Press Fits

[ajoeiam]

Asbolutely! Tripe! That may work, however, for someone who has lazer cutting and a 5Million$ 6 axis machine. But I don't have one--yet.

Sorry - - - - even a $5E06 6 axis machine won't be accurate to more than perhaps 0.0001".
Grinders with that kind of price tag might get into the 0.000010" range.
I would not like to use such a machine - - - - the degree of control needed in the heating and cooling of the work area would be tremendous!!!!
I think you would have to control the temperature around that machine to within maybe 0.5 degrees C - - - - and that's not easy to do!!!!

Laser cutting is most often to +/- 0.001" and I'd bet that most of the shops running such would be quite happy to hold +/- 0.002" day in and day out.

 

[fltenwheeler]

I redid the file with more realistic sizing.

Here is a treasure trove of information.

Tim

 

[ajoeiam]

The Machinery's Handbook, if I am reading it correctly, seems to go out usually to 4 decimal places on fits, but sometime 5 decimal places.

There is a whole lot of different fit types; pages and pages of them.

I don't think I need to know that much, but good to know for a reference.

.

I pointed out the reference literal because there are just so many different levels of fits.
IMO its actually a huge subject that is often distilled into a quite simple rule of thumb that whilst it is often quite useful - - - - well - - - it isn't always!
(I prefer the rule of thumb over the overly belabored highly turgid information that a 'CAD' program provided!)

 

[ajoeiam]

I redid the file with more realistic sizing.

Here is a treasure trove of information.

Tim

Thanks mr Tim.

I want to apologize to you - - - - you may be thinking I was ranting at you - - - - nothing further from the truth - - - - just frustrated by 'tools' that purport to really solve every problem literally creating more problems than they're actually solving.

Excellent machine tools (I can name more than a few common brands that are anything but!) will even have a hard time consistently holding even to accurate +/-0.0005" day in day out.
More machines with smaller work envelopes can achieve high tolerance than can those with larger work envelopes.
When you look into tools that have very large work envelopes and still hold high tolerances - - - - that's when my sense of awe goes way high!
(You really don't want to ask for pricing on that level of tools though - - - - its also sorta rarefied country!)

 

[trlvn]

Supposing a '5/8 inch' hole, how does one measure it with tenths accuracy (0.0001")? Can you get gauge pins in increments of a tenth?

In a controlled manufacturing situation, I suppose you would have Go/NoGo gauges for the exact press fit requirements of the application? What do these 'feel' like when checking the hole? How much force is normal or permitted? As the force gets larger, the hole is being minutely stretched, no?

What about roundness? Do you have to check the hole in more than one orientation to ensure it isn't egg shaped or some other aberration of "round"? If it is egg-shaped by just a couple of tenths, that probably is the difference between a heavy press fit with heat versus a cold light press fit?

Challenging stuff!

Craig

 

[fltenwheeler]

Hi Craig

If you would click on the link it will explain a lot.

I can measure to the 0.0001” of a 5/8” hole with a snap gauge.

My reamed 5/8” hole is 0.6256”.

If you look at the second column you will see that a 5/8” reamed hole is allowed to be 0.0007” over sized and still meet the H7 limits.

So with an s6 shaft limit you would have from 0.0004” to 0.0015 press fit. This meets the ANSI B4.2-1978 limits for manufacturer.



Tim

 

[Steamchick]

As I use a Vernier Caliper that is repeatable to 0.002" (Not a digital caliper!), and my MODEL workshop is about capable of the same, there's a lot that relies on "fitting" by feel of the fit, and even using a VERY slight taper lead on a pin, so when inserted in a bore (with marking blue) I can identify the contact size and extract the dimension, to machine more accurately by touching this point with the tool. (Sorry, a bit hard to describe the method!). The methodology suits me for pistons, or interference fits! I do have some "ex-works" GO-NOGO (conformity) gauges for some sizes, I just haven't found an engine that I am making that has those sizes... Craig, when you use these gauges it becomes easy to recognise when the Go actually goes, and the NOGO doesn't. Handling measuring equipment is always like handing your girlfriend. GENTLY, Gently! - The delicate touch is best. It's how a "Gentleman" succeeds.
K2

 

[Richard Hed]

Sorry - - - - even a $5E06 6 axis machine won't be accurate to more than perhaps 0.0001".
Grinders with that kind of price tag might get into the 0.000010" range.
I would not like to use such a machine - - - - the degree of control needed in the heating and cooling of the work area would be tremendous!!!!
I think you would have to control the temperature around that machine to within maybe 0.5 degrees C - - - - and that's not easy to do!!!!

Laser cutting is most often to +/- 0.001" and I'd bet that most of the shops running such would be quite happy to hold +/- 0.002" day in and day out.

I'm sure yhou're right, however, the point I am making is those numbers could be used in EXTREMELY expensive equipment, otherwise they are essentially meaningless except to truncate them at 4 places.

 

[Zeb]

I love how we're talking about a model train but end up learning all kinds of neat stuff. The old 'Murica National Standard I have for interference locational fits only goes to ten thou.
I would agree on using pins or even a drill bit shank for reference. On thinner material I'll see three sided holes from drills and 5+ star patterns depending on the reamer flute count. If the hole was ovaled on one end or drilled, all of the hoop stress (new term for me library Steamchick, thank you!) would concentrate on those points, even with embedment.

Another interesting thought is, some places will draw a BB to strengthen a cylindrical surface. You would get a similar effect if you deformed the cast iron plastically a bit with cold installation. Too much and you'd start tearing metal. With a hot install, you could get away with more interference and no tearing but then you're left with a lot more tensile stress on the wheel.

I ythinkpf the biggest service to humanity...or if Keith Appleton runs into this, is to use the spidey sense and make it so the wheel is serviceable down the road.

 

[Richard Hed]

I love how we're talking about a model train but end up learning all kinds of neat stuff. The old 'Murica National Standard I have for interference locational fits only goes to ten thou.
I would agree on using pins or even a drill bit shank for reference. On thinner material I'll see three sided holes from drills and 5+ star patterns depending on the reamer flute count. If the hole was ovaled on one end or drilled, all of the hoop stress (new term for me library Steamchick, thank you!) would concentrate on those points, even with embedment.

Another interesting thought is, some places will draw a BB to strengthen a cylindrical surface. You would get a similar effect if you deformed the cast iron plastically a bit with cold installation. Too much and you'd start tearing metal. With a hot install, you could get away with more interference and no tearing but then you're left with a lot more tensile stress on the wheel.

I ythinkpf the biggest service to humanity...or if Keith Appleton runs into this, is to use the spidey sense and make it so the wheel is serviceable down the road.

Proof of Evolution: ythinkpf

 

[Charles Lamont]

As I use a Vernier Caliper that is repeatable to 0.002" (Not a digital caliper!), and my MODEL workshop is about capable of the same, there's a lot that relies on "fitting" by feel of the fit, and even using a VERY slight taper lead on a pin, so when inserted in a bore (with marking blue) I can identify the contact size and extract the dimension, to machine more accurately by touching this point with the tool. (Sorry, a bit hard to describe the method!). The methodology suits me for pistons, or interference fits! I do have some "ex-works" GO-NOGO (conformity) gauges for some sizes, I just haven't found an engine that I am making that has those sizes... Craig, when you use these gauges it becomes easy to recognise when the Go actually goes, and the NOGO doesn't. Handling measuring equipment is always like handing your girlfriend. GENTLY, Gently! - The delicate touch is best. It's how a "Gentleman" succeeds.
K2

I suggest turning (and if necessary, polishing) a bit of scrap to a firm push/wringing fit. The wheel seats (or whatever) can then be machined to an appropriately larger dimension. This way, the tentative feely bit is done in fitting your home made gauge. No need for a gauging lead on the actual parts, and no need to buy expensive gauges.

This method does need an instrument with better than +/- 0.002" accuracy, though. I have fairly recently bought a 0-1" micrometer with a tenths vernier. I use it with a pinch of salt, as I do not have a temperature controlled inspection room, but it is a bit better than estimating the tenths with a standard mic. (It is 4.7°C and 56% humidity in the shop this morning.)

 

[Harglo]

What would the proper machined size for a shaft that will be pressed into a 5/8” reamed hole in a cast iron wheel that you want to stay put?

I know about Loctite but I am interested in press fits.

Thanks

Tim Meyer

Been flowing this press fit. Try measuring the hole after you have reamed it likely its a bit larger than the stated dimension on the reamer due to some miss alignment of the tail stock or drill press. for sure the chuck its self it doesn't take much of run out due the most of the reamers have such a long shank which I don't know why. Length of the fit has some bearing on the fit all so. Best to bore if ya can.

Harvey

 

[fltenwheeler]

Been flowing this press fit. Try measuring the hole after you have reamed it likely its a bit larger than the stated dimension on the reamer due to some miss alignment of the tail stock or drill press. for sure the chuck its self it doesn't take much of run out due the most of the reamers have such a long shank which I don't know why. Length of the fit has some bearing on the fit all so. Best to bore if ya can.

Harvey

Hi Harvey

The H7 hole size takes into account the tolerance that a reamer can produce. From 0.6250" to 0.6257" diameter. For most machinists it is easier to measure the diameter of a shaft rather than a hole.

Tim

 

[ajoeiam]

Hi Harvey

The H7 hole size takes into account the tolerance that a reamer can produce. From 0.6250" to 0.6257" diameter. For most machinists it is easier to measure the diameter of a shaft rather than a hole.

Tim

Hmmmmmmm - - - - methinks it is not so much 'easier' as it is possible.

Using telescopic gauges - - - I'd be willing to bet that the 'feel' could easily range over 2 or even more tenths. Then the measuring tool itself is at best +/- 1 of the smallest unit (and enough are at 2 if you have a vernier scale) so measuring an inside bore - - - - not only not so easy but its accuracy is, at least in my experience, also less.

 

[Asm109]

Oh, its possible, just not cheap. Mitutoyo Bore Gage Set, Holtest, 0.8-2 In 368-918 | Zoro

I have used this type of gauge for boring holes for ball bearing fits. They work great. Spoiled me for what I can afford at home.

 

[Charles Lamont]

In our home shops we are not operating production lines making parts to fit some other item that may be made elsewhere, so as far as I am concerned, an H7 fit and its range for a particular size are of no concern or interest.

In any case, the standard fits tables are merely a simplified codification of what works, and it is thinking the wrong way round to take them as the basis of all wisdom.

 

[Steamchick]

Hi Charles, I just use the tables a a "target fit". And yes, I have an uncalibrated mic with vernier on the barrel, but it is a compatit8r, not an "absolute" measurement device. By comparitor, it compares dimensions measured to "standards" - such as anything close with a non size, like the shank of a drill, milling cutter, or reamer. Or comparing measurements of the bore gauge to shaft, etc, cut-to-cut,,etc. And I only use the mic when I get close to size as the venerable caliper is so quick and convenient And if I drop the caliper it is less to cry about than a broken mic.
Each to his own, as the "best practice" is the one with least operator error, fewest mistakes, etc.
K2

 

[Harglo]

Hmmmmmmm - - - - methinks it is not so much 'easier' as it is possible.

Using telescopic gauges - - - I'd be willing to bet that the 'feel' could easily range over 2 or even more tenths. Then the measuring tool itself is at best +/- 1 of the smallest unit (and enough are at 2 if you have a vernier scale) so measuring an inside bore - - - - not only not so easy but its accuracy is, at least in my experience, also less.

Yes I agree about measuring the shaft vs the hole.
Harvey

 

[davidyat]

I built a crankshaft from machined parts with Loctite 680. Two of the 8 joints have loosened up. I had the holes bored with a 0.376 reamer and the shafts are 0.375. Six of the joints seem solid and two loose. Should there have been more room in the fit larger than 0.001 for the Loctite to work properly? I did use Acetone to clean up the hole and shaft prior to assembly and I didn't touch anything for 24 hours to allow for curing. The close up shows the shaft no longer flush with the joint.
Grasshopper

 

[trlvn]

Oh, its possible, just not cheap. Mitutoyo Bore Gage Set, Holtest, 0.8-2 In 368-918 | Zoro

I have used this type of gauge for boring holes for ball bearing fits. They work great. Spoiled me for what I can afford at home.

That is a beautiful tool set but the minimum size it will measure is 0.8 inches. The OP's application was a 5/8" hole.

Craig