Boiler feed pump, or injector, or?

[Bob WorsleyParticipant @bobworsley31976]

The latest ME had a letter about injectors and boiler feed pumps, this referred back to a previous article. Having read this article, with such a silly title I hadn’t noticed it before, I have some comments.

I am slowly reading through a large pile of Mechanics Magazines from the early to mid 1800’s. Quite by accident I came across an article that I noted down, that discussed boiler feed pumps to high pressure boilers. Copy of article attached.

What caught my eye was the pump design that didn’t need to force feed water into the boiler against the steam pressure. Briefly it worked by filling a valve up with water at atmospheric pressure and then swapping the valve to be open to the boiler pressure. At that point the entrapped water would simply run out of the valve and into the boiler. It works on the principle that if the valve is completely filled then there is no sensible change in the volume of water.

There are obviously losses associated with the rotation of the valve, less that forcing water against the boiler pressure? Both a feed pump and an injector are lossy devices. Either from the raise in pressure to force the water into the boiler, or the loss in energy of the steam as it is condensed. I have yet to come across an article that actually defines these losses.

With this rotating valve there is a requirement for the feed water to be above the valve, and the valve above the boiler, so that gravity is used transfer the water in both cases. This is probably not ideal, some form of pump pressure in and out would be better. But, again, because water is incompressible then there will be minimum losses.

Something to think about?

 

[Howard LewisParticipant @howardlewis46836]

Sounds rather like perpetual motion, lubricated by snake oil.

Any pipe or vessel, at a pressure below that of the boiler, when connected to it will see an outflow FROM the boiler.

So a “one sided” stop cock, filled with water from the hot well, will contain water that is below boiler pressure and temperature.

How is the water propelled from the hot well into the pipe to fill the “stop cock”?

I really cannmot see how the water could then be induced to enter the boiler against the pressure.

Boilers are topped up by means of pumps, or Ms Giffard’s clever use of cones and venturis to convert fluid velocity into pressure, in an injector.

For a flow, there must be a pressure differential, so that the external pressure exceeds the boiler pressure.

If I am wrong, please explain the theory

Howard

 

[noel shelleyParticipant @noelshelley55608]

My first thought would be wear ! If it fails what stops the contents of the boiler from being discharged ? Was it a good idea ? Since it did not come into general use may be NO ! The injector Howard mentions looses almost no power(steam ), The mechanical feed water pump (weir or crank driven)properly set up will keep the boilers requirement for feed water satisfied. Noel.

PS if this WAS early 1800s then the Giffard injector did not exist  and we had not long been using STRONG steam ?

[duncan webster 1Participant @duncanwebster1]

In principle it works, BUT

as soon as the water filled void is exposed to steam a lot of steam will be condensed. Not a loss as it heats the water. As long as the void is higher than the boiler gravity will drive the flow

After the water has run out, the void is full of steam. When you switch back to refill, this steam will exhaust back into the environment. I suppose you could somehow pass it through a heat exchanger.

A injector sounds a lot simpler to me

[duncan webster 1]

On duncan webster 1 Said:

In principle it works, BUT

as soon as the water filled void is exposed to steam a lot of steam will be condensed. Not a loss as it heats the water. As long as the void is higher than the boiler gravity will drive the flow

After the water has run out, the void is full of steam. When you switch back to refill, this steam will exhaust back into the environment. …

Looking at it in reverse, when the cavity is open to the boiler, it is filled with boiling water at full steam pressure and temperature.   Thus, when the cock is turned a slug of energy fires back into the cold-water system.  High pressure in 1833 might have been manageable, but I’d expect violent results from a slug propelled by modern high-pressure where a 10 atmosphere boiler is very much on the dinky side.

As a result, I think the 1833 mechanism would put a small amount of cold-water into the boiler in exchange for blowing a lot of heat into the cold-water tank.  Very inefficient compared with a force pump or injector; thermally, these are one way systems that don’t give the boiler much opportunity to leak heat out into the feed system.

Keeping a rotary valve steam tight would be an engineering challenge too.  I also think the system would consume a lot of power because the turning part is pushed hard into the body by steam pressure.   Took a few Horse Power to move full-size locomotive slide valves, which is why effort went into designing balanced and mushroom valves.  In comparison, the bearing surface of a force pump is exposed to far less friction and probably easier to lubricate.   An injector has no moving parts at all.

So I think the 1833 system would work at low pressures, but waste a lot of heat, and I strongly suspect it becomes less practical with rising boiler pressure.

None of my reading on steam and boilers mentions anything like this system on a real boiler.   Perhaps they were so common, no-one mentioned them!   But being a suspicious Old Henry, I guess the idea didn’t work as well as the alternatives.

Dave

 

[Charles LamontParticipant @charleslamont71117]

SillyOldDuffer:

I1) As already explained, the principle works if the cock and water supply are above the boiler water level.

2) The dollop of steam in the valve is of much smaller mass than the incoming water. Given a reasonable head of water above the valve, the steam would be immediately condensed (possibly fast enough to hammer).

3) The condensate would go back into the boiler.

3) The boiler pressure is tending to blow the valve out of its seat, not in.

No, it does not seem a very practical proposition, and that is most likely why we have never heard of it.

[Nigel Graham 2Participant @nigelgraham2]

Few full-size locomotives were fitted with feed-pumps, though those that were, used steam-driven ones whose overall efficiency must be about the lowest behind axle / crankshaft driven pumps and injectors.

Instead they used two or three injectors, which if designed properly for the boiler should give lower losses than anything else. (Live steam plus exhaust-steam injector, a very complicated device as it needs a supplementary live-steam feed controlled by an automatic relay valve; and an oil-separator.)

It’s instructive to read D.A.G. Brown’s book on making injectors for miniature locomotives. Presumably too, traction-engines whose own crankshafts can revolve unsuitably fast for directly driving pumps; but the prototype practice was to fit one injector and a mechanical pump.

He makes the point that there is no objective reason why such a locomotive cannot be fitted with two injectors of appropriate flow-rates.

The boiler inspection code specifies at least two independent feed methods, not what type they are. You can still hide a hand-pump in the tank or tender, or under the footplate, if you want; but I am not sure if they are really quite as effective as we hope.

‘

That suggested “valve” is of historical interest, but probably best kept to replicating early, low-pressure engines such as those made for draining mines. These worked at such low pressure that some seemed to have used pure gravity feed.

[duncan webster 1Participant @duncanwebster1]

Quite a lot of early locos had water pumps driven off the crosshead. Big problem was you can’t put water into the boiler if you’re stuck in a siding. Mf Giffard injector got over this problem. Only people who are very good at driving models manage without some kind of mechanical pump. If you let the pressure get down your injectors won’t work

[JasonBModerator @jasonb]

Sounds no different to the oil pots used on many a steam engine. Open top cock and fill with oil then close. Open bottom cock and oil finds it’s way into the cylinder or valve chest.

Main reason not to use it on models is it won’t look scale on a loco or traction engine. But could be an option for those that are not worried by looks

[Bob WorsleyParticipant @bobworsley31976]

Thank you for your comments. The first article was in 1826 but I couldn’t make any sense of the woodcut, so just noted it down. The second article, as copied, was from 1838.

The injector consumes a huge amount of energy when working, but so far not seen any figures. These articles pre-date the injector, so all locos must have used feed pumps.

If you rotated the valve at a certain wheel speed, then also had a small force pump running in synchronism to push atmospheric water into the valve, then 1/2 revolution later another pump to pump the water out? The whole idea here is to not have to use the pump to raise the pressure of the water in the valve from atmospheric to boiler. If the pump is full then there is no loss in raising the pump contents from atmospheric to boiler pressure. That is where the sneaky part of the pump works.

Not certain the idea of using gravity to both fill and empty the valve is really workable, hence the addition of a small force pump to empty and fill the valve.

A displacement lubricator isn’t the same, whilst it runs at full boiler pressure it is just the replacing of oil by condensed water to force the oil out.

Does the mentioned injector book talk about the losses?

[noel shelleyParticipant @noelshelley55608]

From where have you got the idea that an injector uses a HUGE amount of energy ? Look at the thermal capacity of water – work out the mechanics and theremo dynamics of this device. I don’t think it will look good compared to other methods of boiler feedwater suply. Noel.

[duncan webster 1Participant @duncanwebster1]

With an injector, nearly all the energy which leaves the boiler as steam goes back in as hot high pressure water. There will be a small heat loss from the pipework.

An exhaust injector is even better, it uses steam which would otherwise have gone up the chimney to waste. I’m waiting for one of you clever people to make one at 1:12 scale. Not me, I can’t even make live steam ones

[JasonBModerator @jasonb]

I was not talking of displacement lubricators but oil/fat pot types.

[Bob WorsleyParticipant @bobworsley31976]

I see that the Brown book on injectors is over £100, but the Laurie series on injectors in the ME looks a very comprehensive alternative. In 1986, volumes 156 and 157. Says that efficiency is 98.5%.

 

 

[noel shelleyParticipant @noelshelley55608]

DAG Browns book on injectors £100 ? I think mine was about £12 new ! I think it will be available from Tee unless out of print. From the above figures do you now accept that far from consuming HUGE amounts of energy the live steam injector is in fact a very efficient device for the supply of feed water to a boiler in steam ? Noel.

[ChrisLHParticipant @chrislh]

DAG Brown’s “Miniature Injectors Inside and Out” from Tee Publishing £14.95 plus £2.65 p&p. Tee have a very fussy search engine, needs all the i’s dotted and tees crossed before it will have anything to do with your query !

[Nigel Graham 2Participant @nigelgraham2]

I verified TEE’s catalogue for this.

D.A.G. Brown’s Miniature Injectors Inside and Out is still listed at the £14.95 as on the sticker on the back of my copy!

Someone wanting £100 on Eek for it is taking the whatsit. Never assume these Internet general-sales sites are the first options for anything just ‘cos they are on-line and big.

 

You don’t need to use TEE Publishing’ search bar.

Use the catalogue’s own contents list to open the relevant section, then scroll through the offerings till you find what you want. I had no problem at all finding this book there.

 

The exhaust-steam injector Duncan mentions is a complicated beast indeed, and it does not use only steam from the blast-pipe, via an oil-trap. It also has a supplementary live-steam cone fed via a relay valve that opens when the exhaust pressure drops (as when the driver closes the regulator). I have never heard of anyone making a working miniature version, and it’s possible the exhaust pressure would be far too low. That more so downstream of the oil-trap – I think many of us tend to err on the side of safety and slightly over-oil our engine cylinders. It might be an interesting challenge, perhaps better at one-eighth-ish (7.25″ g) than one-twelfth-ish (5″ g) scale, for the builder prepared to experiment.

For fine-scale work you could make a “simple” live-steam injector disguised externally as an exhaust-steam unit, and I believe that has been done by some builders.

 

Incidentally there is another book about injectors on the TEE list but it may be a more general one, and possibly about full-size practice. I’ve not seen it for myself.

…….

 

Re the facsimile introducing this thread, the author or editor has already one spotted one flaw with that proposed “Rotary Valve” – enough to elicit a terse footnote.

Intiguingly because we can see only a vertical half it, the next letter is about a “Revolving Slide Rest” – not relevant in this forum about boiler fittings, but perhaps so on another. What’s visible seems to hint at ball-turning among other uses…. perhaps for making Boiler Feed-water fittings.

[Author]

Posts