question about aluminum shrinkage

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

Jasonb

Project of the Month Winner!!!
Project of the Month Winner
Joined
Mar 30, 2008
Messages
3,094
Reaction score
783
Location
Surrey, UK
FYI
It simple 3/16" shrink is.
12.1875 ÷ 12 = 1.05625 for computing .

Dave

Are you sure about that Dave as that is 5 1/2% .

12.1875 divided by 12 = 1.015625 near enough 1 1/2%

For a bit of home casting you can print out a modified rule and stick it to a strip of wood or metal, cover in clear tape.
 

Bazzer

Well-Known Member
HMEM Supporting Member
Joined
May 14, 2017
Messages
143
Reaction score
38
Location
Southampton, England
Some good advise above but be careful to understand that you are not applying the solidification shrinkage to X.Y.Z dimensions, Jason and others explained this well.

My 1/77 figure came from actual investment castings that we were doing in the foundry under the conditions that we were working to.

Please don't forget to work out a shrinkage factor for the 3D printed material as well and then add the two together. Although surprisingly with my recent OS 60 front housing project the 3D print shrinkage was not high.

The wax in the attached photo is quite a bit nicer than it looks, the translucent nature allows the internal structure to be visible and those are the long lines.
 

Attachments

  • Wax5.JPG
    Wax5.JPG
    402.8 KB · Views: 0
Last edited:

Jasonb

Project of the Month Winner!!!
Project of the Month Winner
Joined
Mar 30, 2008
Messages
3,094
Reaction score
783
Location
Surrey, UK
That 1/77 is as near as makes no difference to the 1.3% in that BS document as it comes out to 1.2987%, maybe the 1/75 was just simpler to work out back in the day

The 3/16" / foot was also an easy one to work with as that equates to 1/64" per inch and comes out at 1.56%

Nicely printed wax Barrie, was that for a change of bearing spec/size?
 

Bazzer

Well-Known Member
HMEM Supporting Member
Joined
May 14, 2017
Messages
143
Reaction score
38
Location
Southampton, England
Hello Jason

The front housing for that OS 60 is no longer available but we have a number of crankcases that we want to use in control line speed applications, I was going to machine some new ones from solid.

However we have a bigger project to make crankcases for a 40 size pylon race engine and wish to do those by investment casting.

The OS 60 front housing is to dip our toe's in the water of the 3D printed/investment casting process, the investment casting I understand well it is just the 3D printed part for casting that is a bit less clear.

The housings were printed on a Dremel 3D20 using a material called polycast.
 

abby

Well-Known Member
Joined
Apr 13, 2009
Messages
515
Reaction score
173
Too much confusion being talked here , the shrinkage is the difference between the size of the pattern and the cooled casting .
I have made aluminium castings for a company that produces extruded sections, these were right angle joining pieces and needed to be VERY accurate.
I have also cast threaded aluminium castings with an M40 thread that had to screw on to a male thread.
Both jobs were cast in LM25 alloy and the shrinkage allowance for both was 2.0%.
I have checked the shrinkage on thousands of non ferrous castings over several years and found that 2.0% pretty much covers everything.
Dan.
 

SmithDoor

Well-Known Member
Staff member
Global Moderator
Joined
Feb 12, 2009
Messages
1,462
Reaction score
232
Location
Clovis Ca
Are you sure about that Dave as that is 5 1/2% .

12.1875 divided by 12 = 1.015625 near enough 1 1/2%

For a bit of home casting you can print out a modified rule and stick it to a strip of wood or metal, cover in clear tape.
I put 1 in place

Thank you
 

GreenTwin

Well-Known Member
HMEM Supporting Member
Joined
Jul 2, 2021
Messages
1,468
Reaction score
686
Location
MidSouth
When I was making patterns in wood and steel, I would use the copy machine to print out a drawing, and adhere it to the pattern piece with a picutue framing iron-on sheet adhesive.
I used the copy machine to enlarge the print larger by a factor of 1.015.

I generally use between 1.015 and 1.02 as a multiplier, more often use the 1.015 factor, and that has been getting me plenty close enough to my desired cast dimension.

These days, I 3D print most of my patterns, and so I use the 1.015 multiplier in the Prussa slicer program, so that the 3D printed pattern is slightly larger than the final cast part.

rIMG_2256.JPG
rIMG_2259.JPG
 

ddmckee54

Well-Known Member
HMEM Supporting Member
Joined
Jul 8, 2009
Messages
562
Reaction score
135
GreenTwin:

Just so I'm straight on this, when you print with Polycast you scale the model to the 1.015-1.02 range?

Just wondering, because when I print with PLA I have to scale the model to the 1.03-1.05 range to get an accurate model.

Don
 

Bazzer

Well-Known Member
HMEM Supporting Member
Joined
May 14, 2017
Messages
143
Reaction score
38
Location
Southampton, England
GreenTwin:

Just so I'm straight on this, when you print with Polycast you scale the model to the 1.015-1.02 range?

Just wondering, because when I print with PLA I have to scale the model to the 1.03-1.05 range to get an accurate model.

Don
Don
Green Twin did not print Polycast, that was me.

So you are dealing with two shrinkages

1) The 3D print shrinkage, this is determined by printing simple test pieces and measuring those and applying that factor to any model that you create.

2) Then you have the aluminium shrinkage, there was varying advise but something like 1/75 to 1/77 is what I used when making parts for Boeing, Airbus, Chobhams etc. so this shrinkage factor has to be applied over and above what ever you have already grown the model for 3D print shrinkage.

This is all being a bit fussy of course you could just work with no shrinkages but you will end up with slightly undersize parts.

I( hope that helps.
 

Gedeon Spilett

Well-Known Member
Joined
Jul 24, 2010
Messages
194
Reaction score
35
Location
near Paris France
with Cura, it is simple to change the size of the print, and I print my pla models at 102 % to have cast parts in aluminium dead size...
otherwise something around 0.5mm for a 20mm dia is missing on the cast part.
 

GreenTwin

Well-Known Member
HMEM Supporting Member
Joined
Jul 2, 2021
Messages
1,468
Reaction score
686
Location
MidSouth
GreenTwin:

Just so I'm straight on this, when you print with Polycast you scale the model to the 1.015-1.02 range?

Just wondering, because when I print with PLA I have to scale the model to the 1.03-1.05 range to get an accurate model.

Don

Don-

I printed in PLA, and yes, I scaled the model up by 1.015 in the Prusa slicer program before I started the 3D print.
Sometimes the shrinkage is not exactly linear in every direction, but shrinkage is not an exact thing, you just want to get the casting size in the ballpark so that it falls within a size ranges that will work when you start machining it.

I use aluminum 356 alloy, and so perhaps if others are using a different alloy, the shrinkage may be more or less.

It could also be that you are measuring your shrinkage more accurately than I am, such as more precisely on a smaller part?

Anyway, I use a 1.015 multiplier for both gray iron and 356 aluminum, and it seems to work.
If you have more accurate measurements, obviously use whatever multiplier gives you the desired results.

I have seen some use a 1.02 multiplier, but I seldom see a value higher than that used.
It would be simple enough to cast a few test pieces at the approximate length you will be using and check the shrinkage results, but as I mentioned, the shrinkage may not be exactly linear.

Pat J
 

ddmckee54

Well-Known Member
HMEM Supporting Member
Joined
Jul 8, 2009
Messages
562
Reaction score
135
Pat:

So far my 3D printed parts are used directly, not as patterns. My 1.03 scaling factor is just to get the finished part accurate to the dimensions of the 3D model.

I recently purchased a non-working electric jeweler's furnace and got it working. I do intend to cast some small engine parts in the future, and maybe some parts for RC trucks, so my questions are more to find out what works and what doesn't.

Don
 

terryd

Well-Known Member
HMEM Supporting Member
Joined
Jun 15, 2010
Messages
356
Reaction score
188
Location
South Leicestershire, England
I think the 6% figure in your reference refers to the volumetric shrinkage. Linear shrinkage is more like up to 2% for aluminium, 1% for iron, somewhere in between for brasses and bronzes.

A typical paternmakers' rule(r) for cast iron would be have a scale increased by 1/8" per foot.
Sorry to mislead but the shrinkage I referred to should have read 7mm to 10mm per metre depending on the alloy. I got carried away with the mixing of medieval and Enlightenment systems of mensuration.
By the way in another post someone mentioned shrinkage of 3d printing. In my test pieces and general experience of 3d printing , mostly ABS, any shrinkeage is not discernable so is not really a problem and that is on my completely home built, from the ground up, machine . There are stl files for test pieces which can be downloaded from Thingiverse to check your machine.

TerryD:mad:
 

Jasonb

Project of the Month Winner!!!
Project of the Month Winner
Joined
Mar 30, 2008
Messages
3,094
Reaction score
783
Location
Surrey, UK
Now that is going the opposite way, 10mm/m would be 1% which is what most allow for iron.

Aluminium ,your 7mm is half what everyone else here suggests and uses eg 13 to 15mm/m is 1.3 to 1.5 %
 

Bazzer

Well-Known Member
HMEM Supporting Member
Joined
May 14, 2017
Messages
143
Reaction score
38
Location
Southampton, England
Jason we could do with one final statement on aluminium shrinkage, there is so much misleading numbers in this thread as to render it useless.

The shrinkage is really the two figures that we talked about, my figure of 1/77th was on LM25 or L99 which is practically the same with L99 having less iron. The iron is often picked up from the crucible walls and shows in the post cast tests to certify the melt/batch.
 

Bazzer

Well-Known Member
HMEM Supporting Member
Joined
May 14, 2017
Messages
143
Reaction score
38
Location
Southampton, England
By the way in another post someone mentioned shrinkage of 3d printing. In my test pieces and general experience of 3d printing , mostly ABS, any shrinkeage is not discernable so is not really a problem and that is on my completely home built, from the ground up, machine . There are stl files for test pieces which can be downloaded from Thingiverse to check your machine.
Terry
This is quite misleading, pretty well any material that is heated to form it and then cools down to it's finished state will shrink.

The point you make about having STL test pieces to check the machine and part kind of tell you that this is the case. Let's be clear PLA SHRINKS AFTER COOLING.

How much and if this important is another discussion.
 

Jasonb

Project of the Month Winner!!!
Project of the Month Winner
Joined
Mar 30, 2008
Messages
3,094
Reaction score
783
Location
Surrey, UK
Bazzer, Most of what has been said in this thread agrees, it's just one persons posts that are wrong, hopefully people can sort the wheat from the chaff and not go by that persons latest post which despite their earlier posts that iron has a much smaller shrinkage than aluminium now posts figures that say aluminium shrinks less than iron:rolleyes:

If you want a statement then the linear shrinkage allowance for aluminium alloys could be taken as follows, treat this as your starting point and do some test castings if the final cast size is critical.

Shrinkage as a fraction 1/77 to 1/75
As a percentage 1.3 to 1.5%
As an imperial measurement 5/32"/ ft to 3/16" / ft
As a metric measurement 13 to 15mm per meter

This is how much larger the pattern should be to allow for the metal to shrink to the desired final size.
 

Bazzer

Well-Known Member
HMEM Supporting Member
Joined
May 14, 2017
Messages
143
Reaction score
38
Location
Southampton, England
Bazzer, Most of what has been said in this thread agrees, it's just one persons posts that are wrong, hopefully people can sort the wheat from the chaff and not go by that persons latest post which despite their earlier posts that iron has a much smaller shrinkage than aluminium now posts figures that say aluminium shrinks less than iron:rolleyes:

If you want a statement then the linear shrinkage allowance for aluminium alloys could be taken as follows, treat this as your starting point and do some test castings if the final cast size is critical.

Shrinkage as a fraction 1/77 to 1/75
As a percentage 1.3 to 1.5%
As an imperial measurement 5/32"/ ft to 3/16" / ft
As a metric measurement 13 to 15mm per meter

This is how much larger the pattern should be to allow for the metal to shrink to the desired final size.
Jason
That is a very good rounding up of the thread, thanks.
B.
 
Top