expose the coil. So, I'm just going to clean the coil with a solvent. Most of the soot was burned off near the top because the temperatures were a lot higher there. I could probably clean it up by running at a low output with LOTS of excess air and NO water in the tubing. But the ability to pull out the combustion chamber and expose the coil for cleaning is a very useful feature!

ADDENDUM: I pulled the combustion chamber and cleaned the coil with carburetor spray cleaner. Easy. Although, I can see how a final design might be serviced on a regular basis by firing the furnace at a LOW rate with LOTS of excess air with no water in the tubing coil. Keeping the furnace temperature at 1000F with plenty of excess air should clean up things nicely.

 

[buenijo]

Planning an extended test run this weekend if time permits. I intend to find the highest clean-burning output for the burner, then operate the steam generator at about 300C and 500 psig. This temperature and pressure are roughly what I have planned for the first expander design. Gonna get it dialed in and just let it run.

 

[buenijo]

I just completed a two-hour run of my steam generator. I maintained steam temperature at approximately 600F (315C) and pressure roughly 500 psig (these are the steam conditions I plan for the first expander test). I measured the condensate mass flow rate to determine the feedwater absorbed almost exactly 40,000 BTU/hr. I then measured the steam generator furnace exhaust temperature at 250F. The peak temperature on the steam generator casing near the top was approximately 300F. The top center of the casing at the removable plug was approximately 350F. I'm just ballparking, but these numbers suggest on the order of 85% efficiency (lhv basis). Assuming this is reality, then a recuperator + additional thermal insulation I have planned for a later prototype could take the efficiency to 90%+. The thermal loss at the steam generator/boiler is the first loss in a steam power plant - so optimizing efficiency there goes a long way to increasing overall efficiency.

I will soon advance to the next step in the process which is to purchase and mount the alternator on the crankcase. I also intend to mount the oil pump even though I will not be using it for the first expander design. That noted, I intend to add some insulation to the steam generator casing and continue testing.

ADDENDUM: My goal was to get 40,000 BTU/hr into the feed water. So, I'm happy with the results. In theory at least, this will allow the expander to develop the maximum power rating of 1500 watts electrical from the alternator. This corresponds to about 2.5 shaft hp. That noted, I will likely limit the output to 1 KWe to protect the alternator. In summary, the steam generator has more than enough power.

 

[buenijo]

Someone in another forum made me aware of a couple of very small operations that have been testing Tesla turbines with good results. One claim suggests their turbine has significant torque at surprisingly low rpm such that 1-2 KWe can be generated at well under 3000 rpm or so. That would be very impressive. But I have to check it out to see if it's legit. I'm very skeptical.

(02/24/25) No reply as of yet. I don't expect anything useful. I do not expect a Tesla turbine at low cost to handle sufficiently high temperature steam required for useful efficiency. Furthermore, extensive testing of these turbines in the past showed at best 40% of theoretical max efficiency. Whereas, the single-acting bump valve uniflow can show 70-80%. I can build a uniflow piston engine with less effort than a good Tesla turbine. But I do like the simplicity.

(03/01/25) Still no replies. I left emails and phone messages. I also called several times. Nothing.

 

[buenijo]

I have to purchase the alternator soon. Once it arrives, I will have to install an adapter plate on the crankcase cover via standoffs. The oil pump will mount on the plate and be coupled to the camshaft extension. The alternator will have a separate set of standoffs attached to it to bolt to the adapter plate (these are necessary to make room for the oil pump). A rigid coupling will be installed on the crankshaft, and the alternator shaft will slip into the coupling. It will then bolt on the plate via the standoffs.

After the alternator is mounted, then I have to disassemble the condenser and replace the needle valve installed in the pot still with a heating element that has the right resistance. The alternator will wire up to the heating element via a rectifier which will mount on the aluminum crankcase (a good heat sink). Finally, I need to remove a check valve on the feed pump discharge line to the steam generator and replace it with a cross fitting that will include a pressure relief valve and a pressure gauge.

After all this is done, then I will start assembling the steam expander. No timeline. God knows what nonsense could pop up again to delay things!

 

[buenijo]

(02/28/25) Tested the steam generator again today. No problems. It's working well. I set the burner a bit lower. Feed water absorbed 32,000 BTU over an hour.

I will not likely have the expander in place until later this year. I will be purchasing the alternator soon and I will install it over the next few months along with the oil pump.

(03/01/25) Purchased the alternator. Should be here within the next two weeks. Gotta do a minor mod on the burner just to reinforce things. Nothing serious. I was going to augment insulation on the steam generator. But I decided against it. This can wait for a final prototype. The unit is surprisingly efficient as is. I inspected the tubing coil. Still clean.

 

[buenijo]

I modified the burner slightly. Nothing fundamentally different. Just a more robust assembly. Seems to work well. I fired it up (just the burner) at full power at night (so I could see the flame well). Nothing but blue! Very hot!

The alternator arrived! It's pretty beefy at 16 pounds and 8" diameter. I also dragged out the oil pump as it needs to mount to the adapter plate and couple to the small shaft extension I threaded onto the end of the camshaft. It's a tight fit. I'm not in a rush to mount this thing. I don't think it will be terribly complicated. But I am taking my time to get it right. So far I'm well on track to have a chance of actually testing an engine later this year.

 

[buenijo]

I am corresponding with someone who is testing a very small aluminum Tesla turbine available for purchase on Ebay. Very low cost unit well under $150. He contacted me hoping I would do the testing. But I declined as I have enough on plate with my current project. Besides, I honestly do not expect the unit to perform well. That noted, I offered to him advice on how to go about testing. I am highly skeptical about the performance of these small low cost turbines. But if a small turbine were to demonstrate high performance on low pressure saturated steam, then I would be interested to power one off the steam exhausted from my engine (placed between the piston engine and the condenser in a compounded configuration). As it stands now, the configuration I have in mind is to operate the condenser at the lowest practical pressure that still supports hydronic heating. I'm thinking about 5 psia which has a saturation temperature of 162F. This will comfortably heat a hydronic loop to around 130F.

 

[buenijo]

The system I designed calls for charging a 24v battery with the PMA via a rectifier. However, it is possible to place a charge controller with MPPT between the rectifier and the battery. This would allow the PMA to operate at a lower RPM. This will likely have some benefits. I prefer to eliminate fancy electronics. But it's worth consideration. Besides, these charge controllers are low cost these days. This approach would also allow for charging batteries at different voltage (i.e. not just 24v) with the same PMA.

Also, there is a company in the UK selling a fairly large Tesla turbine with an integrated small three phase PMA they claim is rated for 1400 watts. They claim the turbine can handle 150 psig steam. They also claim the alternator puts out 1 volt per 500 rpm with optimal rpm 10,000 - 15,000. This is good. But again, NO TESTING DATA is available! I might opt for this unit if rigorous data on steam was provided. What a shame. That noted, I did contact them and made an inquiry.

NOTE: All test data I've been able to uncover on these Tesla turbines suggest 20-40% of Rankine efficiency. So, in the absence of test data that suggests otherwise, then I must decline. My steam engine design can show 70% - and use high temperature steam at high pressure for higher theoretical efficiency. No, it's NOT all about efficiency. But the efficiency of small steam engine systems are generally FAR too low to be useful. My system does not have to be highly efficient. It just has to be EFFICIENT ENOUGH to be useful. I want 10% conversion of fuel lower heating value to DC electricity. That's not "high", but it IS useful. The 3-4% one can expect from a traditional small steam engine set up is NOT so useful.

For reference, my system at 10% overall efficiency would consume right at 5 pounds of air dry wood (15-20% moisture content by mass) for every KWh of electricity that charges the battery. The absolute highest efficiency my design could provide (within reason) is 15%. So, anywhere from 3.3 - 5 lbs of wood is the goal. A typical small traditional steam engine set up would require 15 pounds or more.

 

[buenijo]

I got a reply from the manufacturer/inventor of the Tesla turbine I mentioned in the previous post. He says the turbine was never tested on steam. Well, actually, he said there was a brief test, but they were not able to complete testing. So, that's a NO. Typical. My friends, you have to be careful of BULLSHIT because it's EVERYWHERE! I'm not saying the Tesla turbine does not have merit. I just saying don't trust. VERIFY!

(03/14/25) I offered to test the turbine if he sends me a unit. I doubt he will agree. We'll see.

 

[buenijo]

Progress on the project will be slow or stop altogether over the next few months. The wife and I are moving to a new house. Good news is the new place has a much larger garage with a lot of counter area. I may be able to get a few decent tools which will be handy for the expander builds to come.