Topic: Matching water pump to steam engine (Read 12460 times)

kno3 · « **on:** April 15, 2009, 06:15:56 pm »

Recently I came across a Cheddar water pump which I have fitted to my Puffin engine. I was surprised to see that the pump's piston is tiny, just a very thin metal rod, around 2 mm in diameter, yet the pump is able to overfill the boiler without a bypass valve fitted.

Since I am thinking of adding a water pump to my Graham TVR1A steam engine too, I was wondering if there is way to calculate the pump's displacement in relation to the engine displacement?

Thanks for any advice!

Bunkerbarge · « **Reply #1 on:** April 15, 2009, 08:58:51 pm »

The trouble is that the load on the boiler is varying depending on the load and therefore steam required by the engine. Consequently you want the water pump to be capable of overfillig the boiler so that you can then control it to suit the demand, either manualy or automatically.

As for size don't forget that steam is about 13,000 times the volume of water so the steam being used in the engine relates to a very small quantity of water.

flashtwo · « **Reply #2 on:** April 16, 2009, 12:33:03 pm »

Hi,

Assuming the pump ram is connected directly to the engine and its strokes per minute is the same as the engine RPM, then you work out the steam VOLUME consumption per rev:-

Volume = stroke x (diameter/2)squared x PI x number of cylinders

and double the Volume if double acting. (PI = 3.142). To be more accurate you should also consider the thickness of the piston and the volume it takes up.

Looking up the "Model Engineers Handbook" by Tubal Cain, section 8 we have the skeleton steam tables (you don't want to see the complete steam tables !), then for a steam pressure of 50psi (gauge) water expands 414 times, therefore dividing the steam volume by 414 will then give the water volume per stroke. If the steam was being exhausted at atmospheric pressure, then water expands 1665 times.

Once you have the water volume, you multiply it by a pump efficiency value (I use 1.15), since the pump clack valves do not close immediately and the pump may be only 87% efficient. From the water volume you can then work out the pump ram diameter and stroke.

The above calculations will get you in the right ball park, but steam leaks, pressure drop across the regulator valve and varying boiler pressure will complicate matters.

Ian.

kno3 · « **Reply #3 on:** April 16, 2009, 03:44:59 pm »

Thanks!

I'll look up the bore/stroke and try to calculate this tonight.

Bunkerbarge · « **Reply #4 on:** April 16, 2009, 11:40:02 pm »

Actually I meant 1,300 not 13,000! to atmosphere of course but you're right anyway, it's nearer 1,600.

benjaml1 · « **Reply #5 on:** November 28, 2009, 09:43:07 am »

I'm new to this steam lark & find this gem of a forum invaluable...

benjaml1 · « **Reply #6 on:** February 06, 2010, 06:47:50 pm »

Something I don't understand.... OK theres a commercial water tank of 2 1/2" in diameter X 3" tall. ( <14.7 cubic inches) Recommended for 2 cylinder engines up to about 1" bore, including double action.... Using the above formula my single action 2 cylinder 3/4 X 3/4 bore stroke @ 2000 RPM ( 3.2 cubic in/min ??) would eat that up in less than 5 minutes... Where am I going wrong ?

SteamboatPhil · « **Reply #7 on:** February 06, 2010, 07:43:18 pm »

Arh but remember its a steam engine using steam, which of course is expanding. The total water content is not what the engine is using, its just topping up the exhusted steam consumed (in a nutshell)

benjaml1 · « **Reply #8 on:** February 06, 2010, 07:58:34 pm »

So instead of using the factor of 414 you use 1665 in the calculations for water replenishment if indeed it is not closed circuit recovery ?

benjaml1 · « **Reply #9 on:** February 06, 2010, 10:25:04 pm »

Quote from: Steamboatphil on February 06, 2010, 07:43:18 pm

Arh but remember its a steam engine using steam, which of course is expanding. The total water content is not what the engine is using, its just topping up the exhusted steam consumed (in a nutshell)

I think I understand.....

benjaml1 · « **Reply #10 on:** February 21, 2010, 03:32:44 pm »

Built from scratch to my own design, 12VDC electric motor ( in a dummy casing) powered boiler feedwater pump, 8mm ram x 25mm stroke... 30-15 RPM depending what choice of motor I use. Total platform length 11 inches....Bottom picture a mock up, water tank, separator tank & pipe work yet to be installed

kno3 · « **Reply #11 on:** February 21, 2010, 06:38:28 pm »

If I understand correctly, the whole pump body moves up and down with the crankshaft and piston? It's looking very good.

benjaml1 · « **Reply #12 on:** February 21, 2010, 06:41:40 pm »

Exactly, I didn't want side thrusts on the ram seal so mounted the clack valves remotely & connected with a flex hose. Everything was made from metal stock apart from commercial clack valves & flanges. ...

SteamboatPhil · « **Reply #13 on:** February 21, 2010, 10:12:05 pm »

So the flexi pipe works as suction / force with the clacks (inlet/out) remoted (as it were). Nice idea I like it. Will the electric motor be controlled or just be "on" all the time.
A good set up as it removes a fair bit of load from the engine. I'm am doing something similar for my new flash steam plant (watch this space details to come) although I will be gearing the motor and running it from a ESC.

benjaml1 · « **Reply #14 on:** February 21, 2010, 10:19:58 pm »

Quote from: Steamboatphil on February 21, 2010, 10:12:05 pm

So the flexi pipe works as suction / force with the clacks (inlet/out) remoted (as it were). Nice idea I like it. Will the electric motor be controlled or just be "on" all the time.
A good set up as it removes a fair bit of load from the engine. I'm am doing something similar for my new flash steam plant (watch this space details to come) although I will be gearing the motor and running it from a ESC.

Exactly, one hose with flow in & out. The important seal is the water/air interface which is a lip seal, the ram doesn't need a seal/O-ring as it's pure volume displacement. It is important to match the torque of the motor to force required to move the ram against full pressure. Roughly, I have a factor of 3 on the 15 rpm & a factor of 2 on the 30 rpm motors as it also has to overcome seal friction. Motor may be controlled with a timer or a level control device, I'll go with the timer first....

kno3 · « **Reply #15 on:** February 22, 2010, 12:06:10 pm »

What exactly is a lip seal?

benjaml1 · « **Reply #16 on:** February 22, 2010, 12:32:51 pm »

Quote from: kno3 on February 22, 2010, 12:06:10 pm

What exactly is a lip seal?

A lip seal is an annular "L" shaped seal with the flange towards the pressure side. Line pressure forces the flange onto the shaft to obtain a seal. There is also a spring around the lip to maintain it's position without line pressure... The joy is ( unlike O-ring or packing seals) it doesn't require constant adjustment....

http://www.lipseals.net/

kno3 · « **Reply #17 on:** February 22, 2010, 09:12:22 pm »

Thanks! Learned something new

benjaml1 · « **Reply #18 on:** February 22, 2010, 11:22:48 pm »

You're very welcome, I have learned a lot from this site... Just trying to share...

Bernhard · « **Reply #19 on:** February 27, 2010, 07:29:14 pm »

Technische Daten:

Breite 20,5 mm http://cgi.ebay.de/Regner-20220-RC-Speisepumpe-Neuware_W0QQitemZ270523446323QQcmdZViewItemQQptZDE_Modellbau_Modelleisenbahnen?hash=item3efc743433
Länge 85 mm
Höhe 50 mm
Gewicht 130 mm
Pumpendruck bis 7 bar
Hub 10 mm Working http://www.youtube.com/watch?v=Loy2qdXYaCk
Kolbendurchmesser 4 mm
Fördermenge 360 ml/h

HS93 (RIP) · « **Reply #20 on:** February 27, 2010, 10:09:33 pm »

I'm going to be out in the shop tonight building one, thanks for that

Peter

Bernhard · « **Reply #21 on:** February 28, 2010, 07:42:14 am »

Hi..Great rember photo,,,,And ,,This one work reale great i have 2 ,,one in lady Jane
http://www.home.earthlink.net/~bfindus/index.html

HS93 (RIP) · « **Reply #22 on:** February 28, 2010, 08:44:57 am »

has anybody got any plans or links to a small pump of this type, one thing I have always had problems with is why you need a one way valve on the output side and one on the boiler the same as i have always been told water will not compress so they are both doing the same job ?

Peter

benjaml1 · « **Reply #23 on:** February 28, 2010, 10:37:34 am »

Quote from: HS93 on February 28, 2010, 08:44:57 am

has anybody got any plans or links to a small pump of this type, one thing I have always had problems with is why you need a one way valve on the output side and one on the boiler the same as i have always been told water will not compress so they are both doing the same job ?

Peter

I have installed two non return/check valves in series also. In the case of a single unit failure, you won't rapidly lose boiler level. My reasoning is for safety, both equipment & personal... Maybe overkill but in the real world of industry, check valve failures are common, especially gravity ball checks....

benjaml1 · « **Reply #24 on:** March 07, 2010, 09:38:19 pm »

Apologies about the poor quality of the vid but I think you will get the idea of the workings...It works/pumps well

http://s18.photobucket.com/albums/b136/benjaml1/?action=view&current=P1010875.flv

News:

Author Topic: Matching water pump to steam engine (Read 12460 times)

kno3

Bunkerbarge

kno3

Bunkerbarge

benjaml1

benjaml1

benjaml1

benjaml1

benjaml1

kno3

benjaml1

benjaml1

kno3

benjaml1

kno3

benjaml1

Bernhard

Bernhard

benjaml1

benjaml1