Locomotive valve events

Daft question time...
What causes a locomotive to sound "off-beat"?
A steam locomotive has its valves set by shimming the valve heads up and tightening adjusting nuts up, so that the valve heads can't move relative to the valve rods on a piston valve loco. I'm guessing on a slide valve loco it's more or less the same as in the valve face is shimmed so that the length of the valve rod & valve slide is set correctly...

So, apart from broken valve rings on a PV loco, what causes some locos to sound really off-beat? Is it mainly down to wear in the pin joints on the valvegear? Bent rods will contribute, but it's got me thinking...
Listening to some of my SoundTracks CDs (Ex-Audicord), and there's some wonderfully off-beat sounds, as well as plenty of locos in fine fettle...

Richard.

 

Sometimes its intrinsic. A Duchess, AIUI, will never be as even as a King, because of the design of the rockers and linkages that transfer between inside and outside cylinders.

 

Can I refer you to my posting no. 22 here which should answer your question.

https://www.national-preservation.c...n-and-other-cylinder-questions.1417686/page-2

In both cases, at one end of the cylinder the valve uncovered the steam chest port earlier and for a greater distance than the other end. Putting it another way, one end received more steam than the other resulting in a loud exhaust from one end of the stroke and a soft one at the other.

 

Thank You! Will have a look at this. I am wondering also how locomotives become off-beat whilst in service too, as some of the extreme examples can't have been turned out of a works' overhaul in that state, surely?

Richard.

 

The effect on exhaust beats caused by the difference between the front and back of the cylinders would lead to two pairs of different exhaust beats. I have never noticed this difference. I think that this is because the human ear cannot detect what is a relatively small difference. What is usually noticed is one strong beat. I suspect that locomotives rarely came out of the works in that condition. It is almost certainly caused by wear, damaged valve rings liners or valves or incorrect assembly after examination. There are also other possibilities, such as a return crank moving.

Sent from my SM-A105FN using Tapatalk

 

I can assure you that in the cases of both locos referred to above in post 3, one could definitely hear the different loudness of the off-beat cylinders.

 

The ear will notice two things. The first is how loud each exhaust beat is; the second is the gap between each exhaust beat. UK locos seldom have tail rods so the volume of steam admitted to the front and back of the cylinder will differ slightly resulting in two slightly louder exhaust beat every revolution for two cylinder locos. The valve timing is manually set so is subject to human error. However, once built there is little a valve setter can do other than equalise port openings at each end of stroke, as has recently been discussed on another thread. Even slight errors in the manufacture of the valve gear can affect valve events.

 

Sorry, Steve, but would tail rods influence the volume of steam admitted? surely that's down to the time and length of the port opening; all the tail rod does is reduce the piston area on which that volume of steam acts.

 

One thing which is over looked is the affect of the eccentric rod. If the loco has axel boxes off centre to the original position say after replacement then that affects how long the eccentric rod needs to be. If to short or long the maximum opening at mid stroke becomes un equal and thus also affects the overall valve timing. With this fault locos have a tendency to stall and hard to start.

 

To which I would add that the reduction in area is very small. I don't know what the diameter of a tail rod would be, but say 3 inches. With a cylinder diameter of, say, 18 inches, that makes the rod one sixth of the diameter, so one thirty-sixth of the area, less than 3%. Even if that did produce a corresponding difference in loudness of the exhaust beat, it would be much less than 1 dB, quite inaudible.

 

I have used the Allan Wallace simulation to help me design the valve gear for my 5" loco. I spent hours tweaking various dimensions I an attempt to get the "best" valve events. There is a facility in this simulation to add backlash which simulates wear. You then find that the benefits of the fine tuning are are lost as the valve events change. The horizontal axle box position will certainly have a significant effect but the vertical axle box position will also change events to a smaller extent.

I suppose the moral is to try to get as close to "perfection" as you can when the loco is built or overhauled then things will not be too bad after wear and slack has developed.

I think the Caprotti valve gear gets round most of the problems of Walschaerts or Stephensons but no doubt adds a whole load of different ones



Sent from my SM-A105FN using Tapatalk

 

Is another point the effect of heating up of the valve spindle? That will cause it to expand; since the rear end is essentially "fixed" but the front end is "free", that will cause the valve head to creep forward relative to the ports as the valve spindle warms up and expands. For a piston valve engine with inside admission, that would tend to advance the shut off of the exhaust valve and opening of the inlet valve at the front of the cylinder, and have the opposite (but lesser) effect at the back of the cylinder.

You can compensate for it in the design but it is still another variation, not least between a loco starting out and one that has perhaps been running long enough for everything to warm through properly.

Tom

 

I can't remember where I read it, possibly John Powell, but that came up with a batch of brand new Black Fives built at Horwich Works. They rode very roughly, so roughly as to generate crew complaints, and the investigation found that the valves had been set cold, but as shown in the drawings. This represented the valves when hot, but no allowance for expansion had been made on assembly.

 

It is small but it is there, as is angularity - the point at which the connecting rod and crank are perpendicular - and therefore delivering the peak torque - occurs at a point where the piston is slightly forward of the cylinder's mid-point. Neither of these effect the valve events, though they might have a small influence on the exhaust pressure to the blast pipe.

 

I think what Steve means is that the empty volume of the cylinder on the piston rod side is going to be slightly less than the other, due to the presence of the piston rod. Admittedly, it's only a few cubic inches, but there will still be a difference in volume.

 

No, not really. Steam is a gas and therefore has an indefinite volume: it's volume will always be equal to the space in the cylinder with or without tail rods, and at every point of the stroke. The pressure will reduce as the stroke extends and in proportion (leaving out changes in temperature) to the increased volume. I took it that Steve meant the mass of steam, not the same thing.

 

I understood the original question to be if the valve gear was set up properly at overhaul, and assuming that generally meant that there was a fairly even exhaust chuff in that condition, what would cause it to go offbeat, rather than a general query about the limitations and issues in converting a rotational motion to linear etc which has been covered at length elsewhere. Which is a good question e.g. was wear in the joints of the Gresley gear responsible for the exaggerated syncopation of V2s, or was this also to do with shifted valve heads, broken valve rings, severe carbonisation blocking passages etc? Did return cranks shift as someone suggested? I would have thought that the eccentric rod is likely to bend first.

 

This is more what I was getting at, Thank You for communicating my question more clearly! I'm wondering if some of the recordings I have feature bent eccentric rods, combination levers, etc. There's a recording of a Black Stanier which is horribly off-beat, and I can't imagine it having been outshopped in anywhere near that state. To me, something must have happened to it in service. I can't imagine it being a shifted return crank, as it's a 4-stud job on the LMS. The crankpin will be keyed into the wheel centre too...

Thank You all for your answers Gents, all very much appreciated!

Richard.

 

If, and I repeat if, an axle box is off centre, then the adjacent ones would need to be off centre by the same amount otherwise it is doubtful that the coupling rods could be fitted on.

If off centre (in reality, doubtful to any great extent) it would then be a case with Stephenson's gear of adjusting the valve to get near equal port openings.


However, with Walschaerts gear this would need altering of the length of the eccentric rod to compensate as there would be a difference in the dimension between the axle and the pivot of the expansion link.

In reality, the axle centres would be pretty close to the intended original dimensions.

 

With these LMS return cranks they are not only held on with 4 studs, but there is a tongue on the back of the return crank that engages in a machined slot across the face of the crankpin as well. The studs and nuts merely hold things together, so it is a pretty rigid and positive arrangement.