5" Britannia Injector

Have just (almost) finished my 5 " Britannia. It had its first run in steam today.

All was well with the exception of the injector which refused to work.

It is fitted with a brand new 40 oz Reeves injector piped with 1/4" pipes and a Don Hewson Clack valve into the boiler top.

Clack valve is OK and identical to the one fitted on the axle pump input to the boiler. Finding nothing wrong with my build I am trying to find a solution to the problem.

Is the injector I have selected , too high a capacity for the a Britannia sized boiler ?

40oz is a bit steep isn't it.?

A 12 oz would easily keep up with a Britannia at full chat. If you need the calculations see DAG Browns book on injectors.24 oz maximum I'd suggest, but a 12 won't cool off the boiler - but you might need a bit more forethought and planning using it. I just run with the injector on most of the time, because its sized quite well and one can largely forget about the pump. Its a lot more efficient than the pump too.

Clack - you don't need to restrict the lift to 30 odd thou, since the ball is not trying to be caught up by the pump frequency. You can afford to let it lift 1/8" or a touch less to allow it to get out of the way and minimise restriction.

Having been through injector woes, I can tell you what I did. First I tested the injector on anohter model. It worked.

Next step was to look along the lines for a blockage - I used Locktite 574 in the joints. There weren't any. Then I checked the water lines were big enough to keep water velocity down. They were (3/16 pipe foir a 12 oz injector.) See DAG browns book about sizing lines.

So it had to be the steam supply. (since it wasn't water and boiler feed lines) Sure enough, making sure the steam feed was properly straight, and removing a nice scale elbow sorted all- because by then the penny had dropped. Its a pressure recovery device, and if one is inducing pressure drops in the plumbing then its not available for pumping water.

Exit all injector woes, and it lights first time everytime.

I don't think there is anyhtng magic about injectors, or the diagnosis of basic problems - one just has to plod logically through the steps to sort the problem. (borrrow a working injector etc) Personally I was convinced there was a blockage, because the injector would strike but then almost immediately vent steam . Well there was - the ball wouldn't lift becase the pressure had been lost in the line. (Just not the normal type of blockage!) So the injector was seeing a blocked line.

Many thanks. It looks like I scored an own goal by fitting the wrong size injector.

Further work last night indicated that the aperture through the Hewson Clack valves was smaller than the originals designed for the Loco. I bought the Hewsons because they looked the part but I failed to do a flow path check.

I have a new injector en route and I will search out a copy of DAG Browns book

Again many thanks for a very informative and helpful reply

Fitting an oversize injector will just mean that it feeds too much water and knocks down the boiler pressure too fast - unless grossly too big it will still work .

Most common cause of injector problems on all Britannia , 9F , West Country and similar engines is the location of the injector when put in its scale position right next to that enormous grate and ashpan . Apart from the constant source of dirt the injector can get very hot and even a just moderately warm injector won't start unless flooded with water first to cool it down . Have the biggest practical airgap between ashpan and injector - the difference between nearly touching and a 3/4 inch gap could make all the difference between sucess and failure .

There is a fundamental difference between a mechanical pump and an injector . A mechanical pump which sees some flow restriction on the output side will just generate more and more pressure until it forces fluid through the restriction or something breaks or the pump stalls . An injector on the other hand can only generate pressure with a quite low capped value and if it sees any restriction on the output side there is very little reserve pressure to force fluid through .

Restriction in the feed lines mostly means holes too small somewhere . The flow velocities are not high enough for pipe bends to matter too much unless particularly tight . The main problem will be in fittings with drilled holes meeting other drilled holes at acute angles or small holes suddenly becoming big holes (or vice versa) ..

Michael Williams .

Ianmac. If I could sugest a couple of other things.

As Michael Williams suggests the injector must be kept cool - up to a point. The logic is that the presure drop in an injector is very great, which means that warm injectors will cause boiling. Or more precisely warm water will boil  in hte injector, to the detriment of all!

I'm less convinced about keeping the injector rather than the feed cool. My experience (not great ) is that if you have a decent water supply, then it doesn't matter if you start steam first or water first. (ie with a red hot injector) mostly on the TE I actually open the steam valve first - less mess. The suction draws the water, and bang , she lights first time. The point being that steam first establishes a decent suction jet. People will argue., but thats how mine work. Anyway, heat is a point to watch. Small heat shield if necessary?

Pipes - a pipe is only as big as its smallest dia, so I am careful about fittings olives and bends.

You need a filter in the line. Being a tender loco I imagine you have one, but if not, it might be wise to fit one.(of sufficient capacity)

DAGs book is I think very good. If it has a weakness, he is less positive on the need for a good steam supply. If you look at the flow factors for various fittings, its horrendous the equivalent line size drop they create. One is losing pressure all the time. Good smooth run, and I wouldn't ever use a globe valve for a steam control again. The injector ones impose far less restriction.

all the best.

Edited By mgj on 17/04/2012 19:10:01

The injector is fitted at least 3/4" out from the ashpan on a mounting plate that acts as a heatshield with free airflow around it. I will ensure that the incoming water supply is as far away from the ashpan as possible.

I have replumbed today in readiness for a 24 oz injector designed to operate with 3/16 pipe. DAG's book is en route. I look forward to reading it - perhaps I should have had it to hand much earlier in this build.

My conclusion to date is that I fitted an injector with an output that exceeded the flow capacity of the boiler clack valve.

As a side issue on inspecting the clack valves supplied by Hewson they were fitted with standard steel balls not stainless or bronze as we have to use in this part of the UK.

Many thanks for the feedback

On the positive side the Brit (Robert Burns 70006) steamed like a witch to 80 psi. all present were suggesting ways of slowing it down a bit.

Will get there

Nothing to do with original posting but Ianmac's mention of a protection plate for his injector reminded me of this :

In Swansea where I live there were once several major railway stations . One of these was the LNWR Swansea Victoria station . The line out of Victoria went out on a causeway right along the seafront , parallel to the Mumbles train for the first part , and then turned inland to Pontardulais where it joined the LNWR/GWR joint Central Wales line to Shrewsbury .

There was a suburban service worked for a long time by LNWR Coal Tank 0-6-2 's . The causeway carrying the track was completely exposed to sea and large quantities of sea and sand got washed over the line , engines and train almost all the time .

The poor Coal Tanks had to work in this and experienced very bad wear of working parts and sand in the feed water causing all sorts of problems . Apparantly local shed officials had several times approached Crewe for some protection devices to be fitted on the allocation of engines for Swansea but had they been turned down flat . It seemed Crewe thought it more important to keep all engines standard rather than do something to stop them wearing out .

The problem was finally addressed (very) unofficially by the shed staff . Over a period of time the engines acquired an almost complete underpan , sand guards all over the place , water filters on everything , side screens to cab and specially made bungs to stopper all hoses not connected up to trains . All made out of scrounged material and rubbish . Problem solved .

After numerous years the engines got sent back went back to Crewe for major overhaul . You can guess what happened next - Crewe stripped off all the protection gear , scrapped it and sent the engines back in nice shiny paint in completely standard condition .

Apparently the shed staff at Victoria were a bit cross .

All this was before my time and I doubt whether it appears in official records but the tale was told to me by a retired driver and is almost certainly true .

Edited By MICHAEL WILLIAMS on 18/04/2012 10:51:50

New Blackgates 26 oz 3/16 injector fitted. Piping reworked to accomodate as per DAG's book. No air leaks in the injector supply water path. (excellent book)

Trial yesterday was only partially successful. Got the injector to work but far too fiddly and still some water emanating from the injector drain

I'm back to the boiler Clack valve not sealing correctly . It was supplied by hewson and has a white ring inside it which the ball seals against. The ring is held inside by and internal screwed collar which in mine partially obstructs the flow hole into the boiler which cant help things!

Next move will be to remove the ball from the boiler clack and fit an inline clack close to the injector output ..  Anyone gone down this route ?

Edited By Ianmac on 28/04/2012 09:54:09

(1) If boiler clack is faulty it may be in one of two ways :

(a) Constricted so that high input pressure needed to force water in .

(b) Leaking so that steam bleeds back to injector .

(2) Test the clack 'as is' .

(a) Connect up a small handpump with pressure gauge and just see how difficult it is to pump water in against boiler pressure and how high the pump delivery pressure has to be before water enters boiler .

(b) See if pump feels spongy , gets warm or if bubbles form in feed tank . These indicate steam leaking back .

(3) Just possibly a fault exists somewhere else . A common one is that the steam valve supplying injector is blocked or inadequate size .

More work done .

The boiler clack was leaking. Bought from Hewson for £50 ea. I was a bit .. off. This clack uses a hard nylon seal for the ball. It relies on its underside meeting a perfectly smooth surface to provide a seal within the clack chamber. This surface on both clacks had the marks of 4 tap flutes nicely imprinted on it.

A cutter was made and both clacks cleaned up .Centre hole reamed to size and new balls fitted to make a conventional clack.

Ready to retest as soon as I can after my impending unmoveable holiday.

Final test done today.

Hewson clack was still unreliable. Ball removed and it is now being used only as a entry into the boiler.

In line clack fitted in the pipe between the injector and boiler, just after the injector in line with the slope on the firebox and all now works.