Third Steaming
Tuesday 4 April - Success Day

ModelWorks 4 inch Scale Burrell Agricultural Traction Engine 

Tuesday: I hope to take the Burrell for it's steam test on Sunday and today's exercise was to prove that everything is ready.  

Transport to the test site is essential so I took the Burrell for a short test drive.  I am relieved to say that the trailer  and load behaved itself.  I got a bit too relaxed and took some speed bumps a little too fast and the trailer went airborne (speed about 20mph).   On the plus side this confirmed the lashing and suspension of the trailer.   

I used to own a house with open fires and a coal fired Rayburn.  When this is your main form of heating you get quite skilled at making and maintain fires.  There is much hot air expended on how best to light a Rayburn but the truth is that it is best to adopt a style that works for you and your Rayburn.  Previous when firing the Burrell I tried to slavishly follow other peoples advice.  While this was relatively successful for this steaming I decided to follow my own instincts.  After all the Burrell's fire box is about the same size as my old Rayburn.

A friend has found an excellent source of firewood in the form of rough wooden crates thrown out by a local firm. This timber is dry and appears to be untreated.  Since this has always been my preferred method of ignition I decided to dispense with the barbeque charcoal and use wood instead.   I never used Paraffin in my Rayburn and as I believe this is responsible for much of the smelly smoke I used the wood dry.  I did however dip one or two small pieces of in paraffin and after lighting these on the shovel I dropped them onto the kindling already in the fire box.  I added a few more pieces of dry kindling closed the firedoors and waited.  

I could see through the crack in the firedoors that the wood was catching and I was pleased by the reduced quantity of smoke issuing from the chimney.  What's more it was less smelly.

The fire was more controlled and did not burst out of the firedoors when they were opened to add more wood.  Over the next hour I slowly built up the fire adding smokeless coal every five minutes or so.

After about Ninety Minutes the safeties lifted.  At  120 psi they were only hissing but by 121psi a plume of steam was reaching up to the top of our four story house.  There pressure did not rise any further.  As the pressure dropped the safeties required a gentle tap to reseat and stop the flow of steam. 

Having used up some water it was time to test the two methods water feed needed for the test.  I closed the pump bypass and I am pleased to report that it worked.  PTFE appears to be a good replacement for the O-rings but I did have to shorten the pump gland to enable the 3 C's of PTFE to fit comfortably.

After confirming that water was going up the glass I opened the pump bypass and tried the injector.  Since my last steaming I have removed the filter and this has made it easier to get the injector to pick up.  Throughout the remainder of the steaming I used both the pump and injector to maintain the water level with total success.  

I was pleased to find that the fire was much more controlled and I suspect this was in part due to not using Barbeque charcoal.   Previously I used a fair quantity to light the fire and I suspect it maintains its heat for a long time.

Steve Traill has successfully modified  his water lifter.  Since his modification is relatively drastic involving changes to both male and female cones I decided to try a simpler modification.  I made a new female cone with slightly steeper cones.  Unfortunately this did not work and in an attempt to discover the source of the problem I removed the lifter from the tender and set it up to blow water across the garden.  My experiments showed that the lifter only worked when the water source was raised above the lifter.

Steve Traill has explained his modification to me and for my next steaming I will try and follow his lead.

As the day progressed I became concerned by an annoying click coming from the motion.  After a thorough investigation I concluded that it was coming from or near the pump eccentric.  As I could not investigate this further whilst in steam I made a decision to stop adding coal to the fire.

I let the engine stand maintaining the water level with the injector; it did not need much.  After ninety minutes when the fire was no longer showing any life I dropped the firebox.

For the first few steaming ModelWorks have recommended that the boiler is completely blown down after each steaming to rid the boiler of any foreign matter.  I am beginning to wonder if this is necessary as each time only clean water emerges and the engine shows no tendency to prime.

I am very pleased with today's steaming.  I felt in control throughout and am looking forward to my trip to the park on Sunday.  With the fire under control I found I used much less water and the tender float was a great success.   I have some work to do before Sunday, most importantly to  investigate the click from the pump eccentric.  A quick look this evening suggests that the pump eccentric is being twisted by the pump's connecting rod and this is the first thing I will check tomorrow.

Wednesday: When I refitted the pump after fixing leaks I had forgot that the connecting rod fitted one way better than the other as the mounting holes had been drilled slightly off centre.  Rotating the connecting rod improved things but to make sure I removed the pump and elongated the mounting holes slightly.  This enabled all twist to be removed from the eccentric.

Examining the inside bore of the piston I was surprised to see dents caused by the connecting rod hitting it.  Could this have been the source of the click?  To ensure no contact I filed away the connecting rod in the region where it emerges from the inside bore of the pump.

I had a chance to examine the PTFE packing and it looked the business.  I am sure this is the right solution.  I believe I now have all the Burrell glands packed with PTFE C's.

Thursday: Having completed the essential repairs I set about making replacement cones for the water lifter.  Steve Traill had sent me a drawing of a full size Clayton & Shuttleworth water lifter which he used as the basis for his successful replacement.

As I have not yet modified the water lifter's steam cock I decided to scale the full-size lifter down by a factor of 3.5 in the belief that this would make it less steam hungry.  The notable differences are that it is has a steeper cone on the inlet side, the minimum diameter is much less and is much longer.

My first job was to modify the male cone through which the steam is blown.  As supplied it had a tiny hole and I opened this out to 1.4mm.  To smooth the flow of steam I stepped drilled in from the steam pipe end using 1.7mm and 2mm drills.  The full size had a hole size of 5mm with a steady taper out to 12mm for the steam pipe.  In the full size the tip of the male cone was position 25mm away from the female cone but Initially I am leaving it at it current length which means it tip is inline with the start of the female cone.

Before I could make the female cone I had to make tapered D bits in order to cut the cones.  I have made several of these before from Silver Steel but never any this long.   I made the cutters slight longer than necessary so that if necessary I can Increase or decrease the diameters.  The outlet cone is long and has a very shallow taper and it was a bit tricky to mill the flat on the cutter.  Once complete I harden and tempered the cutters though for a one off in brass I might have got away leaving them soft.

Once the cutter were made it was relatively straightforward to make the new cone.  I roughed out the cones by step drilling finishing to size using the tapered D bits.

I reassembled the water lifter and before refitting it decided to see if it would work on compressed air.  I was very pleased to find that it did and it had enough suck to pull  water up the full height of the delivery hose (2M).  I was a bit disappointed with the rate of delivery of water but I imagine that it will be much better on steam. 

Steve Baldock has just contacted me to say that ModelWorks will be supplying replacement cones for the water lifer based on Steve Traill's research.   There will be one or two other goodies and these will be manufactured and dispatched as soon as they can be slotted into the factories busy manufacturing schedule.  

Sunday: Today I took the Burrell to my local club to get it tested.  The testers decided to conduct both a hydraulic and steam tests.   The hydraulic test sprang leaks I never knew I had.  As part of the new test code it is necessary to establish the capacity of the boiler so the boiler was drained down into a measuring jug.  For the record the capacity was determined to be 26 liters. 

Raising steam took exactly one hour from lighting though it felt much longer.  The safety valve was checked first.  After building a very generous fire the blower was turned on until the safety valve lifted.  The distressing part is that the test continues until the testers are satisfied that the pressure will not rise any further.  I was relieved when the blower was closed and the test past.

Next came the water feed arrangements.  The pump worked though it was suggested to me that the pump efficiency was being reduced by bouncing ball valves.  The injector was not on it best behavior needing continual tweaking to keep it working.

With the test successfully completed I was persuaded  to go the long way round the club track to avoid a short steep hill.  This was my first chance to drive any significant distance.  My education started immediately when the engine started to prime.  Either it was me, the motion of the engine or stuff dislodged by the test but the engine primed several times during my circuit. Each time I had to stop and recover the water level.  I used about 50 liters of water to go less than 1 mile.

I tried my modified water lifter and while it worked it was not a efficient as expected.  It took about 30 seconds to empty a bucket of water.

Tuesday: My NTET disc came through today and I must congratulate Walker Midgley on their excellent service.  This will come in handy as I am hoping to go to my first rally this weekend.  With the prospect of a rally this weekend I wanted to get on and attend to the snags uncovered on Sunday. 

It was suggested to me that the crank pump efficiency was being impaired by the bouncing of the balls inside the pump as a result of excessive lift.

I can now remove the pump in my sleep and my first act was to measure the ball lift.  The inlet ball had a lift approaching 1/4" whereas the outlet ball had a lift of around 3/16".  Since these lifts were greatly in excess of the 1/32" recommend to me decided to have a go at reducing them.

It was relatively easy to limit the rise of the outlet ball.  I simply drilled and tapped the pumps top cover and made and screwed in a brass button.  I then trimmed this to give a rise of 1/32". 

The inlet chamber presented a few challenges. The biggest was that it's roof does not present a even surface to act as stop for the ball. My experiments showed that when the ball hit it, the ball was pushed sideways.  In the end I decided to improve this as much as I could using my mini drill.  Making a new inlet bush was straight forward except I had to allow for the fact that the chamber was not tapped to its full depth.

I have now reassembled the pump after checking that the ball lifts were 1/32".  When I next steam it will be interesting to see whether these changes have made any difference.

Wednesday: I am still having trouble with the bolts coming loose on the reverse reach rod and on Steve advice I have decided to weld it up.  I carefully set the length of the reach rod so that the die block did not knock in either forward or reverse settings.  Removing the reach rod whilst full assembled was tricky.  After removal of the injectors steam feed pipe I was able to jiggle the reach rod out backwards.  With the bolts still in place I welded the parts together and then removed the bolts and welded up the holes.  Once I had cleaned up and filed away the excess weld the reach rod looks as if it was made in one piece.