The Final Bits

ModelWorks 4 inch Scale Burrell Agricultural Traction Engine
The safety valve and other bits to complete the Burrell arrive Tuesday 10th January 2006

Tuesday 10th January 2006: The arrival of the of the safety valve today is a bit of an anticlimax as I am in no hurry to steam the Burrell.  For once I will not be amongst the first with news so if you have steamed you Burrell, successful or otherwise,  please let me know and I will share you experiences on the web.

send report to:

The main reason for my delay in steaming is that I am waiting for the warmer weather.  It also seems silly to get the boiler wet in the middle of the winter when the normal practice would be to lay it up dry for the winter.  I am pacing my self so that I can get a few steaming under my belt before a boiler test and then straight into the rally season.

As well as the safety valve the kit included the makings of a leather belt for the governor, the safety glass for the water gauge, the seat's base plate and a mystery grub screw.  The note with the grub is not much use as no instructions were included.  I have just heard from Mike and this shipment did not included the gauge cocks for those customers who returned them for repair.  Apparently they are waiting but need to be claimed!

While it would be tempting to polish the safety valves I have had a good look at my full-size picture archive and black paint appears to be the norm.  Many miniatures leave the top section of the cap bare and polished but I can't find  any example of this in full-size.  

Thursday: I have now painted  the safety valve with high heat gloss.  This was made more difficult as it was not appropriate to dismantle it as ModelWorks supplied it preset and tested.  Yesterday I carefully masked of the operating lever, ports and spring and gave it a coat of etch primer.  Today I spayed it with the high heat paint and am now waiting for it to dry.  I also painted the seat's base plate.

Debbie answered my grub screw query; apparently the grub screw is to replace the one in the regulator nut as it is believed to foul the bottom of the safety valve.  I have had a quick look and my grub screw does not come up much past the top of the nut.

Since the paint is still wet I have not been able to try the safety valve in position but I suspect it will clear the bush on the underside of the valve.  I wonder if the control model had an extra long grub screw fitted?

Friday:  Now that I have the gauge glass I could sort out the length of the protector.   I had previously counter bored the top plate to allow it to clear the top nut but I was still having difficulty fitting.  Today I ground about 25 thou from each piece of glass and shortened the rods by a similar amount.  This has made fitting easier.

I think the cause of the difficulty is the variability of hornplate  alignment relative to the boiler discussed in kit 12.  The top cock is secured to the 2nd shaft which takes it datum from the hornplate whereas the lower cock is fitted directly to the boiler.  As my hornplates are lower that the nominal position I have a slightly shorter gauge glass.  There could well be someone other there who finds the gauge glass protector too short!

I have just tried the safety valve in position and the original grub screw does not foul the bottom of the safety valve.  I am puzzled as the replacement screw was slightly longer!  I shan't worry any more about a problem that I haven't got.

I probably won't permanently fit the governor belt which is just as well as it is slightly too wide.

I could finally fit and try out the seat.  The foot rest would be in just the right position except that the winch rollers get in the way of my leg.  I found it more comfortable to sit with either one or both feet inside the tender but I suspect this will not be appropriate once steam is raised.

Over the next few day I plan to search through my kit residues and fit all the remaining items.

The first job I tackled was the fitting of the cylinder drain pipes.  These pipes were supplied unbent and it took me a while to get the bent to an correct shape.  The pipes are held in place against the cleading using a clamp secured to a plate on inside of the cleading.  To avoid any problems I had previously Loctited this plate in position prior to fitting the cleading.  This worked well and I was able to fit the pipes with ease.

Monday:  Over the weekend I trawled through all my kit residues and fitted the few remaining parts. 

I have been troubled by the operation of the regulator since fitting it in kit 20. - It needs to be pull backwards to go.   Whilst at Dorset I snapped Burrell Road Locomotive YA 366 which had a regulator similar to our Burrell.  Note: shut  is fully back

I believe this is more intuitive and wanted to follow suit.  This is easy to accomplish by changing the position of the side valve on the regulator operating rod but first I needed to file a new notch.   I took the opportunity to make new notch bigger as I have been finding difficulty setting a slow idling speed.  

This afternoon I struggled to find anything else to do and so I must conclude that the Burrell is finished.  To celebrate the occasion I took a few photos inside the garage.  Here is another view at high resolution 1600x1200. 

I don't intend to do any more until the spring when I will give the tender and boiler a good wash out, test the axle pump and then steam. 
 
Monday 30: Even though I have now got the original lubricator working I am nervous about the long term reliability  of the roller clutches.  The POP lubricator was temperamental until I discovered the need to prevent the build up of "oil gunge" inside the clutches.  I found the quickest way to dislodge this was a quick blast of WD40. Whilst at the Ally Pally exhibition I purchased a new lubricator from the stand of RA Barker.   The new lubricator is slightly bigger and will not fit on the existing mounting bracket.  While I could have extended the existing bracket I decided to machine a new bracket from BMS; shown part completed in foreground.

The order of the lubricator pump components as delivered

Since I want the maximum oil flow for the first few steamings I reassembled the pump with both spacers between the o-rings.

The new bracket worked very well and I didn't even need to adjust the position of the delivery pipe.  At the suggestion of David P I have placed a fiber gasket between the cylinder and the bracket to reduce heat transfer that might otherwise thin the oil reducing the efficiency of the pump.

I am not totally happy with the geometry of the linkage and this will need to be addressed.

Saturday: I had thought that I would have the extend the arm from the valve gear upwards so that I could get the intermediate linkage between the valve gear and the lubricator arm horizontal but after a lot of musing I decided that it did not matter if this linkage was at an angle.

In the sketch right I show the position of the linkage at each end and in the centre of the stroke.  Provided the linkage is set so that the lubricator arm swings equally about vertical the swing of the arm will be the same as achieved with the intermediate linkage horizontal.  (Since the intermediate linkage at the beginning and end of the stroke is parallel the horizontal distance moved by its ends are the same) I have exaggerated the geometry so that the effect is clear.  The case shown is not practical as at the right had end of the stroke the lubricator arm and linkage nearly come in line so putting excessive strain on the mechanism.

The grass is always greener on the other side
On checking out the function of the new lubricator I discovered that is was tricky to get sufficient motion when fully notched up to click over one ratchet.  I can now see the advantage of the original lubricator's  roller clutches which, having no teeth, are capable of working with a smaller stoke.  

I have recently learned that ModelWorks have updated the design of their lubricators to use 4mm roller clutches in order to ensure reliable operation.  Since the Burrell's lubricator is to this improved design I might have been hasty buying a replacement.

Thursday 9th February:  Last week I received a letter from ModelWorks explaining that constructors would be receiving new pistons as there was a concern that the rings could get trapped in the ports.  I suspect this was the source of Arthur's trouble when he steamed his Burrell.  While I have have not yet received my replacement piston, Mike has, and it is worth sharing what he found when he removed the old piston for replacement.

The picture speaks for itself.  Fortunately for Mike, damage was confined to the piston and ring.  It is particularly fortunate that the cylinder bore was unmarked.

Given the seriousness of the problem the Burrell should not be steamed or run on air until the piston has been replaced.  It is reassuring to know that ModelWorks have taken a proactive approach and have already ship replacement pistons - don't put this one off - fit the replacement piston!  Mine has not yet turned up either it has been lost in transit or ModelWorks are shipping them as they are produced.

The problem arises because the length of stroke of the piston causes the piston rings to pass over the ports at the side of the cylinder.  If the end of a ring get trapped in the port slot it is shattered when the piston starts its return stroke. 

The solution is the change the piston to one with only one centrally positioned ring.  This avoids the ring ever overlapping either port.  Since the Burrell's stroke is 97mm and there is 103.3mm of unobstructed cylinder between the ports the centre 6.7mm of the piston is never exposed to the port.

Friday 17 Feb: I have been musing on whether the piston trouble is indicative of a more serious problem.  In sort the answer is no but I will also give the long winded explanation.  It is clear that the piston will partly cover the steam ports at the end of stroke and momentarily impede steam flow.  This lead me to check my Live Stem Models 4" Burrell drawing where I found very similar geometry.  In this design the port is eclipsed by 7/32" at the end of each stroke. Clearly partial port covering at the ends of stroke is nothing to worry about.  To avoid the piston covering the port either the ports would have to be made much narrower or the cylinder bore lengthen  thereby increasing the dead volume at the end of stroke.  Both there two option would have a much more serious impart on performance than the original "problem". 

The Cluputs are the culprits
The ModelWorks Burrell is fitted will Cluput piston rings and I believe this improvement triggered the problem.  The Cluput rings consist of two ring joined together.  This means they are relatively fat but more significantly have and addition edge.

As soon as the first half of the ring goes over the port it is liable to spring out and become trapped.  This is particularly likely if the gap has rotated to be over the port.  In the diagram on the right have have tried to illustrate how this reduces the "safe" stroke of the piston.  In both cases the margin is small and there is the addition complication that the boiler will stretch when hot moving the piston towards the crank.

Today I fitted the new piston which has a single centrally mounted Cluput ring.  As instructed I  removed the old piston by removing the front cylinder cover.  Curiously I found the piston rod easy to unscrew from the cross head.  The reason turned out to be that I had never tightened the locking grub screws.  We we told to leave these lose when fitted during kit 11 and I can find no instruction requiring them to be tightened.  While it is obvious that they should be tighten I could find no guidance on setting the piston rod length.  An overlong or short piston rod will exacerbate the problem with the piston rings entering the ports which perhaps explains why some constructors have been badly affected.

I had expected to have great difficulty removing the piston rings and transferring one to the new piston.  In the end I found it easy to unwind them and refit.

What I was not expecting was the trouble I had getting the old piston off the rod.  The lock nut was easily removed but even with the rod held in the vise and the piston gripped with a lathe chuck I could not get it to budge.  As I suspected Loctite I gently heated the piston with my blowtorch.  I guess I took the temperature  to around 160 C and at this temperature it unscrewed easily.

When I reassembled the engine I went to great pains to set the piston rod length so that the piston sweep was central between the ports.  Having found this position I then slightly  lengthened the rod to allow for expansion of the boiler and did up the grub screws to lock the piston rod in this position.   Once I had refitted the end cover I ran the engine on air and I was rewarded with a nice even chuff.  

Thursday 9 March: I plan to steam the Burrell in the next few days and it preparation I have been giving the engine a through check over.

In correspondence with Steve Traill I have been alerted to problems caused by the loosening of the hornplate/boiler bolts.  The boiler is secured using a M8 mushroom headed bolt in a 10 mm clearance hole.   

In an attempt to resist sideways motion I made brass bushes to reduce the clearance between bolt and hornplate.  These bushes were custom made with the hole offset to suit the alignment at each hole.  Before tightening the bolts I removed the paint from the side of the hornplate. 
 
In order to get at the bolts it was necessary to remove the rear wheels and gear guards.  On seeing the differential exposed I was inspired to have a go at getting the diff lock working properly.  When I visited Steve Traill I was surprised how freely his lock operated and he explained that he had straightened the operating horse shoe.  I removed mine and checked it on my surface plate.  Sure enough there was a slight  twist that was causing a fore aft twist of the locking ring.  I tried straighten this using brute force but this caused the brazing to give way so I was forced to get out the brazing torch.

Fitting the upper gear guard is made very fiddly by the need to fit 4 nuts in the inaccessible recess between the hornplate and the gears.   To overcome this I decided to make "nut bolts".  In essence these are M7 bolts with a M5 tapped hole through the centre.  I drilled out the hornplate and tender fixing holes, tapped them M7 and then screwed the "nut bolts" in place. 

The nut bolts are shown in close up on the right and may be seen fitted to the hornplate in the shot above.

These changes made refitting the upper gear guard very easy.  When I retested the diff lock it slid in very easily in sharp contrast to the extreme force needed before.

When it came to fit the reach road I noticed that the bolts joining the two parts had come loose again!  Since it is difficult to get a good grip on the M4 bolts I decided to replace these with M4 caps.  If they come loose again I will be getting the MIG welder out.

14/4/06: They did and out came the welder.

Saturday 11 March: I have been contacted again by Steve Traill who has been working with Steve Baldock to iron out the problems with the injector.  It has transpired that there is a design issue with the clack valves.  There is excessive lift in the boiler feed clack valves which can allow the ball to move such that the feed to the boiler is blocked.  Not only does this prevent the injector working it can place excessive strain on the crank pump.  It is therefore advisable not to steam until this issue has been resolved.  

I removed the feed pipe from the injector's clack giving me access through the clack to the base of the ball.  Using a nut and bolt as a depth gauge I was able to determine the ball lift.  I then removed the top plug and by reinserting the bolt was able to see how far the ball can lift before hitting the stop.  As the picture on the right shows the ball is lifted high enough to block the outlet to the boiler. 

Steve Baldock is aware of the problem and I am sure  Modelworks will be issuing a fix in due course but as I am keen to steam I decided to have a go at my own fix.

The obvious solution is extend the stop on the top plug.  Rather than make new plugs I drilled out the existing stop and threaded in a length of 1/4" brass.  I then turned this to finished dimensions giving a lift of 1mm for the crank pump and 2mm for the injector.  Less lift on the crank pump's clack is desirable as this reduces the tendency of the ball to bounce between strokes.

The picture on the right shows the before and after for the injectors stop.

I have never filled the tender with water and in preparation for steaming I filled this to the brim.  This also gave me a chance to check out the crank pump and flush the injector's pipes.

Sunday:  I checked the weather forecast for tomorrow and as it is sunny I will steam tomorrow.  I spent the day doing final checks and preparations.  

I had not applied any gasket sealer to the valve chest cover so this was one of the jobs I tackled.  When removed the cover I was engulfed in tarry steam oil which had accumulated at the bottom of the chest.  Without the heat from the steam it was not all reaching the cylinder.  Had I not removed it I am sure in the first few stokes on steam it would have been all over my drive.