Print

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

[steamboatmodel]

I studied PID loops back when I did building maintenance, never thought I would have to think about it for model boiler control. I wonder if my notes from 35 years back are still relative.
Regards,
Gerald.

[flashtwo]

Hi,

Benjamin, you obviously have a control background - Honeywell Avionics?

I understand now, it is what I knew as adaptive control where the control terms varied according to the load of the system.

The other I knew as matrix control, where a data matrix held all the data throughout the control range and updated it under varying plant conditions.

I've never really used derivative (anticipation) control, because it amplified noise on the signal and could cause erratic control. I much preferred feed forward, which is what I use between the feed flow control loop and the temperature control loop; when the feed is increased the temperature set point is raised in anticipation of the temperature dropping and vice-versa.

I was thinking of varying the temperature control loop control terms relative to the gas valve opening value; it would be a simple linear relationship rather than an additional PID loop.

Gerald,

Yes, exactly the same principles are involved. You measure the thing you are trying to control, compare it with a desired value and then adjust an actuator which controls the thing you are measuring. It doesn't matter whether its the H&V of a building or a missile control system.

Apologies to all other Mayhemers who haven't a clue what what we are talikng about.


Ian

[gondolier88]

Apologies to all other Mayhemers who haven't a clue what what we are talikng about.


Ian

Thanks!!!

Give me springs and diaphragms any day!

Just joking, I for one, and I'm very sure I'm not alone here, respect your take on what is a most versatile and dynamic part of model engineering/mosel boating, and you never know what else these control systems may one day manage...

Greg

[benjaml1]

Thanks!!!

Give me springs and diaphragms any day! Greg

Interestingly enough PID control systems were made/functioned using springs/bellows/restrictor & levers in the good ole pneumatic days...Well before micro processors & are still in use all around the world... 

[gondolier88]

Ahh, i'm stuck in the good ole days, I don't feel so bad now

Greg

[benjaml1]

I have been offered a position with the nuclear energy peeps, not based on my neuvo technical intellect but my archaic knowledge of so say redundant pneumatic control systems...  

Tried retirement & found that boring, 61 years old & going back to work...EH! I love a challenge    

May be able to buy more steam goodies now...  

[gondolier88]

I see, so our 21st century powerstations recognise that steam should be mechanical not electric, but us model boaters think better, Mayhem to a 'T'!

Greg

[benjaml1]

I see, so our 21st century powerstations recognise that steam should be mechanical not electric, but us model boaters think better, Mayhem to a 'T'!

Greg

  

[flashtwo]

Hi,

Wasn't it annoying when you found that tiny hole drilled in the flapper caused by the fine rust particles in the compressed air - it always used creep up unexpectantly.

Honestly, the old pneumatic systems were superb in a hostile environment of heat, humidity and dust and IF the compressed air was maintained to good quality they would last for years unattended, but the slightest bit of damp air and the compressed airlines would rust internally and you would be on the downhill path.

I started my career in a small power station that had no electronics at all, not even thermionic valves - it was purely hydraulic and pneumatic and, if you talked about using electronics, the old guard thought you were from another planet!

Saying that, despite all the computing power, the last link in the chain is still the hydraulic/pneumatic actuator though increasingly being replaced by electric actuators.

Can you imagine one of the old Taylor controllers on a model boat! No.

All the best

Ian.

[flashtwo]

Hi,

Just to let you know that "Vital Byte" will be displayed and demonstrated at the Guildford Model Engineering Society steam rally this weekend, 10th & 11th July, 2010.

This V.B. will have its new control system and other modifications in operation.

See http://www.gmes.org.uk/find.htm for details of the Society's location.

Hope to see a few Mayhemers there as usual.

Ian.

[flashtwo]

Hi flashsteamers,

New failure mode!!!!@~###

Yesterday and today at the Guildford Steam Rally, Vital Byte was put through its paces, but only for about 10-15 minutes.

Yesterday V.B. experienced a heavy clunking sound after about 10 minutes of near silent running. The sound seem to come from the prop shaft rather than the engine, but finally the power train seized and the boat was hauled out of the water.

I removed the cylinder covers and, although one piston needed tightening slightly, there was no apparent engine problem. I then found the prop shaft shear pin had fallen out after over a years service. I replaced then pin but found that the shaft seized at certain positions, but, if I lent on the stern, it would free up. I concluded that the very hot sun ( it was 31degC!) had caused the hull to warp during it's open air display.

Today I ensured that the shaft would turn whilst it was on diasplay and everything was ok. Then after several hours it was time for a pool demo and again after about 10 minutes of perfect running the shaft seized again!

What a puzzle.

Suspecting the pool water temperature, I put my max/min thermometer (its amazing what useful junk one carries) in the pool and recorded 27degC! The only thing I could think of was that I have some reiinforcing aluminium strip under the stern that was expanding and causing the hull to "hog", thus causing mis-alignment of the engine to prop shaft coupling. Normally, I run V.B. in the local lake where the temperature is quite low thus avoiding this problem.

Solution: open up the gap on the dog-tooth coupling to increase its tolerance.

There was another chap there with a tug powered by a D10 engine and his loc-tighted prop became loose on the prop shaft - whether this was to do with water temperature, I don't know.

Ian.

[flashtwo]

Hi,

Here is a photo showing the fully tested and working AE-35 controller now being used on Vital Byte. With it is last years somewhat larger prototype and a 4inch piece of 7/8th brass for scale.

The AE-35 has basically the same functionality as the prototype, but uses smaller PICs for the pumps and engine RPM processing. The display processing and control functions are each in their dedicated 16F877s PICs and the PICs talk to each other via a serial link that enables a lot more data sharing.

The AE-35 still has the capability of being re-programmed and has had it's software code tidied up and simplified in parts. Having a dedicated display PIC gives a more reliable response to the push buttons located on the operator's control console.

Ian.

[benjaml1]

Impressive.... 

[flashtwo]

Thanks Benjamin...

No sooner was the AE-35 unit in V.B. then it was out for another software mod.

With the old system I could have the desired steam temperature set by the steam pressure, via an equation, to give the steam saturation temperature appropriate to the prevailing pressure. It did, however, suffer from instability if the steam demand suddenly changed causing the pressure to change and consequently the desired steam temperature.

The latest mod links the desired steam temperature with the desired feed flow (set by the radio transmitter). As the desired feed flow increases, it is anticipated by the controller that the steam and pressure will increase with an appropriate increase in saturation temperature. If the heat input isn't increased accordingly, then wet steam conditions would be experienced.

The new mod addresses this problem by adding a proportion of the feed demand value as an offset on the manually set desired steam temperature. Since the feed demand is a constant, though manually adjustable, the system is stable in its operation.

Next project a computer controlled flash steam calliope? Then I could really start playing some tunes on it!

Ian.

[benjaml1]

It amazes me the level of technology that has found it's way into steam model building. I'm still in the stone age....

[steamboatmodel]

It looks impressive. I haven't used a 16F877 PIC yet, I am still working out the basics (in assembly and C ) with a PIC 2 starter Kit with the 16F690 pic. I also have a 16F627, 16F628 to play with once I get them down I will then advance to a 16F886 28pin pic and then the 16F887 44pin pic. So far I have done the "Hello World" bit and got the leds to flash without letting any magic smoke out, my next step is to do servo positioning and reading RC signals.
Regards,
Gerald.
PS It was much easier learning things forty years ago when I was in my twenties. Then I would write notes and remember things without having to look at the notes, now I not only have to write the notes, I have to remember where I put the notes and the right glasses to read them.

[flashtwo]

Hi,

Vital Byte produced her best performance yet at the Blackheath Club meeting on Sunday.

The meeting was really for the straight runners, but Steamboat Phil encouraged me to get V.B. out on the water before the competition started.

Before launching, V.B. had the honour of being inspected by Alan Rayman one of the authors of "Experimental Flash Steam" - I hope he approves of the modern technology!

Several of the members were proud owners of Alan's boilers and engines, some being 30 years old or more.

All the changes made to V.B. over the last six months came to fruition - a good quick start and then she was off with her new prop, new computer and new boiler coil.  I gradually pushed her up through the power range, by remotely increasing the feed, to a very good speed and fine hull waves. She did many full power change-overs from ahead to astern and back with the engine accelerating to 730RPM during the change (maximum recording is working ok now);  not bad for a 72lb displacement hull shaped like a box. The average RPM for the run was 480.

A couple of mid-pond shut-downs were made and she auto-restarted on both occasions. Three point turns - no problem.

A second successful run was made in the afternoon, though with one minor problem of the feed inlet plastic pipe coming adrift and causing her to ship water; I had inadvertently dislodged the pipe when installing the lead ballast. Despite the minor hic-cup, Steamboat Phil and his son took a video for one of Phil's talks on flash steam.

One remaining problem - I've now got to build a proper boat!

Ian

[SteamboatPhil]

A good day was had by all at Blackheath on Sunday, with Vital Byte steaming around most of the day.

Here is Vital Byte in action 

[flashtwo]

Thanks very much Phill for the photo - its always a bit difficult to operate the transmitter and take photos at the same time.

Its very interesting to see the hull wave, it seems to be one complete wavelength, which may indicate that the boat is travelling at hull speed.

I've got very limited knowledge in that area, so I would be very appreciative of any comments regarding hull speed waveform from Mayhemers with that experience.

Ian.

[SteamboatPhil]

Here you Ian a slightly better bow shot 

[flashtwo]

Thanks Phil, that's much better.

It looks like one trough behind the bow followed by one crest before the prop, i.e. one wave length.

For those not familiar with the boat, there is a large overhang above the water after the prop (see earlier posts at last year's Guildford Show), which gives the appearance of a much longer hull.

Now, is the hull speed indicated by one wave length or one half wave length?

Ian

[flashtwo]

Hi,

I’ve just fitted an oil/water separator and combined feed heater on Vital Byte and tested it with further improved results.

The separator is a vertical brass cylinder 1.75inch diameter by 8.5inch height. Within the cylinder, I've fitted a copper coil 5/32 by 100 inch which gives approx. 50sq.inch of heating area (steam exhaust side).

The exhaust steam /oil enters near the top and condenses on the exposed section of the coil. The drain is located at the bottom and exits above the water-line, the exit height determining the water level in the separation tank. The coil carrying the boiler feed water enters near the cylinder top, passes through the exhaust steam section then into the hot water on top of which lays the separated engine oil.

One problem with the previous exhaust steam collection arrangement was having to manually pump the water/oil out every 20 minutes of running time - with the new arrangement the boat will only be brought in to top up the displacement lubricator having dumped excess (oil free) water overboard.

The unit has been operated on the test pond with the drain water  free from oil and the feed water temperature rising to about 70degC before entering the economiser coil located in the stack.

The economiser outlet temperature is now about 147degC which indicates steam production since it is above the saturation temperature.

The efficiency has improved a lot since the gas valve is only having to open to 48% instead of 100% to achieve an average RPM of 612 over a 18 minute period.

I shall try and take some gas consumption readings on the lake tomorrow.

One thing I’ve got to understand now, is that the inlet temperature to the main boiler coil is 147degC, whereas the boiler outlet steam temperature is 131degC!


Here's a couple of photos.

Ian.

[flashtwo]

Hi Mayhemers,

Vital Byte has been performing very well over the past few months with the oil separator / feed heater preventing oil from poluting the lake.

Since there is now no condenstate tank to empty every 20 minutes, extended runs of 40 minutes without returning to shore are a regular feature.

With a view to building a “proper” boat, I’ve started construction of a vertical boiler to match the scale of an Edwardian Steam Launch that I have the plans of.

 The new vertical boiler is unusual (like everything else in this project!), in that the boiler coil extends up through the furnace and into the 2 inch diameter stack. In this way it will have 200sq.inches of copper-to-water surface area plus the 50sq.inches in the feed heater gives a total of 250 sq.inches.

The water will enter the coil at the top of the stack and steam will exit the coil at the connector located at the bottom of the furnace – how different is that!

To ease the removal of the boiler, connections have been kept to a minimum. The safety valve (not legally required) and the pressure and temperture tapping points have been removed to their own manifold separate from the boiler. The only connections left are the feed inlet, steam outlet, pilot gas and main burner gas – no sight glass or level measurement is required.

The first photo (DVD for scale) shows the skeleton of the boiler prior to attaching the coil and heat shield. The second photo shows the steam outlet to manifold arrangement.

The vertical boiler is in fact the same height of the current horizontal boiler.

After initial trials it is planed to encase the boiler in the traditional wood and brass banding.

Ian.

[steamboatmodel]

Hi Ian,
You say "The water will enter the coil at the top of the stack and steam will exit the coil at the connector located at the bottom of the furnace – how different is that!"
Would not the steam rise to the top of the coil?
Regards,
Gerald.

[flashtwo]

Hi Gerald,

That's a very good point and well worth the experiment.

One thing with the monutube or flash boiler is that the feed water is continuously pumped in under pressure, which has to be slightly higher than the steam pressure to maintain flow, so hopefully that should discourage the steam from rising to the top.

Another feature is the effective tube gradient.  As you can imagine the tube sprials down from the top over a vertical distance of 405mm. The tube length is in the order of  12 metres and so the effective gradient is about 1 in 30 and as you might have experienced with a steam or air lock in a long pipe they can be difficult to shift by natural circulation.

The reason for pumping water in at the top is to maintain the highest temperature difference between the combustion gases and the water/steam in order to maximise heat transfer.. The coldest part of the boiler is at the top, so that is where the cold feed enters, and the hotest part is near the burner where the steam exits. The water/ steam spirals down from the cold to the hot end.
 
Anyway that's the theory - we'll find out what happens in practice!

Cheers

Ian

[ooyah/2]

Hi Ian,
I would have no fear of air locks, your coil is wound in spirals with no sharp corners.
Your coil is similar in construction to my flash steamer with the water going to the cold end and my coil is lying horizontal but in yours with it being vertical the pressure created by the pump and the heat generated will turn any air bubbles into steam along with the air already in the coil.
Any condensate in your cylinders will be shoved out as soon as the valve is opened to the steam.

I am puzzled as to what the large dia tube going up into the funnel is, have you considered making your coil wider to fit into the shell case and then reducing up into the funnel, you could then use a ring burner or a large ceramic one, just a thought.
I look forward to hearing your results, the mechanical ones not the electronic ones !!!

George.

[gondolier88]

Hi Ian,

I have to say your idea re. the feed arrangement, though based on sound theory, isn't based on practice, as I'm sure your aware.

The reason the cold water is injected at the base is to make ful use of natural convection- the steam engineers bedfellow, our hobby wouldn't exist without it.

I would suggest you think carefully about your coil arrangement- if you are looking to make the most heating surface in the space available then why not use a 'twin' coil arrangement- an inner and outer coil?

The outer, say 5/16", tube would be a coil of a fairly flat gradient, taking the cold feed from it's base where it would flash and rise to what I would propose is a stainless 'header', which would the connect to an inner steeper coil of say 1/4" tube that takes the now very dry steam back down through the products of combustion, through the flames and the I would really suggest that you again make use of convection and take the steam outlet back up, through the center and have the outlet at the top of the boiler.

This would make a very compact heating exchanger, but very high heating surface and very dry, very hot steam!

The full size monotube boilers I've seen all use this method, along with every twin coil heat exchanger I've come across.

I very much look forward to seeing this launch make progress- what plans are you using- have you found some original Georgian ones or is it one of Selway Fisher's?

Greg

[kno3]

This is a really interesting project and I am thinking about building such a coil boiler too. Do you hink it could work on a much smaller scale? In a size of about 10 cm length and 7 cm diameter?

[flashtwo]

Hello George,

The centre 15mm copper tube serves two purposes -

1) The feed water connector is located at the bottom diametrically opposite the steam outlet connector and the feed will be directed through a 4mm pipe up inside the 15mm copper pipe to the top of the stack  from whence it spirals back down. Taking this route means it doesn't interfere with the return pipe spriraling down.

2) The gap between the spiral layers needs to be maintained for good heat circulation, so the 15mm pipe will be drilled to accept 1/8th inch copper pipe sections to act as spacers. The lowest spacers are stainless steel (for strength) rod inserted through copper pipe, which will support the weight of the coil.

When I start the system, I always flood feed right through to the cylinders and, since I have not throttle valve between the boiler and engine, as the steam is raise the feed is blown out of the cylinders (I toggle the reversing gear) and the engine warms up with the boiler, i.e. there is no sudden shock of high temperature steam.

I can see what you mean by having a ceramic burner. The design could be changed so the 15mm pipe terminates higher up and leaves room for a ceramic burner.

Are ceramic burners rated by their gas through put? - I've been aiming for 7 grams/min, though I'm getting very performance at 4 grams/min.

I shall post photos as the construction progresses.


Best regards

Ian.

[ooyah/2]

Hi Ian,
Gotcha, as the coil is vertical with the heat being put into it the coils could start to slip down and prevent the heat circulating between the coils, so you St/St rods keep them apart, good idea.
Not sure about the gas consumption with ceramic burners but any that I have made work very well on a No5 jet, here is a pic of one with a No6 jet below a vertical multitube boiler 4.5" dia x 6" between end plates.
I have never been very concerned with the gas consumption as I refill the camping gas cans with Butane fro a 4.5kg bottle, this of course is used in my steam tug not the flash steamer, it has three burners with .025" jets pressure fed from the engine and uses about 3/4 pint of fuel per 4 mins.

Are you going to computerise the boat as the previous one?

George.

[SteamboatPhil]

Nice job Ian, can we have than on the stand at Ally Pally as well ?   

[flashtwo]

Hi Greg,

The lower part of the coil will be constructed in three layers and the part in the stack will be a single layer in contact with the brass tube. The three layer section is wound with a 3.5mm (on average!) gap between the 22 spirals and with the vertical separation maintained with sections of 1/8th inch copper tube. I have a spread sheet that I can input the heating area required, tube size, gap size etc.,  and it calculates the overall dimensions of the boiler - so I can play around with the numbers to get the boiler shape that I require.

On my steam jet engine project (see the other thread) I used a 5 layered coil and I did achieve steam temperatures in excess of 720degC - yes the silver solder did melt!

I think with the horizontally coiled flash boilers, that I’ve been using, couldn't have had much natural circulation, since any density variation causing circulation must have been cancelled out along the length. In a normal boiler, as you know,  the circulation occurs between the lower headers and the drum and requires lots of relatively large diameter tubes. With the monotube flash boiler with a small diameter tube the flow relies on continuous feed pumping to maintain flow through the narrow tube. Anyway, its all part of experimental fun, which, if it is successful, could be applied to larger vessels – all part of modelling I suppose.

I have been wary of using any “header” in the system, since it might constitute being a pressure vessel, which is what I’m trying to avoid with all the testing etc.

Yes, I’ve got the Selway Fisher Edwardian 30 foot launch plans and their detailed booklet on strip planking techniques. I’m aiming to use the 5 inch prop that’s currently installed in Vital Byte. The boat is planned to be of  75 inch length  and the boiler is scaled to fit. I’ve never tried strip-planking before, so its back to the bottom of the learning curve!

Ian

[flashtwo]

Hi Kno3,

I can’t see any reason why smaller versions couldn’t be made. The required copper-to-water (not combustion gases-to copper) surface area needs to be converted into tube length knowing the tube diameter and wall thickness. The are many ways of coiling the tube:-

1)   Single layer results in a long boiler
2)   Multilayer results in a short boiler
3)   Layering the coils by starting at one end on the inner coil over a former then returning the opposite direction over another former (i.e. a strong plastic tube that slides over the inner coil) and repeating this for the number of layers required. The only problem with this arrangement is that you end up with several very hot spots each time the coil approaches the burner.
4)   Making the layers by spiralling out from the centre to the rim, thus making a spiral “disk”, then moving up to the next disk and spiral in from the rim to the centre and repeat until the number of disks is achieved. This takes a bit more skill, but it means that the coil gradually moves from the cold end to the hot end without the hot spots of the method 3) above.
5)   Figure of 8 is used a lot by the traditional flash steam tethered boat fraternity.


Ian

[flashtwo]

Hi George,

I’m only using copper covered stainless steel supports at the bottom and the rest are copper tube spacers. I’m trying to avoid contact of steel against the copper boiler tube to avoid corrosion.

Blimey! that’s a lot of fuel you’re using on your flash boiler it must go like a rocket. On my steam jet engine I’ve been using 25 grams/min of gas and I thought that was a lot.

Yes, this will also be computerised. The computer based controller will simply unplug from Vital Byte and plug into the new model, though I might have to tweak some of the control settings (via the keyboard) for the new boiler depending on how it raises steam in comparison with the current horizontal boiler.

Thanks for the photos. The stack is a little bit smaller diameter than mine, eh! What device is the V10 driving – is it a generator?

Ian.

[flashtwo]

Hi Phil,

Yep, it will only be installed in Vital Byte for test purposes come Ally Pally - no sleak mahogany steam launch as yet I'm afraid.

I'm still working on the steam jet boat, though the vertical boiler is the priority at the moment, and I hope to have that at Ally Pally as well, if you can find some extra table space; I've an idea about converting it into a turbo-steam jet that drives the boat's propellor, if the direct water drive doesn't come up with the goodies.

I'll be seeing you at the Model Enigineering Exhibtion Sandown in December - I've not been there before.


All the best
Ian