Running in TVR-1

[GreeboTheCat]

Have run-in the TVR-1 for the recommended time (about an hour on air) and it's freed up beautifully but for some reason it seems to run slightly smoother in reverse than forward.  :?:
Anybody got any idea why this should happen.

Greebo

[sandy_ACS]

 

Hi PD's

Greebo....

First off: - well done on getting your engine built up and running.... it gives you a real buzz does it not?

Ok, your question..... I think it is fair to say that 1 hr running (on air) will only have begun the running in process... it will take several more hours (on steam) before all the moving parts are fully bedded down.

The main reasons that an engine runs better in one direction than the other are: -

Bearing tightness/alignment

Sliding, or oscillating components not fully bedded down.

And (probably the biggest contribution) valve gear setup/errors.

The first 2 of these can really be considered as a whole and constitute the need for a running in period in the first place.

Take, for example, the crankshaft main bearing.... it is a well documented fact that when the shaft running in such a bearing is subjected to a sideways (radial) thrust, then the shaft will take up a position, within the bearing, which is not in the centre, but the sahft will be deflected in the direction away from the angle of the thrust..... in other words the shaft becomes eccentric with regard to the bearing.

If you were to study the effects of this on a double acting steam engine, then you would see that the crankshaft favours one half of the bearing for one direction of rotation, and the opposite half for the other direction,
therefore, even minute, bearing miss-alignments throughout the shaft (remember you have at least 3 main bearings) will create more or less bearing drag depending on rotation direction.

The same thing applies to all bearings and sliding surfaces.

Sliding parts would include: - the crossheads, the slide valves and, in your case, the slides for the Hackworth valve gear.

Oscillating parts would include all bearings and pins which do not make a complete revolution, such as the Con-rod small end and other parts of the valve gear mechanism.

It may seem, when you turn the engine by hand, that these are not the cause, however, when you apply air, or steam then the thrusts applied to the various parts will be entirely different to those you can apply by hand turning of the shaft.
In some cases these thrust variables will combine and make rotation somewhat easier, but equally they can combine and create additional drag forces.

Over time, with carefull running in, and with lots of good oil, these effects will gradually reduce and will become a very small part of the issue.

I think though, that the biggest culprit is the valve gear, and it's set-up.

First thing to remember is.... there is no such thing as a perfect valve gear... and 'Hackworth' is no exception, there will always be errors in the valve gear operation due to geometric errors between the moving parts.

If you take just one example: say one of your TVR cylinders.... and assuming the valve gear is in FULL FORWARD position and has been set-up exactly... then you would almost certainly find that the actual valve events for the downward steam cycle would differ slightly from the upward steam cycle.

e.g..... on the downward cycle the top steam port would close at 82.4% of the stroke (82.4% cut-off) however, for the upward steam cycle, the lower steam port may close at 81.9% (81.9% cut-off)... this means an error of just 0.5% over the desired equality.

If you now place the same valve gear in FULL REVERSE... the results could well become 89.0% for the down stroke and 74.8% for the up stroke... error is now 14.2%.

Such errors are not unusual and I am sure you can see that this would indeed create a difference in the apparent smoothness between the 2 directions.

These errors are inherent in the valve gear and even the best designed and set-up gears would inevitably have some errors.... such is the way of things.

Some improvement may be found by making very small adjustments to the valve gear on your engine, it is probably worth while re-examing the set-up after you have run the engine on steam for a couple of hours.


By far the best way is... get it in a model, load it with a real propellor and just poodle around the lake with it..... the effects a true load can have are considerable... and will give entirely different results than just running, unloaded, on the bench.

Hope this helps explain things a bit.
 
Best regards.

Sandy.     :luck :computer

[GreeboTheCat]

Thanks for your comprehensive and (important for me as a beginner to steam) easily understood reply to my question  
Your right about getting a buzz seeing the engine, which started off life as a plastic bag of odd shaped bits of metal, suddenly burst into life and can probably only be bettered by making all those bits personally. Next project maybe  
I did wonder about the valve gear being a possible culprit and your explanation of how port opening and closing can so easily be different for forward and reverse has made me realise that I didn't really pay enough attention to getting the valve centered over the ports. Everything so darn small :shock:
So, I am going to re-visit valve timing and then as you suggest, get it installed in the launch and see how it runs.
Cheers

Greebo

[sandy_ACS]

   

Hi pd's,

Ok Greebo.... glad it has helped your understanding.

BTW.. I was not, in any way, being critical of your assembly as I am sure you followed the assembly instructions to the letter, what I was trying to convey is the fact that no matter how carefully you set the thing up, there will always remain some timing error.

Most valve gears are designed, and optimised for FORWARD running, since reverse running is usually considered the lesser function.
As with all things mechanical, geometric angular variations/errors are ever present and whilst they can be designed round, or even out in some circumstances, it is inevitable that some can not.... direction can play a big part in effect of any particular error and therefore cancelling out by design in one direction may well, more often than not, make the same error larger for the opposite direction.

The whole thing is a design trade off.

Give it some time and then take a good look at the set-up, including the location of the eccentrics relative to the crankpin.... this is one area where a degree out can make quite a significant difference and will have a marked effect on the valve timing.

Keep happy.

Best regards.

Sandy.  :computer

[GreeboTheCat]

Most valve gears are designed, and optimised for FORWARD running, since reverse running is usually considered the lesser function.

Does this in any way imply that the valve gears are "handed" i.e. one should go on the front steam chest/cylinder and one on the back? They looked pretty identical but I do take your point about the eccentric/crankpin relationship - The instructions simply said "line up the eccentric strap setscrew with the arm of eccentric" and thinking about it now, I could very easily got that out by a degree or so  :?

This is much more fun than electric motors        

[sandy_ACS]

 

Hi Pd's,

Greebo ask's

Does this in any way imply that the valve gears are "handed" i.e. one should go on the front steam chest/cylinder and one on the back? They looked pretty identical

The short answer is no.

The valve gear on the TVR-1 is, in general, well designed from an assembly point of view, and operates well if set-up carefully, that is not to say that some design compromise has not been made.

The sort of compromises I am referring to relate to the valve gear mechanism as a whole, and how this may effect each individual component.

A couple of examples may help...... take the eccentric strap and arm for example..... the bottom end attaches over the eccentric, as normal, and the top end of the arm is attached to a sliding block, located in a slide bar, the angle of which is adjustable.

A short distance below the slide attachment there is another connection point, which is the take of/connection for the valve motion.
This consists of an arm connected between the eccentric arm and a bellcrank pivoted on one of the side supports.

The distance of this take off connection ,from the eccentric centre, is critical to the valve travel/timing, however, the exact location may differ between FORWARDS and REVERSE.
Let us say the optimum location for this connection in FULL FORWARD is 1.502" and the optimum position in FULL REVERSE is 1.493" (both hypothetical values), then it would be a reasonable compromise to make this distance 1.5"... which tends to favour FORWARD RUNNING.


At the other end of the take off arm is the bellcrank, as mentioned above, which consists of 2 arms of slightly different length and the angle between these arms is another possible consideration.

Again (hypothetically), for FULL FORWARDS it may be that the optimum angle between the 2 arms is 30.3 deg and for FULL REVERSE it may be 29 deg.
Again a reasonable compromise could be 30 deg, which again would slightly favour FORWARD RUNNING.

Whilst it is perfectly feasible to make this components to say 29.65 deg, the mid point in the above hypothetical case)
the actual cost of doing so tends to get higher, due to the perceived higher angular tolerance required..... in actual fact, modern CNC machines can easily work to 0.005 deg of arc, or better... however, as soon as someone asks for a part to be made with anything above 1 place of decimal arc the cost seems to quadruple.... fact of life I am afraid.

I agree you should carefully adjust and set your eccentrics in relation to the crankpin.... once set, correctly, they should not then need to be changed.
It is some time since I last saw a TVR-1 in the metal, so to speak, however,     I believe the crown (high point) of the eccentric should be exactly in line with the crankpin... alternatively it would be at 180 deg to it, I cannot recall which.

Once you have got the eccentrics correct, then small adjustment can be made to the exact position of the slide valve on its port face, in relation to the ports, with the piston at TDC.
On these small engines, a few thousandths of an inch can make a great difference.

Another item/s to take care with is the relationship between the 2 slide bars.... these should be absolutely parallel to each other, otherwise one cylinder will get different valve timing events than the other.

Hope this explains what I was getting at a bit better.

Best regards to all.

Sandy.  :computer  

[steamboatmodel]

Hi  Greebo,
One of the most important things about running in is Lubrication, both inside the cylinders and other moving parts. With running on air the best way I have seen is to use one of the lubrication units intended for air tools, when running on steam proper steam oil is a must.
Regards,
Gerald

[woody]

From: >Woody

AS usual Sandy give a good explanation of all the questions posed to him.
Just a note to say that after receiving instructions from a paddleduck
member for my Krick marine model 2 engine, Krick states that the engine will
run better in one direction than the other.
No explanation of why, Thanks to Sandy I  now know why.

 

Hi PD's

Greebo....

First off: - well done on getting your engine built up and running.... it
gives you a real buzz does it not?

Ok, your question..... I think it is fair to say that 1 hr running (on air)
will only have begun the running in process... it will take several more
hours (on steam) before all the moving parts are fully bedded down.

The main reasons that an engine runs better in one direction than the other
are: -

Bearing tightness/alignment

Sliding, or oscillating components not fully bedded down.

And (probably the biggest contribution) valve gear setup/errors.

The first 2 of these can really be considered as a whole and constitute the
need for a running in period in the first place.

Take, for example, the crankshaft main bearing.... it is a well documented
fact that when the shaft running in such a bearing is subjected to a
sideways (radial) thrust, then the shaft will take up a position, within
the bearing, which is not in the centre, but the shaft will be deflected in
the direction away from the angle of the thrust..... in other words the
shaft becomes eccentric with regard to the bearing.

If you were to study the effects of this on a double acting steam engine,
then you would see that the crankshaft favours one half of the bearing for
one direction of rotation, and the opposite half for the other direction,
therefore, even minute, bearing miss-alignments throughout the shaft
(remember you have at least 3 main bearings) will create more or less
bearing drag depending on rotation direction.

The same thing applies to all bearings and sliding surfaces.

Sliding parts would include: - the crossheads, the slide valves and, in
your case, the slides for the Hackworth valve gear.

Oscillating parts would include all bearings and pins which do not make a
complete revolution, such as the Con-rod small end and other parts of the
valve gear mechanism.

It may seem, when you turn the engine by hand, that these are not the
cause, however, when you apply air, or steam then the thrusts applied to
the various parts will be entirely different to those you can apply by hand
turning of the shaft.
In some cases these thrust variables will combine and make rotation
somewhat easier, but equally they can combine and create additional drag
forces.

Over time, with carefull running in, and with lots of good oil, these
effects will gradually reduce and will become a very small part of the
issue.

I think though, that the biggest culprit is the valve gear, and it's
set-up.

First thing to remember is.... there is no such thing as a perfect valve
gear... and 'Hackworth' is no exception, there will always be errors in the
valve gear operation due to geometric errors between the moving parts.

If you take just one example: say one of your TVR cylinders.... and
assuming the valve gear is in FULL FORWARD position and has been set-up
exactly... then you would almost certainly find that the actual valve
events for the downward steam cycle would differ slightly from the upward
steam cycle.

e.g..... on the downward cycle the top steam port would close at 82.4% of
the stroke (82.4% cut-off) however, for the upward steam cycle, the lower
steam port may close at 81.9% (81.9% cut-off)... this means an error of
just 0.5% over the desired equality.

If you now place the same valve gear in FULL REVERSE... the results could
well become 89.0% for the down stroke and 74.8% for the up stroke... error
is now 14.2%.

Such errors are not unusual and I am sure you can see that this would
indeed create a difference in the apparent smoothness between the 2
directions.

These errors are inherent in the valve gear and even the best designed and
set-up gears would inevitably have some errors.... such is the way of
things.

Some improvement may be found by making very small adjustments to the valve
gear on your engine, it is probably worth while re-examing the set-up after
you have run the engine on steam for a couple of hours.


By far the best way is... get it in a model, load it with a real propellor
and just poodle around the lake with it..... the effects a true load can
have are considerable... and will give entirely different results than just
running, unloaded, on the bench.

Hope this helps explain things a bit.

Best regards.

Sandy.     :luck :computer

------------------------
If you want a STANDARD item I will have to charge you extra!!

 Post generated using Mail2Forum

[GreeboTheCat]

:news
Carried out all the checks suggested by Sandy and what d'ya know? it's purring like a sewing machine now, slightly better forward than reverse.
 :bow  :bow  :yeah

Pics to follow when it's mounted in the hull.

Greebo

[crash93]

Found this on a site in the USA seems to be a Gage TVR-1

Peter


http://www.xten.tv/view_video.php?viewkey=8dad4d52d063967f0a70&page=4&viewtype=&category=&chid=0|1|0