N&W S1a Project - Tender Build

Where users can chronicle their builds. Start one thread and continue to add on to it.

Moderator: Harold_V

User avatar
gamh44
Posts: 81
Joined: Tue Mar 24, 2015 4:03 pm
Location: Australia

Re: N&W S1a Project - Tender Build

Post by gamh44 »

All

Just finished the last wheel and here they are.
55. WheelsDone.JPG
For reference, I found the receipt for the material, which is grade 1035 and referred to as a mid carbon steel (0.35% Carbon). Described as having good machinability but it is definitely harder than the mild steel (grade unknown) I have lying around. Nonetheless, the advice above was priceless - thanks again.

Now back to the bogie frames.

Geof
User avatar
gamh44
Posts: 81
Joined: Tue Mar 24, 2015 4:03 pm
Location: Australia

Re: N&W S1a Project - Tender Build

Post by gamh44 »

Did a test today to see how the shrink fitting the wheel to the axle will go.

Since this is something I have not done previously I used the test wheel I turned and machined up a small test stub axle based on the 0.001" per 1" rule of thumb.

Unfortunately, not all the wheel boring ended up perfect. A couple of the inside diameters ended up a little (maybe 0.0005") oversize, so rather than just using the *drawing* size for the axle, I came up with this plan.

The axle boxes are outside the wheels so I will need a fair bit of material beyond the wheel. So I turned this section down to fit the wheel at hand to a snug fit. Then this was measured and 0.001" was added (or just a little less - the axle diameter is actually 0.625" so it should be 0.000625" but I can only guess that as I only have 0.001" measuring tools). With this ready for shrink fitting, I turned up the outboard diameter ready to fit into the bearing.

Hope that makes sense. Please feel free to point out any flaws in my logic here.

Apply heat to the wheel and drop the axle in. It started well but needed a slight tap with a hammer to encourage it in. I suspect I had it slightly skewed when I originally dropped it in.

Below is the result - just after the fit and after cleaning up. Very happy with the result.
56. AxleTest_1.jpg
57. AxelTest_2.JPG
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: N&W S1a Project - Tender Build

Post by Harold_V »

I'm a little confused with the numbers you posted. If you have a diameter that measures .625" and you add a thou, it would then measure .626". The number you posted represents a miniscule amount- slightly more than six tenths of a thou.

In this case, you'd likely be well served to use a little more interference. For a press fit it's ok, but if you shrink fit, a little more is good (guarantees the parts won't come apart in use) and shouldn't present any problems in assembly (just heat the wheel a little hotter to compensate). I'd shoot for no less than .0015" interference, so long as you're heat shrinking, and .002" would be acceptable.

For a thin section (like the wheel), it's possible to get the axle/wheel not perfectly straight. Something to register the two would be a good idea.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
gamh44
Posts: 81
Joined: Tue Mar 24, 2015 4:03 pm
Location: Australia

Re: N&W S1a Project - Tender Build

Post by gamh44 »

Harold

Thanks for your thoughts and apologies for not being clear with my description.

Using the 0.001" per 1" rule of thumb, the interference on 0.625" would be 0.000625" or just over 6 tenths of a thousandth as you say. I went for 1 thou over just to be sure so the ID of the wheel was 0.625" and the OD of the axle was 0.626". Seems pretty firm but I might redo the test at 0.002" (0.627") and see how it goes.

The real wheels measure 0.949" across so I am hoping that is enough to keep them true without a jig of sorts. This smaller one was pretty good. I'll check the first one and see how it all goes.
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: N&W S1a Project - Tender Build

Post by Harold_V »

A belated reply, thanks to an issue with my tired, old computer.
Here's what I was trying to post when it quit on me:
I must have been in a fog, as I now understand what you meant.
If you have an arbor press, it would be a good idea to attempt to dismantle the test piece, keeping in mind that it may gall when doing so. If it doesn't, it may come apart pretty easily, especially if your diameters had any ridges. The idea behind a little more interference is to ensure that you have a fit that compensates for such a condition. If the diameters were ground or polished, that may not be necessary. Just make sure that the two diameters, when down to solid metal, still offer an interference fit. If not, you risk movement of the parts when in use, which will lead to premature failure. That's the reason a knurl is no substitute for a proper fit.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
gamh44
Posts: 81
Joined: Tue Mar 24, 2015 4:03 pm
Location: Australia

Re: N&W S1a Project - Tender Build

Post by gamh44 »

Harold

Thanks. I don't have a press handy but should be able to get access to one so I will see how that goes.

By my calculations, to expand the inside diameter of the wheel 1.4 thousandths, it needs to be at 230*C, which should be about the first colour change of the metal to light straw. If I go to blue (around 300*C) I get 1.9 thousandths. Going as far as dull red will give about 4.5 thousandths.

I suppose the question is how far can I heat the wheel before risking deformation? Once I press the test piece apart I will do a couple more tests.

Geof
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: N&W S1a Project - Tender Build

Post by Harold_V »

Deformation, unless there's a huge amount of internal stress, shouldn't be a concern. What I would do is avoid heating to the point of causing surface degradation, which happens at red heat. I would make the choice of targeting the straw color, which should provide a generous margin of safety.

Also, consider the condition of the machined surfaces. In sizing the axles (assuming that hasn't been done yet) and you don't have grinding capabilities, take them to size by polishing, leaving less than a thou to bring up the finish and achieve your target diameter. That's a common practice for those of us who have been faced with tight tolerance work without the luxury of grinding. I would consciously avoid trying to turn to the target diameter, if for no other reason, to improve the surface finish.

Edit to include grinding comments.
Many think of a tool post grinder when it comes to finishing such diameters. While they (usually) offer the opportunity for better surface finishes, you're still locked in to the capabilities of the machine on which it would be used. Two things trouble me with that situation. The first thing is that only a fool would use flood coolant on a lathe with a grinder, and grinding dry tends to be difficult because of the resulting heating.

The second issue is the lack of size control. A grinder can hold a tenth without much trouble, while the typical lathe struggles to hold a thou. The ability to fine feed the wheel just doesn't exist on lathes, while on grinders fine feeds are very much a part of the machine's capabilities. As an example, my Overbeck universal cylindrical has a .000039 feed (on the diameter), so holding tight tolerances is routine.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
Greg_Lewis
Posts: 3014
Joined: Wed Jan 15, 2003 2:44 pm
Location: Fresno, CA

Re: N&W S1a Project - Tender Build

Post by Greg_Lewis »

When fitting, are you also going to freeze the axles?
Greg Lewis, Prop.
Eyeball Engineering — Home of the dull toolbit.
Our motto: "That looks about right."
Celebrating 35 years of turning perfectly good metal into bits of useless scrap.
User avatar
gamh44
Posts: 81
Joined: Tue Mar 24, 2015 4:03 pm
Location: Australia

Re: N&W S1a Project - Tender Build

Post by gamh44 »

Harold

Thanks again. I took the test piece to work today and put it in the 20T press. It held well but I did manage to press it out. There was indeed a small amount of galling.

I brought it home and redid the whole arrangement with about 0.002" interference. I needed to get the wheel quite a lot hotter to fit the (dummy) axle. I'll try it in the press tomorrow.

Axles are not turned yet so I will definitely finish them oversize and polish them from there.

Greg, are you suggesting the freezer in the kitchen? Does it help much? I have read to use dry ice (and denatured alcohol for even extra cooling) but I think I'll see how we go with the heat first.

Geof
User avatar
Greg_Lewis
Posts: 3014
Joined: Wed Jan 15, 2003 2:44 pm
Location: Fresno, CA

Re: N&W S1a Project - Tender Build

Post by Greg_Lewis »

Geof:
Your choice as to freezing. I shrunk some tires on drivers by putting the driver centers in the freezer and the tires in the oven. I had to move fast but it worked. Considering the size you're working with I'd probably go for the dry ice.
Greg Lewis, Prop.
Eyeball Engineering — Home of the dull toolbit.
Our motto: "That looks about right."
Celebrating 35 years of turning perfectly good metal into bits of useless scrap.
User avatar
gamh44
Posts: 81
Joined: Tue Mar 24, 2015 4:03 pm
Location: Australia

Re: N&W S1a Project - Tender Build

Post by gamh44 »

So I did the press test with the wheel that had 0.002" interference and the gauge on the press got to 5T before I decided enough was enough.

So onto the real deal and here is the first of four. Three to go. Quite happy with the result.
58. WheelsAxlesDone.JPG
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: N&W S1a Project - Tender Build

Post by Harold_V »

The .002" interference is great insurance. Well done!

H
Wise people talk because they have something to say. Fools talk because they have to say something.
Post Reply