Porosity problem

Discussion in 'Sand Casting' started by Al Puddle, Nov 9, 2017.

  1. Al Puddle

    Al Puddle Silver

    In the course of making air manifolds, it became evident that I'm having porosity problems. My castings aren't holding air.

    I'm going to work this out so, that's what this thread is about.

    My pattern
    IMG_1074.JPG
    IMG_1075.JPG
    The casting
    IMG_1076.JPG IMG_1077.JPG IMG_1078.JPG
    Saw cuts:
    IMG_1079.JPG IMG_1081.JPG

    After milling:
    IMG_1083.JPG IMG_1084.JPG IMG_1086.JPG

    There seems to be a fairly homogenous interspacing of black dots.

    I didn't flux or try to degas. Pour temp was 650C. Aluminum was A356 and previously cast.
     

    Attached Files:

  2. Negativ3

    Negativ3 Silver

    The final casting you've shown in General foundry chat, mounted below a bench, does it hold air ok?
     
  3. Tobho Mott

    Tobho Mott Gold Banner Member

    Was wondering the same about your finished manifold from the other thread, you mentioned no hissing there, so is it a slow leaker then?

    There's too much conflicting info out there about degassing, at least when it comes to "home gamer" foundry setups IMO. Best bet is probably to avoid letting excess hydrogen dissolve into your melts to begin with, by tuning your burner to slightly oxidizing when melting aluminum and pouring the mold ASAP once up to temp.

    I think Al2O3 has posted some info about sealing porous castings here somewhere as well.

    Jeff
     
  4. Al Puddle

    Al Puddle Silver

    It may be a slow leaker; I didn't dunk it in water to check.

    A slightly oxidizing flame would be blue, correct? I've been placing the Al charge in the cold crucible and then heating till melted. It takes about two minutes to skim dross and pour.
     
  5. Remelting sand cast aluminium can have problems associated with it, I read a description of WW2 production of aircraft engines by Ford where the silicon content was too high in the castings due to the aluminium chemically reacting with sand grains attached to the reused aluminium. The silicon dioxide was reduced to silicon and the aluminium was oxidised, just like a thermite reaction. This is a photo of one of my castings with porosity and the odd grain of sand embedded in the casting. The aluminium was completely reused runners and gates and went into the pot with sand coating attached, unlike denser bronze the contaminants are less likely to float to the surface for skimming. Having fresh unmelted metal and a fast way to melt it goes a long way towards minimising porosity and contaminants.

    porosity.jpg

    Spindle bore 1.jpg
     
  6. Tobho Mott

    Tobho Mott Gold Banner Member

    I don't see blue fire in my furnace with the waste oil burning, so I tune it by increasing the oil until the fire shoots out the vent hole, then turning up the air to make it retreat back into the furnace. In theory that's about neutral, so just a little more air than that would be slightly oxidizing.

    (I try not to think about what the fire shooting out the drain hole might indicate about how my burner is tuned...) :D

    Never heard of that problem with remelting sand cast pieces, that is interesting. I wonder how many times that WWII stuff had been re-melted before it had that effect? Yet another reason (besides just saving sand) to be careful to brush all the sand off old gating before it goes into the ingot bucket, I guess...

    Jeff
     
  7. Al Puddle

    Al Puddle Silver

    Okay, the next step will be to use A356 ingots, no scrap. I'll hold all the other variables constant.
     
    Mark's castings likes this.
  8. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    IMO, porosity in sand castings is just a fact of life and beyond that the discussion is just a matter of degree. There certainly are things that can be done to reduce it but not eliminate it. Design standards account for it in safety factor and mechanical properties can greatly improve when you move away from expendable molds and cast in controlled atmospheres and vacuum. But these things aren't as economical as sand casting without significant scale.

    Solid contaminates can be easily dealt with but the solubility of hydrogen in aluminum and associated hydrogen porosity will be present. For us home foundry guys there are some common sense things that can be done to help.

    1. First thing is to design for it. If you have something that needs to be air tight allow for the thickest wall possible.
    2. Don't melt and remelt your casting stock. If you have automotive castings don't cast them into ingot first and don't reuse your gate and sprue stock for critical castings. Virgin ingot is best.
    3. Minimize pour temp and the time the melt is molten. If you preheat your charge and the furnace/crucible, process-wise you can reduce the amount of time your metal spends in the molten state and can help minimize exposure to H2.
    4. Fluxing and degassing aren't the same thing. Degassing definitely helps but burping some Argon from the end of an open tube n your melt will do little to nothing. You need fine bubbles for significant minutes to attract and remove H2. Pro's use rotary degassers. You should take in OldFoundrman's video on degassing. I'm going to build a variation on his degasser for my own use.

    Even with all of the above and commercial foundry practice, casting porosity is still common, so much so, it is pretty much standard procedure to vacuum impregnate castings in the automotive industry because doing so is both more economical and effective than trying to control all the other things in the casting process that produce the same result.

    If untreated, most (but not all) my automotive water necks will "sweat" antifreeze in use at 180F and 10-15psi internal pressure, meaning they were porous to the degree they would not outright leak but would eventually form little beads of water that would puddle and or produce wet spots on the outside surface of the castings. No surprise. I cast and recast that metal probably 20+ times in lost foam castings reusing gates and sprues and fluxing occasionally to cope with the drossier remelted cup stock. After vacuum impregnation they are absolutely sealed tight.

    This is fine for sealing but the reduction in mechanical properties is still present. These are not at all critically stressed parts, but I'm about to embark on a couple parts that are and for those I will need the best metal quality possible and I will cast them from virgin ingot.

    Best,
    Kelly
     
  9. Two other things that have been taught to me are to scrape the crucible with the skimmer to dislodge dross stuck to the crucible below the surface of the molten metal, so that it floats to the surface and can be skimmed off. The other thing is that casting on a very humid day increases risk of porosity and also wastes significant amounts of fuel being converted to steam which carries the heat out of the furnace.

    While I think of it, the reason using steel crucibles used for melting aluminium is frowned upon, is the layer of rust reacts with the molten aluminium and reduces it to elemental iron that then goes on to contaminate your aluminium while the chemical reaction generates extra aluminium oxide dross that would not otherwise be there. This is the classic thermite mix of iron oxide and aluminium, just not in powder form so not as spectacular.
     
    Last edited: Nov 10, 2017
  10. Al Puddle

    Al Puddle Silver

    Good info. Kelly. This thread is all about collecting our shared knowledge regarding this subject.
    In messing around with this air manifold thing,

    I'm becoming more aware of the machining properties of Al.
    One thing I know for sure is... don't melt Cu into Al.
    Cu does dissolves into Al but not quickly at temperatures less than 1200F. It was completely gone at 1400F.
    The aluminum mix did not machine well. It was gummy with lots of hard particles. Threw that stuff away!
    I think the copper was precipitating homogenously.

    I've also begun wondering if Al oxide mixed in the casting was making it difficult to get a good finish and dulling my cutter.
     
  11. Al Puddle

    Al Puddle Silver

    If you're right, I'm going to have to redesign my whole foundry. What should I be looking for to validate the need to change to a graphite crucible?
     
  12. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Another possibility is just use radiator sealant to seal the interior of the casting.

    The active ingredient in radiator sealer is sodium silicate. It was commonly used for impregnating castings for many years before the modern low viscosity resins were introduced. I bought a heat cured methacrylate which is a very low viscosity resin which makes it more effective and easily driven deep into the porosity sites but the process is basically the same. If you plumb a piece of tube to the manifold with enough volume of radiator sealer to fill the inside of the manifold, ideally you would orient the assembly so the manifold was dry, pull high vacuum, reorient so the manifold becomes wet, then re-pressurize to atmospheric pressure or even apply 10-20 psi of air pressure. The rotate the assembly again so the manifold is dry and apply a little heat with a torch to the exterior of the manifold to cure the sealant. You can also just perform the vacuum pressure cycle with the manifold wetted the whole time but it is slightly less effective.

    This is essentially a down and dirty pour man's version of the vacuum impregnation process and the vacuum pressure cycle drives the sealant into the micro pores.

    Best,
    Kelly
     
  13. Al Puddle

    Al Puddle Silver

    Thanks, Kelly. I've got some SS so, I'll probably give it a go next time I'm craving pressurized air.
     
  14. My foundry guru prefers silicon carbide crucibles to clay graphite, he reckons they last a lot longer. Watching him prepare for a session, he gets the cold crucible and peels that thin layer of aluminium left from the last use off the inside and ensures there's no loose material inside. When the crucible is full and hot, the inside surface gets a light scraping with the skimmer so that the contaminants float to the surface and get removed. He said foundry hygiene is crucial, He's at the point where he buys in virgin ingots and no longer melts scrap engine parts. He's been manufacturing parts for one product for 40 years now and has seen certain castings made from high silicon pistons fail around the 30 year mark due to age hardening from the high silicon: one good bump and the 3/4" thick casting breaks.

    I'd always been told iron contamination in aluminium alloys is a bad thing, and accepted that without knowing why: it forms an inter-metallic compound Al13Fe4 which interferes with precipitation hardening and reduces ductility and machinability.
    Iron contamination of aluminium
    Removal of iron from recycled aluminium

    One particular example of contamination of molten metal that sticks in my mind was a car brass radiator tank soldering machine built by a friend. The molten tin lead solder was held in a 316 stainless steel tank and pumped through 316 stainless pipe using a 316 impeller. The machine worked fine, but over time the solder dissolved the stainless steel until it ate through the tube completely. Stainless relies on a surface layer of oxide for corrosion resistance and the solder prevented this. Modern soldering irons have an iron clad copper tip to prevent the copper dissolving into the solder at temperatures well below the melting point of copper. Some solders like "Savbit" had 2% copper alloyed into it to reduce copper erosion. So we have a chemically reactive metal like aluminium which can reduce metal oxides to raw metal like iron and then dissolve and alloy with them.
     
    Last edited: Nov 10, 2017
  15. Al Puddle

    Al Puddle Silver

    I should have waited till morning to read Marks castings' post. Now, I'm going to have nightmares.
     
    Mark's castings likes this.
  16. JoeC

    JoeC Copper

    Did you ever get this resolved?

    It looks to me like hydrogen gas porosity, not an alloy chemistry problem.

    No way around it but to degas, needed anytime you do thick sections fully machined, and especially if pressure tightness is required

    All you need is a steel pipe and a regulator to slowly bubble Argon or Nitrogen through the melt

    Keep the bubbles as small as possible, you don't want to splash it everywhere, move the pipe allover the bath to degas it all, several minutes at least

    I still have some Strontium and Tibor that may help a bit, although it wont cure gas porosity it would help with machinability

    http://www.alloyavenue.com/vb/showt...iner-(TiBor)-amp-Modifier&highlight=strontium

    Joe
     
  17. Al Puddle

    Al Puddle Silver

    Not exactly. I just finished getting around to try sealing the manifolds I have with sodium silicate. I'm not hearing any hissing but the small manifold is still leaking. I'm calling these good enough for now. However, I'll be getting to the bottom of this using the manifold patterns and setting up a decent test rig for monitoring pressure. Yea, I'm making a science project out of this. There's a lot of distraction going on for me right now so, the going may be slow. I'll be posting as I get there.
     
  18. Tobho Mott

    Tobho Mott Gold Banner Member

    Wow, that's interesting.

    Some people have mentioned adding that high silicon piston alloy to extruded/beer can scrap melts at 1:1 to turn those crappy cans and old ladders into something better for casting, more like A356 wheelium. Sounds like now there's a reason to do the same thing in reverse, ie. adding pop can ingots to pistonium melts in order to dilute the silicon.

    Maybe not in a pro setting like your guru's, and it's certainly not the point you were trying to make... But still, this could come in handy for the backyard hobbyist on a strict budget who is working on their own projects with whatever materials they're able to scrounge...

    Jeff
     
    dennis likes this.
  19. That's a valid plan, although the unknown composition could be a pain. Not sure how true it is: I'd heard car pistons with a high silicon content get a heat treatment every time the engine runs and get up to temperature which is why they can use such alloys. Peter also tells me he had a bar of a particular aluminium alloy where the bar broke in half due to a silicon crystal growing across the width of the bar over 20 years or so: He showed the aluminium company rep the bar and it ended up in their lab as part of a display on defects.

    Luckygen had some of his cast iron made from steel scrap sent for analysis, so it may be possible to find a scrapyard with an Xray gun to sample your scrap for a six pack and give you an idea of what you have. I suppose you could digest some aluminium in an acid and assume the undigested solids are silicon for a ballpark analysis.





     
    Tobho Mott likes this.

Share This Page