Press Fits

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Joined
Jul 31, 2010
Messages
305
Reaction score
69
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
 
With a stove and a freezer you could go +.0015 interference on the shaft. So maybe Ø.6265? If the metals are dissimilar you can press with blue loctite.
Just a wild guess. You can bias based on the application, the wheel's ability to take tensile stress from the pressing operation, or torsional pulses.

Any other clues on what the wheel will do?
 
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
A shrink fit has about three? times the holding power of a press fit.
 
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
It is simple
For a 1" shaft for ball bearing use a pree fit .0005 to .00075.
For a shaft to a wheel use a press fit of .001" to .0015.
There are tables on this but this works for most work. I can also use it with calculator.

If putting a ring or band it is great

Dave
 
It is simple
For a 1" shaft for ball bearing use a pree fit .0005 to .00075.
For a shaft to a wheel use a press fit of .001" to .0015.
There are tables on this but this works for most work. I can also use it with calculator.

If putting a ring or band it is great

Dave
Now that we are talking shrink fitting. Does anyone know the formula for shrink fitting bronze bushings Into a cast iron fitting?
 
On limits and fits (maybe these links help?):
https://www.engineersedge.com/mechanical,045tolerances/preffered-mechanical-tolerances.htmhttps://www.cobanengineering.com/Tolerances/ANSIForceFits.aspAs it is quite a complex subject, it is hard to find a "simple " table.
But the "simple" comment is that the cast iron, if it is known to be good, and a particular temper condition, can be calculated to have a certain hoop stress when pressed onto what we assume is an "incompressible" steel shaft. The circumferential (hoop) stress depends on the extension of the circumference, the cross section of the piece being extended, and the permissible fraction of the proof stress of the material of the wheel.
Whew!
( I would try 0.001in of interference, heat the cast iron in the chip pan to 150deg. C. and drop it on to the shaft. NO loctite. If clean, the cast iron will not affect the chips... The chip pan is a good heater for controlled "low" temperatures. The hot oil lubricates the fit, and won't affect the "hold" of the part on the shaft.).
When I calculated that an alloy cylinder block would fail with a steel tube fitted with the "wrong" fit tolerance, I was poo-pooed at work until there was a high warranty incidence (5 in one year - as predicted statistically!) of leaking (cracked) aluminium blocks that all occurred at very low temperature in Finland... - So then they followed my changed fits. The equivalent cast iron block had a very substantial boss where the tube was press-fitted and never failed. But that was mostly due to the differential expansion of the steel to cast iron versus Aluminium to cast iron. (Ignored when they had carried the fit from cast iron to aluminium engine blocks - then reduced the boss size because they thought it would be OK...).
So there is no SIMPLE rule that works.
K2
 
Last edited:
Tim,
The "relevant" comment for me is:
"Locational Interference Fits [LN]:
These fits are used where accuracy of location is prime importance. Locational interference fits are used for parts requiring rigidity and alignment with no special requirements for bore pressure. The parts can be assembled or disassembled using cold pressing and greater forces or hot pressing. Such fits are not for parts designated to transmit frictional loads from one part to another by virtue of the tightness of fit. SHAFT "p6" with Hole "H7" gives a true interference fit. It is the standard press fit for steel , cast iron, or brass to steel assemblies. "
Does that match your needs? - So a max interference of 0.0008" is suitable for your shaft to wheel, at 0.625" dia.
K2
 
I would feel much more comfortable with a press fit on a ductile iron flywheel.

I have seen several flywheel failures on small old spoked engines, and it seems like the breaks are usually at or near the hub, which makes me think perhaps the flywheel was pressed into place, or more likely the hub is where the maximum stress occurs and it was not a pressed-on flywheel.

Standard gray iron is pretty strong, especially in the higher classes.

I did press the green twin cranshaft into my gray iron flywheel, and I did not use loctite, and I have not had any hub cracks.
I think the key is to not use an excessive amount of interference, ie: don't exceed the strength of the gray iron.

.
 
Here is a list of fits that Solidworks came up with. s6 shaft size is the maximum recommended for cast iron wheels. Reamers are allowed to produce an oversized hole and still be in limit.

Tim

FITS003.jpg
 
Last edited:
Here is a list of fits that Solidworks came up with. s6 shaft size is the maximum recommended for cast iron wheels. Reamers are allowed to produce an oversized hole and still be in limit.
snip
Tim
Methinks that I would rather use my trusty Machinery's Handbook (25th ed) and peruse the pages from 635 to 682 (on fits).

I'm not sure that I trust any CAD program's spouted references.
(I realize that this might make me a Luddite but so be it! There are references and then there are tools - - - - the two should not be co-mingled nor conflated imo. Far more detail and even, at least some, reasoning presented as well. Besides - - - - the above table just underscores - - - to me - - - the lack of understanding of the real world that the paper world has - - - - how in the sam crap do you even think of measuring something ACCURATELY to seven decimal places. The costs of even trying to do that to 5 places is scary - - - - yet here to someone who understands only from paper - - - - well I must spec to 7 digits - - - - I'll be generous and only ask for 6 - - - - what tripe that would be for a design!)
 
Last edited:
But I think more importantly, how does the Solidworks chart compare with the info in the Machinery's Handbook?

I think the 7 digit runout was to just illustrate the point, not that you would machine to that or use that many decimal places.

Edit:
When assembling the green twin crankshaft components, I noticed that if you use a 1000/th over ream for a very short section, such as 1/16" or a little more, you can get the shaft to press straight in without trying to press in out of square.

.
 
Last edited:
But I think more importantly, how does the Solidworks chart compare with the info in the Machinery's Handbook?

I think the 7 digit runout was to just illustrate the point, not that you would machine to that or use that many decimal places.

Edit:
When assembling the green twin crankshaft components, I noticed that if you use a 1000/th over ream for a very short section, such as 1/16" or a little more, you can get the shaft to press straight in without trying to press in out of square.

.
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.
 
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 am with Brian and Dave. One part in one thousand for a press, and something more for a shrink. What matters more than a table of standards based fits is how accurately can you machine and measure the parts.

A 4-1/8" wheel at 1/8 scale is presumably a truck wheel not a driving wheel. The OP says the wheel holes will be finished with a 5/8" reamer, so they will probably be a running fit on 5/8". I would make a test stub to gauge the holes, finding out what diameter will just push in by hand, and then try to get the axle wheelseats to 7 to 8 tenths bigger, with a good finish.

I think I would then shrink them on rather than press them, at least partly because my only press is the bench vice, and the workshop door takes me into the kitchen, where there is an electric hotplate and a freezer.
 
I use the chip- pan in the kitchen and a room temperature part (the fridge or freezer doesn''t make much difference). The frier has a pretty accurate thermostat and I don't need to go the smoking temperatures. It also has a nice basket for putting a part in. It doesn't risk overheating the part, especially if it has a particular temper.
Enjoy,
K2
 
But I think more importantly, how does the Solidworks chart compare with the info in the Machinery's Handbook?

I think the 7 digit runout was to just illustrate the point, not that you would machine to that or use that many decimal places.

Edit:
When assembling the green twin crankshaft components, I noticed that if you use a 1000/th over ream for a very short section, such as 1/16" or a little more, you can get the shaft to press straight in without trying to press in out of square.

.
Hmmmm - - - - except to someone who did have diddly idea of what machining was or anything about it they would be quite inclined to spout ALL 7 digits.

I was, and am, railing against the extreme disconnect between reality and paper.
IMO far too many practitioners in mechanical engineering live exclusively on 'paper'.
 
Back
Top