Miserlou57 wrote: ↑Mon Feb 03, 2020 10:20 pm
I don't think this is the correct takeaway. It seems you are mixing up clearance with tolerance. They are not the same. I'm surprised others have not chimed in here, as this has been covered here on Chaski before.
OK, I'm not sure what you think I don't understand here. Tolerance in my shop is the +/- .005 or whatever on the callout.
Mr. Precision made the bushings with perzactly not enough looseness in the fit. It ran fine till it left the nice level track in the shop, encountered the crappy track in the yard, suspsension flexed, bush locked up on the pin, rod bent.
Now the old 0-4-0 in the shop with the really loose brasses never had an issue.
The point that was made is that the ± .005" tolerance isn't the problem. The dimension chosen is. If the fit doesn't function, opening the tolerance most likely isn't a cure, as that means the fit can be all the tighter, or, assuming it went the other direction, too loose
What needs to happen in this instance is the bushings (or the pins) need to undergo a change of size. That would create the required clearance, and the TIGHTENING (not loosening) of the tolerance would ensure that so long as you held the tolerance, the parts would function as required. The tolerance you referenced should NOT be considered the tolerance for running fits, as it is already too loose to be functional. In this case, the clearance desired between the pin and bushing should be determined, then a TIGHTER tolerance applied, so the dimension desired is held. That would assure assembly and function of the parts.
Fits.
One can NOT have a ± .005" tolerance unless it applies to one of the features, then the mating piece is fit accordingly. The ± .005" can not apply to both pieces, as when they are made to extremes, they will either fall together, or they can't be assembled at all (one can not put a .755" pin in a .745" bushing, and that's what the tolerance allows). So then, the fit must have a tight tolerance, not a loose one. For a small diameter (like 3/4"), .0005" spells the difference between a slip and a press fit.
Like I said, tightening tolerance guarantees a device will work, but it comes at a higher cost, as close tolerance work demands greater care, especially if it's done manually. It also demands a higher skill level from the operator, and that comes ONLY from experience. For that reason, one should practice machining by trying to hold tight tolerances until it becomes routine. When it does, one can machine to rigid specifications without dread.
What you should take away from this is that it is NOT a tolerance issue. It is an engineering issue, one of not choosing a size that will provide the necessary clearance to compensate for the undulation of the engine as it moves along a not so perfect track.
H
Wise people talk because they have something to say. Fools talk because they have to say something.
Miserlou57 wrote: ↑Mon Feb 03, 2020 10:20 pm
It seems the aforementioned issues with the engine being too "tight" were with the design / choice of fit. If "Mr. Precision" manufactured his part exactly as the design intended, and the part failed, then the failure was of the design and not the manufacture. Maybe the choice of fit was not right. Maybe the part heated up more than it was supposed to. Tolerance has nothing to do with that. Precision is a good thing!
Except for the fact that many Live Steam drawings (especially the old-school ones) do not show ANY tolerance, just dimensions. It was/is expected that the builder know how an engine works and that they should use their own knowledge and judgement to determine what type of fit components should have; running fit, interference fit, etc.
Therefore if someone uninitiated with how a locomotive should work made an engine to the plans without a thought about how the parts work together it is very conceivable that it might be "too accurate" and would not run correctly.
I must agree with Mr. Shields sentiments:
Bill Shields wrote: ↑Tue Feb 04, 2020 5:28 pm
As Charlie (and others who have actually built a few locos with little more than a lathe and drill press) used to (and still do) say:
'runs best when it is about worn out..clanking all the way...'
If you go to a track and spend some time around running engines it may come as a surprise how crude some of them really are...and those are the ones that seem to just run and run and run...
FKreider wrote: ↑Wed Feb 05, 2020 6:56 pm
Therefore if someone uninitiated with how a locomotive should work made an engine to the plans without a thought about how the parts work together it is very conceivable that it might be "too accurate" and would not run correctly.
Nope!
That's still an issue of improper size selection. It has nothing to do with being "too accurate". How accurate one builds an engine is determined by the tolerance selected. In other words, if one were to determine that all sizes were to be held to ±.0005" (excluding fits, which must be closer), the engine would have been built to high accuracy, and, assuming the proper sizes were chosen, it would assemble and run as expected. By sharp contrast, if the wrong sizes are chosen and the engine is built to the same rigid standards, it won't operate as desired. Both are built to the same accuracy. One of them is just dimensioned wrong.
This type of discussion is a good example of why a novice should do due diligence when undertaking a difficult project like building a steam locomotive. Without the proper knowledge of the fits required, he is unlikely to succeed on the first attempt. He MUST come to terms with the idea that the sizes of the parts machined determine if they perform, or not, it is not the tolerance he chose to use. That simply makes no sense.
The harsh reality is an engine could be built with extreme tolerance (something as absurd as ±.0002") and operate perfectly well, assuming the proper dimensions are chosen.
Please do not continue to discuss how an engine operates as being determined by the tolerance used. That has nothing to do with the issue. It's high time the folks who discuss this matter get it right.
H
Wise people talk because they have something to say. Fools talk because they have to say something.
I think you are missing each other's point of view. Harold is saying you can build a locomotive with lots of clearance but to tight dimensional tolerance and it will work fine. This is true. Bill is saying if you need lots of clearance (which you do) then you can build with loose tolerances. This is also true.
Steam locomotives were built in the old days with much looser tolerances than engineering is done these days. But if you want to be technically correct then Harold is right. Me, I just struggle trying to make parts to any kind of tolerance.
KenG wrote: ↑Thu Feb 06, 2020 8:23 am
Bill is saying if you need lots of clearance (which you do) then you can build with loose tolerances. This is also true.
No, it isn't, and that's the point I keep trying to have readers understand.
Tolerance refers to the amount one can deviate from a nominal dimension, and has nothing to do with the amount of clearance one desires or needs (aside from assuring that it exists). Surely you can understand that if tolerance is too open (in lieu of being "tight") there's the distinct possibility that parts won't fit, or that they will be too sloppy. For that reason, tight tolerance is not a bad thing, as it guarantees that they will be useable, and that's the reason tight tolerances exist. Readers are simply using the wrong terminology, and while they may agree amongst themselves that they understand, they don't. Your post is another example that it's true.
Steam locomotives were built in the old days with much looser tolerances than engineering is done these days.
Again, they were, perhaps, built with greater clearances. The time frame involved makes no difference. Clearance and tolerance are not one and the same. Engines were built with the necessary CLEARANCE, and they would have suffered the same consequences models do if the tolerance would have been wide open. Those engines were built with the necessary tolerances, just as engines of today must be built with the necessary tolerances. To ensure that the engines did and will run, the tolerance, of necessity, must be reasonably tight. If not, parts will not fit.
Tolerance and clearance are not the same thing. Until you understand that, you're going to continue to get it wrong.
H
Wise people talk because they have something to say. Fools talk because they have to say something.
What you should take away from this is that it is NOT a tolerance issue. It is an engineering issue, one of not choosing a size that will provide the necessary clearance to compensate for the undulation of the engine as it moves along a not so perfect track.
H
Which is exactly what I intended to convey.
Nowhere in my original post did I say the word "tolerance".
I replied to a post talking about someone using too tight a fit, and said I had seen the same thing in full scale. I then clarified that comment by specifically stating what you state in the second sentence.
What you should take away from this is that it is NOT a tolerance issue. It is an engineering issue, one of not choosing a size that will provide the necessary clearance to compensate for the undulation of the engine as it moves along a not so perfect track.
H
Which is exactly what I intended to convey.
Nowhere in my original post did I say the word "tolerance". I replied to a post talking about someone using too tight a fit, and said I had seen the same thing in full scale. I then clarified that comment by specifically stating what you state in the second sentence.
I recognize that you didn't use that term, Rudd. How I wish others didn't, as it's very misleading to those who understand the issue.
Perhaps a repeat will help.
Tolerance does not relate to clearance. Tolerance is the amount of deviation from a dimension that is permitted (or acceptable).
Clearance is the size difference between two pieces---the amount of space that separates them. If that spacing/fit is critical, tight tolerance (along with proper engineering, or size determination) assures that the parts will fit properly. Loose tolerance does not, therefore an engine is not built to "too tight of a tolerance". That makes no sense.
When an engine binds, it has clearance issues, not tolerance issues.
H
Wise people talk because they have something to say. Fools talk because they have to say something.
Enlightening discussion!
I made a test pin from hot rolled mild steel and case hardened (kasenit) the rod bushing end and left the end to be pressed into the wheel 'soft', just to see how much the pin diameters changed after heat treating. After buffing the treated pin with emery cloth, both the case hardened and untreated ends of the pin lost 0.0015" to the diameter.
John
jscarmozza wrote: ↑Wed Feb 19, 2020 5:09 pm
Enlightening discussion!
I made a test pin from hot rolled mild steel and case hardened (kasenit) the rod bushing end and left the end to be pressed into the wheel 'soft', just to see how much the pin diameters changed after heat treating. After buffing the treated pin with emery cloth, both the case hardened and untreated ends of the pin lost 0.0015" to the diameter.
John
I really appreciate you posting your results.
I happened to make the first crank pin last night. It's approx .280" diameter x 0.535" long cylinder, with 1/8" diameter hole bored through it's length, so it's a fairly small part with not a ton of mass. For some reason, I did cut it to *exact* dimensions (~0.0004 tolerance on the diameter), so you loosing 0.0015" is a little disheartening.
On the other hand, what choice do I have? Maybe I go home and blast it with my MAPP torch, and drop it into the Cherry Red I have, and just see what happens? I guess I have nothing to loose, and really no other way to find out how this needs to be done, eh?
