Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length.

[1967Brutus]

Continueing the "pimping"...


My feedwater preheater only returns about 25~30% of the total steam at best (and I think closer to 25 than to 30) and I want to increase that number.

For that I need of course a cooling water supply, and it should (as a theoretical minimum) be around 70 ml/min

I don't want to mess with pumps (they rob a tiny bit of power from the shaft) and gears/belts/whatnot to drive that pump (robbing typically even MORE power) so I am going to use a "double" water pick-up: A water inlet in the high pressure area behind the propeller, whose "delivery pressure" will be supported by a water return in the low pressure area in front of the propeller. With a bit of luck, this will also reverse the waterflow when running astern.

Now the easiest way for such a pick-up is to drill a hole in the boat and simply glue a piece of brass tubing in. But the most "streamlined" thing to do would be to integrate it in the rudder. The existing rudder is a no-go, but a new rudder is easily made.


https://www.modelbouwforum.nl/attachments/20240223_143237-1-jpg.572784/


While I'm at it, Borkum loses a lot of speed with the rudder hard-over, and does not turn very tight at rudder angles that do NOT rob speed. So I thought I would like to experiment with a Fishtail rudder while I'm at it. No particular reason, just to see what it does...

End of tomorrow I hope to be able to report on at least the water flow...

[SteamboatPhil]

Bon Courage 

[1967Brutus]

Merci beaucoupe!

[1967Brutus]

The curing of the resin to build the rudder took a bit longer than expected, but the rudder is taking shape.

It is a bit shorter than the original rudder, because I started the waist of the fishtail profile a bit too early and I had to cut a narrow ridge off the trailing edge to prevent the fishtail to become too shallow and too blunt. The entire effect of the fishtail is in that the water is sharply accelerated sideways very close to the trailing edge, so it is imperative that the waist is not too far from that trailing edge.

https://www.modelbouwforum.nl/attachments/20240223_235559-1-jpg.572889/


The profile is getting there, it still needs to be a bit thinner at the waist, and symetry is not yet what it is supposed to be, but it's midnight here, so I'm gonna call it a day.

https://www.modelbouwforum.nl/attachments/20240224_000134-1-jpg.572890/

Curious how this is going to work out. I have seen Becker rudders in model ships, but I can't remember having ever seen fishtail rudders, leave alone anyone that replaced a normal rudder with a fishtail and reported on the differences.
Fingers crossed...

[1967Brutus]

And since time does not take lunchbreaks either...

The rudder was made to fit in the propframe. here it still needs filler and paint.
https://www.modelbouwforum.nl/attachments/20240224_130517-1-jpg.573014/

On the topside, the water tubing connected and checked if it clears the hatch and allows for the rudder movement:
https://www.modelbouwforum.nl/attachments/20240224_130521-1-jpg.573015/

Next step, while the boat was "opened up" anyway, was the installation of the water exhaust. It is in front of the prop, in order to utilize the low pressure in front of the prop as assistance for the pick-up pressure.
https://www.modelbouwforum.nl/attachments/20240224_142320-1-jpg.573016/

I cut the tube so that it will end as closely to the prop disk as I could get it, all to maximize effect.
https://www.modelbouwforum.nl/attachments/20240224_142415-1-jpg.573017/

Now the waiting is for the glue, resin, paint etc etc to cure...

Boring...

[1967Brutus]

Short report, tried to get video and such, but since the feedpump failed from the start, I could not get "data" (differences in runtime on the feedwater tank, etc etc).


But I managed to determine the following: it looks like the waterflow is just about right for what I hoped to achieve, but "not abundant".
The propwash is NOT sufficient to fill the system by itself. But if pre-filled, the system flows, and the reversal also works.


The fishtail rudder seems to indeed have better steering effect, and this even with a slightly reduced maximum rudder angle.

Pretty bummed out about that Regner feedwater pump.

[rhavrane]

Bonjour Brutus,I was also very disappointed with my Regner feed water pump. this is why I kindly suggested you to use a Microcosm one with an enhanced engine:
https://www.youtube.com/watch?v=FNy_x-RR9mE

[1967Brutus]

Bonjour Brutus,I was also very disappointed with my Regner feed water pump. this is why I kindly suggested you to use a Microcosm one with an enhanced engine:
https://www.youtube.com/watch?v=FNy_x-RR9mE

I was not too impressed with the Microcosm item either. I had one, but I cannibalized it for one of the connections, rendering the rest of the pump useless.

Not sure what I am going to do. The Regner pump appears to be servicable, but I need a magnifier glass and a good light source since the internal parts are very tiny. It seemed there was some dirt in the pump jamming the valves.

Tomorrow, after a good nights sleep another try.

But I am happy that the cooling system appears to work...

[1967Brutus]

Important update:

The pumping element of the Regner pump turns out to be VERY servicable:


The thing is SO tiny, that the suction valve acts a bit like a filter: Fluff in the water catches in the suction valve, messing up its function.

I could not get a better pic, but for reference, the valve disk itself is approx 3,5~4 mm diameter, the valvestem maybe 1 mm dia, and 2 mm long.
There is a tiny plastic (unknown which, most probably teflon or such) seal on the disk.

https://www.modelbouwforum.nl/attachments/20240225_112125-1-jpg.573095/

Recommendation is to use a strong light, and use a piece of A4 printing paper as work surface.
Mark the parts for proper re-assembly (I made a diagonal scratch along the outside before disassembly, marking how things were supposed to go back together again).
The bolts holding the element together are 3 mm hex head, and the only tool I could safely use was a 4" Bahco adjustable spanner.


The plunger gland seal can be made by taking about 3,5 cm of PTFE tape twisted to a tiny cord, wrapped around the plnger and into the gland cap nut.
A bit of PTFE grease and carefully tightening it with running pump does the trick.

The quality and servicability surprised me, actually.

Test run (hopefully with decent video) in approx 30 minutes from now (posting will take longer of course, the test run should be at least 30 to 50 minutes in duration).

[1967Brutus]

Going to do the video at a later time (need some editting/cutting) but first conclusions:

I seem to be having at best about 60 ml/min, which is marginally enough, IF the coil is inside the condenser instead of outside.

One thing I completely overlooked, is that water has air dissolved in it. When water gets hot, this air gets released, leading to bubbles in the warm water return line. This reduces waterflow and at half speed, the waterflow simply stops. It does return when going to full ahead again, but this was something I had not thought about.
A pressure cone might bring some improvement,

Also: the bathtub is NOT 100% comparable with a free flowing hull, the waterflow significantly changes with the hull free moving through the water.

Improvement can be had from placing the condenser "as low as possible" in the boat"


Does it have a measurable effect?
Most definitely, I did the test running full speed all the time, and very roughly calculated, the feedwatersupply lasted about 20% longer (timer was set at 25 minutes, low level was however reached at around 30 minutes).

We're not done yet...

[1967Brutus]

Apologies, the promised video failed.

But meanwhile I did some calculations, and damn! I love it when theoretical physics (Bernouilli) turns out to be true: The smaller I made the inlet cone, the better the cooling system performed.
The straight open pipe (3 mm inner diameter, all subsequent piping also 3 mm inner diam all the way to the end) of the pick-up worked.

 A small piece of 2 mm inner diam silicone tubing slipped over the pick-up increased flow,


 and an even smaller cone (approx equivalent to 1,5 mm diameter) that I made out of heated and stretched Sullivan bowden cable outer tube worked even better.

Based on estimated prop wash velocity, I calculated that approx 7,5 cm of water column above waterline should theoretically be possible, and with the smallest inlet cone I actually managed to achieve that.

[KBIO]

Surprising and interesting ! How to explain that ??
Thanks to share you tests for the benefit of all ! 

[1967Brutus]

Surprising and interesting ! How to explain that ??
Thanks to share you tests for the benefit of all ! 

Bernouilli's law... That is a very tricky part of physics and leads to very counterintuitive results.

If a flowing liquid is forced to change velocity, pressure changes An increase in velocity lowers pressure, a decrease in velocity increases pressure.

The velocity in the propwash is about 1,25 m/sec, the velocity in the ctubing is in the order of magnitude of 0,2 m sec.

Now if the pipe is straight, and water is not compressable, the velocity in the opening of the pick-up tube is identical to the velocity in the rest of the tubing, 0,2 m/sec. This means that the water has to decellerate very suddenly in front of the tube, and thus the high pressure area is (some distance) in front of the tube. A bit like a bow wave.
In front of the tube there is nothing that forces the water to go IN the tube, and nothing to contain the pressure, so most of the water flows off to the sides, a bit like a bow wave in front of a blunt bow: here the high pressure area also occurs some distance before the water actually hits the bow.


Now if we make a pick-up opening that for a velocity of 1,25 m/sec passes the same amount of water, as would the tubing of 3 mm internal diameter at 0,2 m/sec, we would end up with an inlet opening slightly smaller than 1,5 mm diameter. Now the water passing the cooling system at 0.2 m/sec can ENTER the system at 1,25 m/sec, which means that the pressure change will occur IN the tubing, AKA "enclosed, and none of the water and pressure can escape. now the conversion from velocity to pressure occurs IN the tube. If that is a stepwise increase of diameter, turbulence will occur. Not ideal.
But if we make that increase gradual (AKA diverging cone) there won't be turbulence, velocity change is gradual and maximum pressure is generated.

It took me quite a bit of time looking at the formulas that represent Bernouilli's law before I actually understood what happens.

Intuitive, people would want to make the inlet funnel-shaped in order to "catch the water" but it does not work like that, and a funnel shaped inlet only makes things worse: The inlet diameter increases, which means the high pressure area allready starts at a greater distance in front of the inlet, and water has more incentive to escape sideways.

Decreasing the inlet diameter to where the velocity in the opening is equal to the surrounding water of the propwash, there IS NO velocity change in the inlet, therefore no high pressure area BEFORE the inlet, therefore no water and pressure excaping sideways.

As i said, very counterintuitive, but once you "see it", it is actually quite logical.

[1967Brutus]

And suddenly it dawned on me: The intake opening determines the POSSIBLE water amount per unit of time (propwash velocity times intake opening cross section results in a volume per unit of time).
The rest of the cooling water tubing determines  whether that volume can actually pass given the resulting backpressure.
Highest delivery pressure achievable is with the smallest nozzle opening that will allow the required flow.

I know that I need to condense about 7 grammes of steam per minute, and for that I need an absolute minimum of 60 ml/min, but to allow for some headroom, let's make that 75 ml/min, not only for headroom but also because of ease of calculation:
With a propwash velocity of 1,25 m/sec, that velocity can be expressed as 75 metres per minute...
75 ml/min (ml is cubic cm) divided by 7500 cm/min results in a cross section of the inlet of 0,01 cm2
0,01 cm cross section means an inlet diameter of 1,13 mm.
I don't have that drill...
1,1 mm diameter has a cross section of 0,0095 cm squared, resulting in a maximum possible waterflow of 71,3 ml min.

Should be sufficient. I should be able to make something like that.

EDIT:
Just as a check, an intake of 1,2 mm should flow 84 ml/min but probably the current tubing won't pass that (just a hunch).
But the easiest way to make something like this, would be to turn a conical dowel on the lathe (narrowing down to zero), and use that as a mold to make the internal cone from either Silicone or resin, then cut the cone to length to get the desired inlet opening. That would mean I can make several, and experiment.

Going to be a fun and educational project...

[KNO3]

This is very interesting. And leads me to believe there must be an optimum opening diameter for a given water speed and tubing size.
Since you have both ends of the cooling tube under water so there is no visual of the water flowing, how can you check the system is actually working?

[1967Brutus]

This is very interesting. And leads me to believe there must be an optimum opening diameter for a given water speed and tubing size.
Since you have both ends of the cooling tube under water so there is no visual of the water flowing, how can you check the system is actually working?

Theoretically, optimal opening size is where the velocity of the propwash passes the water the system will accept. In reality  bit larger due to the viscosity of water.

For now I can check flow visually by means of airbubbles, and by the temperature difference between inlet and outlet.
When the condenser has its final form, it also should be visible at distance by the presence or absence of a steam plume.

[SteamboatPhil]

Chapeau 

[1967Brutus]

I see that I have not kept up with the updates.

I made a "mold" of sorts to fabricate a decent pitot-nozzle.
Pretty much a simple piece of roundbar 4 mm turned to a cone.




Coated the mold with wax, wrapped a bit of Epoxy soaked tackling yarn, and let it cure.


This allowed me to fabricate several identical nozzles, which I could drill out to size for experimenting.

1.2 mm was what I ended up with.


Secured with a bit of silicone tubing, this is the endresult:

looks hideous, but all that matters is the shape and dimensions on the inside. Those determine effectivity.

Now I still had the coil on the outside of the condenser, and that is a miserable design: Heated water releases its dissolved gasses, these gasses remain trapped in the high points of the windings, and each airpocket represents a bit of height difference, that the pitot tube has to overcome. As long as I held full throttle, the undisturbed system would keep going all by itself for about 20 minutes, then the collected air would stop the waterflow.
But even with a "fresh" system (no air build-up yet) below 3/4 power the waterflow would stall, and was very reluctant to restart on full power.
With a bit of tapping against the lines I could dislodge that air, and keep the system going, but that is not how it should be.
When reversing the engine, sometimes the waterflow would reverse, sometimes not. Also not very desirable.

With this coil (1 metre of 4 mm copper tube wrapped around the condenser shell) I managed to stretch endurance from about 25 to about 30~31 minutes of continuous full speed.

So I decided, just as a proof of concept to see what a horizontal "coil" inside the condenser shell would do. Since inside most space was taken up by WAAAAAYYYY too much preheater coil, I could only fit about 20 cm of condenser tube, just once forward and back.

the result was remarkable: Despite only 20 cm of steam-exposed cooling tube, I gained another 4 minutes of full speed endurance. Absolutely not bad.
What was even better: The waterflow maintained itself, water pressure at full speed was strong enough to dislodge any trapped air. Waterflow easily and reliably reversed itself when the machine was reversed, and at least in forward direction, waterflow kept going reliably at 3/4, most of the time even down to 1/2 power, and restored itself every time above 3/4 power.



Next step will be to shorten and rearrange the feedwater preheater coil, in order to create more space for one extra loop (additional 20 cm) of condensing coil.
And I think I will leave it at that... 40 minutes of full power is enough.

[1967Brutus]

I have hit a snag in that I seem not to be able to keep waterflow over the condenser with the pitot-tubes. It is somehow a combination of things, because the pitot-tubes can do the job, but as soon as steam is introduced in the condenser, things grind to a halt.

I still have slight hope that the pitot-tube will work better in free sailing conditions, but not optimistic.

So I took the liberty to order a miniature electric pump designed for continuous duty.

Meanwhile I made a "lab set-up" with external waterfeed, and the results are pretty much everything I hoped for and then some...

See for yourself:

https://www.youtube.com/watch?v=hnOM9LrzXeM


As I am writing this, the engine has been running for about 20 minutes at the highest speed the fuel supply allows (10 minutes "balls out", the gastank cools down and I have to rely on the tank heating, which can sustain about 90% of max output).

In this 20 minutes, with only feedwater preheating and the condensate from that, the feedwatertank level would drop at least 30~35 mm.
Right now I am looking at a level drop of 5, maybe 6 mm in the same time...

The condensate return filter has no issues keeping up and the condensate is of absolute brilliant clear quality.

The last 5 runs I would, including raising steam, burn about 35 grammes of fuel for an approx 30 minute run, and then the feedwater level would force me to stop, with about 10~15 grammes of fuel left. Meaning there should be somewhere around 45 minutes of "full power" (meaning "as much as the burner capacity allows") availlable.
Now I am planning to run out the fuel, to see how much water I still have left.

EDIT: it took between 45 and 50 minutes (did not note the exact time of start, sorry) to run out the gastank. All this time I kept the steam regulator such that the burner just kept about 100% fuel supply, otherwise the feedpump does not keep in sync with water consumption.

I have a watertank with 50 mm water (appr 360 ml) in it at start. Normally in 25~30 minutes, about, waterlevel has dropped to 10 mm (70 ml) remaining, and I am forced to stop or the feedpump might run dry.
Now I went all the way until empty fuel tank (45~50 minutes) and remaining feedwater level was 37 mm (appr 260 ml).

To bring this in practical terms, IF I get a reliable cooling water supply going, I can reduce the size of the feedwatertank by about half and still be safe for a complete fuel charge. The size and weight reduction of such tank would MORE than compensate for the installation of a cooling water pump.

[rhavrane]

Bonjour,Quite interesting exercize indeed !
I am a little bit suprized by the design of the condenser itself.
To my opinion, as you can't install an efficient cooling system for the condenser, it could consist of pipes installed in the water under the hull.
Furthermore, it would simplify your current piping.An example on one of my boats, at the beginning of the video : https://youtu.be/g3HkvoJdweo   

[1967Brutus]

I ran a 2nd test today, this time a bit more aimed at practical use.

I may have mentioned before, that I have "marked" 3 speedsettings in the TX, for ease of manouvering, which are "dead slow" (lowest the engine will keep running reliably), half, and "cruise".

Cruise is the speed I can run from beginning to end with the burner remaining in the controllable range, and is about 75% percent of max power.


As long as I use lower speeds every now and then, occasional bursts of full power are no issue on the pond, so running continuous cruise settings is fairly representative of normal pond-use.


This test, for measuring purposes, I fitted a throttle valve on the watersupply, in order to be able to measure the required waterflow.


Not gooing to bore you guys with a full description of the test, but testing showed that for full power, I need 100~110 ml/min of cooling water to condense all steam (steam exhaust remaining dry) and at cruise speed, this volume flow was 75 ml/min, give or take. Cool! Means that I won't need a big pump.




I started out with 60 grammes of fuel, and after raising steam, I basically set the engine at "cruise" and just waited to see what would happen.
Engine run lasted 58 minutes, with a waterconsumption from the hotwell equalling 80 ml.
55 ml was accounted for in the oil separator and the lubricator, so I am apparently losing only 25 ml of water in 1 hour of runtime.

About needing a pump:
I ordered this one: https://www.fruugo.nl/dc12v-5w-miniatuur-food-grade-borstelloze-dc-dompelpomp/p-122360316-257205194?language=nl&ac=bing&msclkid=6c5041651ee3169be8fc274272234dc3&utm_source=bing&utm_medium=cpc&utm_campaign=All_NL&utm_term=4575136606006185&utm_content=NLSmallest I could find with best efficiency. Going to "trim it down" as far as possible, so with a bit of luck I am looking at less than 2 Watt while still habing ample flow, which would be acceptable.

[1967Brutus]

Bonjour,Quite interesting exercize indeed !
I am a little bit suprized by the design of the condenser itself.
To my opinion, as you can't install an efficient cooling system for the condenser, it could consist of pipes installed in the water under the hull.
Furthermore, it would simplify your current piping.An example on one of my boats, at the beginning of the video : https://youtu.be/g3HkvoJdweo

I cannot use that system... It won't cooperate properly with my oil separator (a condensing pipe below the waterline would create higher backpressure, and my oil separator needs to be more or less "atmospheric" or it will blow the oil everywhere).

Also: I started with a feedwater preheater initially, which never was thought of as a full condenser. It's location was above the hotwell for easy condensate return.
Initially I miscalculated the preheater, so it was way too large, large enough to be a full condenser AND a preheater at the same time.

The piping is not very complicated, but I have to admit that the water pick-up tubes did not work the way I had envisioned it.

I am not too worried about that, because I found a very affordable, very tiny centrifugal waterpump with low power requirement, designed for continuous use (brushless motor technology similar to what is used in computer cooling fans).

[1967Brutus]

Pump delivery date is now stated as "approx 10 May"...

So for the time being, I'm going through my old discarded fieldbox fuel pumps. There should be one or two that I could get back to life, if even temporary...

[1967Brutus]

Found an old Volvo windshieldwiper pump, that I got from a salvage yard at least 20 years, and which has been used for glowfuel for at least 10 of them.
I tested how low it would go, and it managed to run reliable at 2S NiMh with a very accptable current draw (I can't measure it any more accurate than "maybe 600 mA") and a waterflow of about 300 ml/min, which is 3 times what I need as minimum. Initially it displayed a dissapointing amount of leakage on the shaft seal, but I simply solved that by reversing the polarity AND the in and oultlet: Now the sucion pressure is active on the shaft seal, and air will leak in instead of water leaking out. Problem solved, boat dry again.

Turned an old receiverpack of 1900 mAh into 2S2P (2,4V, 3800 mAh), and this combo has to do the temporary job of cooling water set-up. It provides me with more cooling duration than I need for now.

Tested this set-up on open water, functioned flawless for 40 minutes, and then a big rainshower spoiled all the fun.
45 grammes of gas, less than 100 ml of feedwater consumption, boat functions flawless, it is a bit of a weird visual, a steamboat with ZERO visible plumes above it, running and running and running and a feedwatertank that barely drops in level.

Fantastic!

Whatever my next steamboat will be, it WILL have a full-recovery condenser set-up... Whether engine driven or electrically driven cooling pump remains to be seen, but full recovery, that is a given! It is worth it. Not because of cost, but because of ease of use (I don't need to bring a jerrycan of water if I go to the pond...)

Can't wait until the brushless pump from China will arrive, after which everything can be installed below deck , and the original visuals of the boat can be restored.

[1967Brutus]

Behold! A steampowered boat with NO steam plume above the funnel! The way it should be!

https://www.youtube.com/shorts/u0jqCW-Cn6M