Kit 22 - The penultimate kit
Continued

ModelWorks 4 inch Scale Burrell Agricultural Traction Engine
Kit 22 was delivered Monday 19th September

When I brought water treatment from Live Steam Models I also brought 1/2 metre of their 3mm PTFE gland packing.  It turned out to be square despite the catalogue saying 'dia' (square is good!).   I think this is similar to that recommended by Steve.  The packing from LSM works out at about the same price as RS but you don't have to buy 8 meters.

So far I have only tackled the valve chest as it was leaking badly.  The way I fitted the packing was to wrap it around the shaft and cut rings using a scalpel.  You have to waste a bit at the end so if possible cut all the rings that will be needed  for each shaft in one go.   I managed to fit 3 rings into the stuffing box offsetting the joints between each piece.

Over the next few days I shall remove the gear change assembly for painting and permanently fit the water gauge.  I have found that this kit needs a lot of TLC (tender loving care) and it is well worth spending the time to get it right.

Friday 30th: Time to stop hating the pipes and give them some TLC.  When I fitted the sight glass drain I found that I had a tangle of pipes and I wanted to sort this out.  Now that I have a former I am able to bend the pipes whilst hard which has the advantage that they are less likely to collapse.  I moved the bends of the water lifter's steam feed so that the pipe went the other side of the feed to the injector.

In the picture right the blue felt tip shows the original position of the bend.

The result is a lot neater and I have also been able to arrange it so that the pipes go out of the tender parallel one on top of the other. 

The bracket supplied was designed to support one pipe as it emerges in the centre of the tender's slot.  Clearly I could not use this bracket so I made a new one.  The photo montage right shows the new bracket at various stages of manufacture.  In a bar 1/2 x 5/16" I drilled holes for the 3/8 & 5/16 pipes and them immediately milled off the top half.  I considered this the easiest way of producing the moon shape recesses.   As I could not use the same holes to fix the support to the tender and for the clamp plate I drilled and tapped 4 x M4 holes.  Finally I made a clamp plate from brass sheeting.

I had no intention of totally rebending the pipe where it delvers steam to the water lifter but when my attempts to get the original pipe to run parallel to the top of the tender went awry I decided to have a go.

I completely straightened the "outside" portion of the pipe (without annealing).  As I could see I would have plenty of pipe  I cut off about 5" to have a trial go at forming the bends.  This went well but when I tried to do it for real I found it very difficult to get the bends tight enough to arrive at the union.

In order to get things compact enough I found that I had to run the two 90 degree bends together ie. start the next before the first is finished.   I had a degree of success but I now fully appreciate why ModelWorks chose to route the pipe the long way round.  The original routing can be see in the picture right (courtesy of Mike).

My final task was to sort out the water feed from injector to boiler.  I found that having bent the pipe so as to pass over the two pipe supports it arrived at the injector 1" below the union.  

On rebending the final kink so that it arrived at the union I found I had too much pipe.  Rather than muck around inserting addition bends to lose the extra I opted for the drastic solution and cut off the excess.

My TLC may appear brutal but you have to be cruel to be kind!

Reverser Quadrant:  Mike recently contacted me concerning a problem he encountered after running his Burrell on air.  He was having trouble with the bolts that secure the two parts of the reach road together coming loose.  I to had this problem and nearly welded them together before discovering the route cause.  As it is clear many Burrells will be affected I think it is time I wrote it up - so here goes.

12/4/06: In the end I got the welder out!

Despite the new quadrant the die block is hitting the end of the expansion link whenever full gear is selected.  It only does this for part of the stroke and there is enough give in the linkage for the engine to run OK.  Unfortunately the continual twanging of the reach rod causes the bolts to come loose.

Even if the bolts could be locked the stress on the valve gear would be unacceptable.

As a interim measure I have optimised the length of the reach rod to enable all three forward notches to be used without die block clashes.  To prevent damage to the valve gear I have Loctited a small cube of aluminum to prevent selection of full reverse gear.  I can now run the engine for extended period without problems.

I was curious to see what effect this compromise would have one the valve gear timing so I measured it using the same method I had before.  Perhaps I have written a clearer account for the POP timing here.

This diagram shows the side valve position as a function of piston position (percentage of full stroke).  0% is with the piston at the crank end and 100% the other end, 96mm away.  The valve position is in mm and I have ruled lines at +/- 4.8mm representing the valve lap.

The red ellipses are the three available forward gears and the two cyan  ellipses are the remaining two reverse gears.  In forward gear cutoff occurs at  approximately 25%, 50% and 70% and in reverse at 45% and 70%.  The exact values can be read off the diagram where each ellipse cross the lap line.  Also note there is a small variation depending on the direction of piston travel.  The yellow line is locus for mid gear.

The results show excellent performance in forward gears with a useful range of cutoffs available.  In reverse despite the blanked off notch performance is respectable.  Given that many fully size engines have fewer notches in reverse the final solution may well be to abandon the 3rd reverse slot permanently.

Wednesday 5 October: When I previously ran this engine I did this without pressurising the boiler.  Will most of the fittings fittings supplied I decided it was time to pressurise the  boiler.  Rather than connect the compressor to the cylinder I made an adaptor so I could connect it to one of the clack valves.  I did this to avoid blowing more muck into the boiler.

With only about 10psi on the clock I was disappointed to hear a loud hissing noise coming from below the crank.  Tracking down the exact source was difficult but the "soapy water test" implicated the elbow on the right hand side of the boiler.

Access was difficult so I decided that the crank would have to come off.  I had been dreading removing the crank but in the end it came of quite easily complete with eccentrics and gears. 

I removed the pipe & elbow and carefully refitted it but it was no better.  With improved access I could feel a draft coming from the pipe.  I thought that somehow I had punctured the pipe but on removal for testing in the time honored manner I discovered that it was leaking from one of the fittings.

In the picture below left a small crack can be seen between the pipe and the ferrule.  When I removed the ferrule the failure of the silver solider to penetrate can clearly be seen in the picture below right. 
 

Repairing the pipe was simple but now I have to repaint it (why didn't I decide to leave them copper) and refit the crank.

Following a tip off from Mike I checked the cocks in the water gauge for burrs.  On disassembly I did indeed find small burrs left over from drilling the passage.  I removed these but unfortunately on inspecting the valve seats I could already see that they were scored.  Currently I do not have the gauge glass fitted and leakage around these cocks is my biggest remaining leak.

Saturday: Removing the burrs from the gauge cocks made them worse.  I presume that the burrs were actually helping to form a seal in the scored valve seat.  The original seat was made from a PTFE plug screwed into the brass body.  I like a challenge, had time to kill and some PTFE bar. 

My first act was to make a tapered D bit to cut the new seat.  Having successful completed a dummy run on a scarp of PTFE I grouched out the old seat.  I then plugged the hole with PTFE and bored the new seat in situ.  I then re-chucked the valve body so that I could re-drill the passage.  I finally fine tuned the depth of the seat using the D bit in the pillar drill.  Don't attempt this at home; ModelWorks are asking customers to return any faulty water gauges for repair.

While checking the installation of the gauge I found that when the upper gauge cock was in its normally open position the operating lever hit the 2nd axle tube.  Visually inspecting the passage through the valve I could see that the valve was not completely open (I had to remove the valve to see this).  The effect was made worse, in my case, because the passage was not drilled at right angle to the handle.  Since I was concerned that this restriction could affect the operation of the gauge I put a dogleg bend in the operating level such that when it was resting on the axle tube the valve was completely open.

In addition to a drain valve the gauge has valves in the upper and lower connections to the boiler.  Prior to do some research I had assumed that these valve were there in case of gauge glass failure - wrong - they are there for testing correct gauge operation.  It is possible over time for the bores connecting the gauge to the boiler to become blocked  leading to false indications. 

Wednesday 12th October:  I have just learnt that Modelworks are recalling the affected water gauges so that they can replace the valve seats.  Having replaced the seats myself I know this would be beyond many model engineers.  My advice is to take up ModelWorks generous offer. 

A thought: I am now running the Burrell with the boiler in circuit and the power of the engine is awesome.  Whenever the boiler is pressurised treat the engine with respect - it could leap into life without notice.  Make sure hands are clear and it is out of gear.  I learnt my lesson the hard way with my Bodmers Engine.  It took me two years to build and I had grown accustomed to resting my hand 'in the works'.  The day came to set the valve gear up which has to be done live. 2 1/2" dia piston, 60 psi, crank, thumb - ouch.  My rule now is to release all pressure before putting hands into the works.  I find the blower a convenient method for releasing pressure.