Inspecting for Porosity with a Microscope

Discussion in 'Sand Casting' started by oldironfarmer, Mar 30, 2019.

  1. Following along the bifilm thread and trying different things made me realize I don't know how much porosity I'm getting. Sanding and filing was showing some but I realize without a microscope you'll never see it. I read where 1,000X magnification may be appropriate, and I found a cheap ($16) microscope which claimed to do 1,600X. Plus, it is all electronic and uses a computer screen for a monitor.

    I received it today and looked at three pieces. I don't have polishing paste yet so I had to rely on sandpaper i had, 1,200 grit.

    Left is an engine box section, center is a solid suspension component, and right is a manifold I cast using a square pouring basin and inlet riser.


    Here's what the gate and riser looked like on the manifold


    The microscope came with a clear standard or calibration card. Here are three pictures against my table cloth. The standard does have specks all over it, this is the 0.03 MM line

    0.03 MM.jpg

    0.05 MM

    .05 MM.jpg

    0.07 MM

    .07 MM.jpg

    I think the 0.07 MM image is a little more magnification. I intended to not change the setting but think I did looking at the picture.

    The box section has lots of porosity. Here is one close to the edge, with the 0.07 MM line

    A Inclusion with .07 MM line.jpg

    The suspension part seems clean, here are some of it's inclusions

    B Inclusions.jpg

    The dimple in the center of the top of the manifold part is still visible


    Here it is under magnification, it physically measures about 1-1/2 MM.

    C Top Center 1.5 MM Inclusion.jpg

    Here are some other inclusions on the top of the manifold. This was at the top of the casting.

    C Top Inclusion.jpg

    The inlet end of the manifold was pretty clean, compared to the commercial castings. Here is one inclusion I found.

    C End Inclusion.jpg

    I hope to get some better pictures when I get the polishing compound. I was pretty please with these. I think the table cloth photos show the camera does a fair job. Comments?
    Al Puddle likes this.
  2. Al Puddle

    Al Puddle Silver

    Nice tool. It should be helpful to asses the quality of your castings.
  3. It'll be good to get some consistent hard data with the 'scope, wonder what a cracked open casting looks like under magnification. An interesting effect I was shown is that the freshly broken aluminium surface smells a bit like garlic for a few minutes until it oxidizes. I've also seen casting so porous you could put your lips on them and blow air through them, so your casting isn't too bad porosity wise.
  4. Zapins

    Zapins Gold

    What do the commercial castings look like under the scope? A side by side with yours would be interesting.

    What a cool idea.

    What model microscope did you get?
  5. Now I just need to get some samples polished. My goal is to try different gating systems and potentially determine which make a difference in porosity.

    I'll try some broken cast iron. It will have to be a pretty clean break to stay within the focal length.

    I noticed what smells like urine to me when I grind aluminum on my disc sander.

    I was pleased my random samples were as good as one of the commercial castings.

    In the post above the first picture is a commercial casting, as is the second. I could see the porosity in the first (box section) but could not see any with the naked eye in the suspension part (second picture).

    Here's the box section and inlet to my casting side by side (top bottom):

    A Inclusion Near Surface.jpg C End Inclusion.jpg

    I think finer grinding is going to show more small porosities.

    This is what I bought.

    Pretty pricey. But very easy to use. Supposed to focus from 1x to 1,600x but at lower magnifications the focal point is inside the plastic guard on front. It does not seem removable, so I'm going to cut it back. I bought another just in case I screw it up.

    The stand it comes with is pretty flimsy so I was going to build a stand with an X-Y table so the sample can be maneuvered easily to scan the whole surface of a part. I figured out that stand can be my milling machine, affix the camera to the spindle and use the DRO to record what part of the sample I'm looking at.
  6. ESC

    ESC Silver Banner Member

    It will be hard to focus when it is spinning.
    Al Puddle likes this.
  7. DavidF

    DavidF Administrator Staff Member Banner Member

    screw it. I ordered one too...
  8. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    You need to quit pissing on your scrap pile Andy. :eek:

  9. I'm thinking slip rings will be necessary to maintain a connection to the computer.

    :D It's quite a major investment.

    This is kind of embarrassing, Kelly. I live in the country and people are pretty resourceful. We all mark our own stuff so nobody else gets confused and takes it. Anybody can grind off a stamp or rub out a mark, but your personal scent is indelible. If you don't have time to mark something properly, piss on it.
    Mark's castings likes this.
  10. CLR

    CLR Copper

    I am pleasantly surprised by the image quality considering the low price of the microscope :)
  11. Well this was interesting. I like Melterskelter's full capacity pouring basin so tried it with aluminum. I routinely melt scrap, and have some finer sandpaper now, and diamond paste so I can begin looking at what I do. some of my hammer heads would have big cavities in the face where the gate enters.

    Small runners, open riser straight through from the sprue, and a large orange blind riser (it's what I had laying there, it's an old pouring basin).


    Pouring basin should hold all the pour, 6 cubic inches


    Short video of the pour

    Somehow I couldn't bring myself to stop pouring when the basin got full. I'll try harder.


    This porosity really did not show up before significant sanding. Six months ago I would have claimed it was clean.:D


    Here's my fixture for holding the microscope.


    And here is the picture from the screen.

    Thu Apr 04 18-17-01.jpg

    It's easy to measure stuff to 0.0005", just scroll a cavity to the edge of the screen and zero the DRO, then scroll to the other edge and note how far you've moved. It's at least accurate to 0.001". The DRO is very repeatable with the image on the screen.

    I don't really know what I'm looking at. Maybe someone can help. I know I still need to do a better job of polishing. This is typical. 0.0115 " wide by 0.0165" high.

    0.0115 by 0.0165.jpg

    And this, 0.0315" wide

    0.0315 Wide.jpg

    This looks suspect, same scale as above pictures.

    Fri Apr 05 18-59-35.jpg

    And this, it may look a lot worse with more polishing

    Fri Apr 05 18-55-43.jpg

    Or is that just grain boundaries (except for the little sperm in the middle), or is the magnification too low to see aluminum grain boundaries?
    Melterskelter likes this.
  12. That's certainly a lot of tiny voids on the round casting in the vise, are you using a mystery scrap mix melted a few times in a rusty steel crucible for hours on end?. :cool:

    Edit: I took an ingot of wheelium (601 alloy?) that was cast into a cold steel mould. I'd already sawn a corner out to inspect for defects and had patted myself on the back for such a high quality homogenous casting when checking the sawn area.......these photos are the same spot freshly filed and sanded with 80 and 280 grit to remove any smearing of the material :eek::oops:. This was melted in under an hour in a clay graphite crucible and skimmed a fair bit before pouring. Cooling would have been quick in a steel mould.

    The photos were taken by holding a magnification loupe over the camera lens while it was in super macro mode.

    sawn polished ingot 1.jpg

    sawn polished ingot 4.jpg

    sawn polished ingot 2.jpg

    sawn polished ingot 3.jpg
    Last edited: Apr 5, 2019
  13. Zapins

    Zapins Gold

    This makes me not want to cast aluminum. Is this all just gas dissolving in then being forced out as it freezes? Are these samples all degassed?

    Does bronze do this too? I can't say I've ever noticed tiny pits like this on sculptures even with high gloss finishes but maybe I'm missing them?
  14. I suspect in my case it's shrinkage caused by fast cooling of the metal in a steel ingot mould. Bronze will do it too under the same circumstances, you just have to have the casting solidify faster than the runner/feeder.
    Last edited: Apr 6, 2019
  15. No degassing in my sample. The crucible feed was a lot of swarf and window frames. I missed one steel screw so the crucible was boiling from the steel in it. I think that had to be a contributing factor.

    If you're seeing that at 280 grit you'll likely find lots more at finer grit. Mine was power sanded with worn 80 grit, then wet hand sanded on glass with 220, 500, 1,200, 2,000, 2,500, and finally 5,000 grit. Then stroked on 10 micron diamonds, 5, and slow power buffed with 2.5 micron diamonds. You can see mine needs more intermediate grits to get rid of all previous sanding marks.

    My intention is to test same again with remelted ingots (and keep the steel out), then try eutectic salt degassing, then clean commercial cast stock. I'm in process of polishing an automotive casting.
  16. Zapins

    Zapins Gold

    Man you guys are taking one for the team for sure. Lot of work involved in this, but very interesting. Will definitely help guide others when melting aluminum.

    I wonder if there are porosities in bronze too or if it is just an issue with aluminum?

  17. You probably had a fair bit of oxides in the melt with swarf to begin with. I should have up to 3000 grit and some cerium oxide <5 um glass polish to refine the aluminium surface with....also a few bronze feeders I can section and polish: they look pretty good when machined.

    Hmm I may have to cut through the middle of the ingot, the last bit to solidify and see what it looks like, the ingot is sunken in the middle from shrinkage.
  18. I don't know what causes these pits, but I am surprised they are not mostly round. It supports the bifilm theory that defects are folded oxide layers which separate with gas collection. The key to that is a folded but tight bifilm layer still is a crack with no adhesion so it really does not matter whether there is gas to open it up or not.

    I suspect that in polishing your sculptures you are smearing the metal and making a slick surface. But using high quality ingots to start with and good melting and pouring practices helps. Bronze may not oxidize as rapidly as aluminum. Aluminum stays bright because of a molecular layer of oxide which seals the surface.

    It will be interesting to see whether you have the same defects, but smaller, in areas of slow cooling.

    I can't see how bronze is much different than brass, and I definitely see porosity in brass. I think it is common to melted materials refrozen. Even water has gas bubbles under the right circumstances as it forms ice.

    Yes, this melt was a mess, lots of dry crap in the bottom of the crucible after pouring. Actually the sample is pretty clean considering what i started with. I wanted to start at one end of the spectrum, but using a pouring basin.

    It would be very interesting to see what you come up with.
  19. Al Puddle

    Al Puddle Silver

    I see your riser at the end of a runner and your shrink bob before the part(hammer head). I keep thinking the riser and maybe the shrink bob should be after the part in the flow path. This how I implemented that thought. There is a slight shrinkage at the top of the block. I think the feed to the riser might have froze to soon.


    My reasoning is the part should solidify from top to bottom forcing out gasses during the dendrite forming process. These gases will go into the flow of the melt at the bottom of the part and be swept towards the riser.

    Attached Files:

    Last edited: Apr 6, 2019
  20. Melterskelter

    Melterskelter Gold Banner Member

    It looks like the last part to freeze in your setup was the large block and the last portion of that part to freeze was the center. There was no place for the block to get molten metal after the runner and large thin flat portion froze. So, it had to collapse and suck down in the middle. A good-sized riser closely located at the entry point into the block would slow freezing there especially if the entry point were on the long side of the block and if it were located in the center (not on a corner). The center of the block might still be last to freeze though. Turning the block on edge and having the riser parallel to the block flat surface and centrally located with a centrally located edge gate would be pretty sure to keep enough heat on the mid portion of the block to ensure that the mid portion of the block would stay warm enough long enough to allow good feeding.

    Just thoughts...


Share This Page