Kit 21 - Steam Head & Water Pump

ModelWorks 4 inch Scale Burrell Agricultural Traction Engine
Kit 21 was eventually delivered on Friday 19th August.
 

Friday: For some reason I noticed that there was no 90 degree elbow SP59031 and this inspired me to carefully check the contents against the packing list.  In addition to the pressure gauge pipe mentioned in the covering note I found that the injector to check valve pipe BU42155 and water lifter pipe BU21159 to be missing.  After a tip off from Mike I checked the fasteners and discovered that the fasteners for the Steam head BU42151 and the Water Pump Gland BU42134 to be missing.

I have spent the evening trying to fit the crankshaft gears.  I cleaned up the spines in the gears using a fine file which exactly fitted in the slots.  I was then able to get the larger gear  to fit.  I was not so fortunate with the small gear.  It jammed when it was only partway on the shaft.  I could not find where it was binding and so I assumed it was on the splines. After further fettling of the slots there was no improvement.  I searched elsewhere for the problem and I began to suspect that the bore in the gear was slightly too small for the shaft.

I measured the internal bore of the small gear and came to the conclusion that it was slightly undersize at one end. I decided to see if I could get the bore of the gear to run true in the lathe.  I mounted the gear in the four jaw and centred it using a dial gauge.  I then coated the inside of the bore with felt tip and advanced the boring tool to the bore and various points along its length.  Inspecting the rub marks in the felt tip confirmed that I had indeed got the bore running true.  I was surprised I accomplished this so easily.  I then took one extremely light cut (1or 2 thou) and removed the gear to try on the crank. 

Its fit was better but not perfect however I decided to call it a day as it freely moved over the outer portion where it needed to operate.

With hindsight it would have been better to either not fit or remove the splines and check the fit of the gear on the shaft without the complication of the splines.  It would then have been much easer to test the fit of the gear and file/scrap/abrade the offending part to size.

Saturday: The next task was to fit the brake to the differential.   M3 bolts were supplied to fit the lugs to the brake band but no nuts.  I think the lugs were meant to be tapped M3 but I found the bolts just about long enough to use nuts.

Sunday

I also made a start on the steam head but as the fixing screws are missing I did not get very far.

Monday: I found some long bolts which enabled me to temporally fit the steam head (top holes only).  I was pleased to discover that on refitting the spectacle plate that the steam pipe fitted nicely.

Assembly of the water pump was straight forward and it was only when I fitted it to the hornplate that I encountered a problem.  I found that the pump rod was out of alignment by about 1/16 inch on the "vacant eccentric". I thought of bending the pump rod but  realised that since I had already permanently fitted it to the piston there was no satisfactory way of gripping it.

In desperation I tried rotating the pump rod through 180 degrees and to my amasement  found it now aligned perfectly.  On inspecting the connecting rod I could see that the holes were not drilled centrally.   As  I don't know whether this is a lucky accident or by design  I would recommend trial fitting the pump before locking the piston to the connecting rod.  11/9/05 also see below

The were a lot of burs within the check valve casing and I spent some time carefully removing these.  The ball valve seat required special attention as it is important that it has perfect edges.  If found that passing a 1/4" reamer in from the top did the trick.  I achieved nicely defined clean edges.

I have made a start fitting the pipes.  Unlike the POP Modelworks have supplied the pipes pre-bent.  It took a while to figure out how to fit the injector steam pipe BU42154 as the ISO is "taken" from a strange angle that makes the vertical portions disappear.  BU42154 has been supplied in two parts to enable it to be threaded into position without dismantling the Burrell.  Amazingly it fitted without any rebending. Note: The top section looks like it would fit either way round but I discovered that only one way was "perfect".  I am not having so much luck getting the pipes to line up with the pipe supports.  The injector water feed pipe BU42156 is a tad too short.  Rather than put a joggle in the steam pipe I plan pack the joint with the water valve to "lengthen" the feed pipe.   The Pump to check valve pipe BU42158 completely misses one of its supports.  Since I am missing the elbow that connects it to the pump I will not attempt to adjust the pipe until it is fixed at both ends. 

Tuesday: When I tried to fit the water feed pipes to the pump I found both to be 1/4" too long.  Rather than attempting to lose the extra by bending the pipes I cheated and shorten them by 1/4".  Mike has managed to fit these pipes but he had to anneal them in order to get them sufficiently pliable to bend.

Without the missing items from Modelworks this kit is grinding to a halt.  The only job available is to prepare the pressure gauge for fitting to the spectacle plate.

STOP PRESS I have just heard from Steve that there will be a mail out of the missing fasteners tomorrow.  The long pipes are to follow with the next kit - it seems they did not fit in the box!

Thursday: Today I received the missing fittings from Modelworks.  This enabled me to fit the elbow to the water pump and check the fit of the final pipe.  While it lined up with the water pump and check valve it completely missed one of its supports.   Bending the pipe to pass the support while still arriving at the pump would require the top bend to be moved down the pipe by about two inches.  I tried straightening top bend but it didn't feel like it wanted to go so I decided to anneal it.  I heated the portion of the pipe from the old bend down to the position of the new bend to a dull red using my brazing torch and left it to cool  (cooling rate is immaterial, the important part of annealing copper is to heat to dull red).  The change in the pliability of the copper was dramatic and I had to be very carefully when applying even moderate force not to wreck the pipe.  Care is also needed when manipulating the un-annealed sections of pipe that their stiffness does not cause the softened positions of pipe to bend unintentionally.  All this said, it did not take me long to get the pipe to fit; now all I need to do it to find a way of hardening it again. 

Before and After

I have been kept occupied these last few days preparing the many brass parts for painting.  

Looking at my picture collection I could not establish a consensus as the which fittings would normally be painted.  In the end I decided to paint virtually everything as it is relatively easy to remove the paint from any that I later decide should be left unpainted. 

Friday: I finally got round to etch priming after spending 3 days on preparation.

Saturday:  I painted everything gloss black using a high heat paint.

Sunday: Fortunately the paint was sufficiently dry to commence assembly.  I started with the brake as this had the prospect of enabling a lot of left overs from previous kits to be permanently fitted.

Modelworks have recalled the brake shaft and screw so that they can silver-solder the two parts together.  Not wanting to be held up I did this myself.  I experimented with the cork and found that using both the thick and thin band together did not allow the brake to be released.  As a result I just used the new think cork however the brake bites close to the limit of its travel and may spell trouble later on.

My next task was to permanently fit the rear spectacle plate.  The four holes at the bottom needed to be deeply counter sunk and the steam head fitted before fitting to the Burrell.  I had to substitute longer M3 screws when fitting the pressure gauge.  

It would have been better to have fitted the spectacle plate before fitting the brake thereby enabling the differential to be slid out. This made access to some of the bolts difficult.  22/9/05  Consider fitting water gauge if available. See

The part I most wanted to fit was the upper gear guard as this would enable me to get the valve gear going again.  This was very fiddly and it is a shame that ModelWorks  did not tap the hornplate holes M5 thereby dispensing with the need to fit nuts.  I also had to substitute some of the fixing with longer bolts.  I was also able to check out the new reversing quadrant and was able to get all notches to engage without any problems with the die block hitting the end of the expansion link. (11/9/05) I am not confident that this statement is correct.  When I attempted to set the valve gear up correctly I again experienced clashes.  (30/9/05) I now know I have a problem with the new quadrant which is described here.

Monday

I have seen many model traction engines where the oil reservoir is remote from the axle and I decided to do the same.  It was a easy job to turn up two adaptors to enable the a section of 1/8" diameter copper pipe to be inserted between the oiler and the shaft.

Tuesday:  When I came to fit the injector's steam feed pipe I discovered that it clashed with the reach rod.  When I previously checked the fit of this pipe the problem did not show as the valve gear was not fitted. Now that I had painted the pipe I discovered it did not fit!

To clear the reach rod the bend would need to be moved about 1/2" closer to the globe valve.  This was not going to be easy as the bend had to start as soon as it leaves the fitting.

As I would be bending the pipe by hand I annealed the top of the pipe .  Had I had suitable bending formers I would have preferred to have left it hard as I had great difficulty in preventing it collapsing.  

I have now got the pipe to fit but it will have to be repainted.

Saturday 10th September: I have beefed up my compressor and have been running on air again.  The method used to plug the holes and connect up the compressor is described here.  As a result of having more air (15cfm) the engine runs faster but this has highlighted some problems.  After a few minutes I noticed that the lower valve eccentric was getting quite hot.  I stopped the engine to investigate.  I found that the eccentric rotated freely except for one or two slightly tight spots.  I introduced more packing into the eccentric strap but even with a total of 30 thou of packing the eccentric was still running hot.  Inspecting the inside bearing surface of the strap I could see a band of what look like rubbing but on closer inspection this turned out to be a raised ridge.   I had wondered whether prolonged running would fix the problem but Steve advised me that if a bearing is running hot it will eventually seize.  Since add packing afforded no relief I decided to take the drastic step of using valve grinding paste.  I reassembled the eccentric with some fine grinding paste inside and gave the engine a short run.   I then spent the next half hour scrupulously clearing the eccentric to remove all traces of the paste.   On reassembly I could immediately see an improvement as the engine ran much faster.

After another extended run I observed that while the valve eccentrics were running cold the pump eccentric was getting warm.   I checked out the eccentric and found that it rotated freely.  Baffled I tried running the engine without the pump eccentric and found that it ran even faster.  This proved there was a problem with the pump eccentric but where was it?  I  tried shimming the strap to no avail and when the valve lapping paste did not improve matters I was ruining out of ideas. 

Careful inspection of the eccentric strap showed signs of polishing on the sides which led me to wonder whether the strap was being twisted.  The only way to be sure was to remove to pump casing so that I could check the natural position of the pump ram - was the pump bore pulling the ram sideways?  In order to make a precise measurement I found just the right thickness of packing to just fit between the ram an the hornplates.  I ended up with a 1" block and two 20 thou shims (total 1.040").  On my surface plate I measured the equivalent dimension on the pump at 1.100". No wonder it was rubbing it was being pulled sideways by 1/16 ". 

Live by the sword die by the sword
This problem is largely of my own making.  In order to get more room for the eccentrics I machined a slice off the governor pulley.  This has allowed the pump eccentric to be fitted closer to the hornplate.  Unfortunately the room has now been used up and without major surgery I cannot simply move the eccentric away from the hornplate.  The correct course of action is probably to put a crank in the pump rod but I chose a more drastic solution!  I would shave 60 thou off the foot of the pump thereby moving it closer to the hornplate.  I did this with the aid of my mill plus fly cutter - I didn't even scratch the paint.

On reassembly the eccentrics ran cold but the engine did not run quite as fast due to the additional load of the pump.  Turning the engine over by hand was now quite easy.  The only significant resistance coming from the piston as it pushed air through the cylinder.  The engine has lost it's 'bark' as the exhaust air is coming out at a much lower pressure.  While this does not sound as dramatic it is good news as it demonstrates a significant reduction in friction.  The engine now only barks when accelerating the flywheel up to speed.