A thin hotface medium/low mass beer keg (drum size) furnace for melting Iron

Discussion in 'Furnaces and their construction' started by PatJ, Aug 24, 2017.

  1. PatJ

    PatJ Silver

    Video from the dashcam is here:

    This is the 2nd iron pour with the new furnace, but the first iron pour using the new domed lid, and the first iron pour at 2.75 gal/hr.

    The domed lid worked very well as did the furnace, and the iron was extremely hot when I poured it.
     
    Last edited: Nov 30, 2018
  2. Melterskelter

    Melterskelter Gold Banner Member

    Way to go, PatJ!

    Great results all around! I can see why you were ecstatic.

    Denis
     
  3. PatJ

    PatJ Silver

    Video of the same pour as above taken from a cell phone is here:



    Thanks Denis, it really went extremely well.
    I am glad it all finally came together.
    Successfully pouring cast iron has finally been demystified for me.
    I can finally change the focus back to working on patterns, and start casting things in iron now.
    That has been the dream for many years.

    .
     
    Last edited: Nov 14, 2018
  4. That's fantastic!, congratulations on getting a working furnace system together. Now the trick is to make some stuff your wife needs for her hobbies...or even cast some wax sculptures she has made :D.
     
  5. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Congrats Pat. Looks like all the hard work paid off.

    Best,
    Kelly
     
  6. DavidF

    DavidF Administrator Staff Member Banner Member

    Well...its about damn time ;):D:D
    How did the furnace hold up??
     
  7. PatJ

    PatJ Silver

    Thanks much Kelly and Mark.
    I really appreciate all the encouragement I get here; that helps a lot with motivation.

    You can say that again.
    I am going to write up what I did right (very little) and wrong in the past, so perhaps others can avoid some or all of my numerous blunders.
    There are definitely some tricks to melting iron that one needs to be aware of.

    The furnace looked much better today than I expected.
    The interior of the furnace except for the inside face of the lid were coated with ITC100, and the coated surfaces looked pretty pristine, and almost unused.

    I forgot to coat the inside of the lid and the new plinth, and the surface of those was rough with tiny pits.
    I will coat those surfaces too.

    The ITC seems to reflect heat back into the furnace, but more importantly it seems to slow down the degradation of the refractory significantly.
    ITC100 is not cheap, but sort of a pay-me-now-or-pay-me-later thing.
    I think the mix is two parts ITC to one part water (check me on that), and a thin coat will stay in place. A mix of ITC that is thicker than that will crack off.
    The ITC/water mix is generally sprayed on with a cheap sandblaster.
    ITC can dry out completely too, but don't throw it out, put it in a heavy plastic bag and beat it with a hammer until it is back to a powder, and then add a little water to rejuvenate it.

    The crucible on one side (outside) looked unused, and the exterior glaze is in perfect condition after perhaps four iron pours.
    The other side is pitted from slag that ran over the side.
    Slag is highly corrosive.

    The stainless needles near the surface of the refractory can be seen as black lines.
    Eventually the entire interior of the furnace will be coated with slag splatter.

    I am pretty impressed with how 1" of refractory has held up.
    We will see if it continues to hold up, but so far the signs are good.

    The ceramic blanket in the last photo is what I added during the melt to stop the lead at the lid in that location.
    I will cut the IFB's down a bit, and I think that will stop the heat from going up onto the lid shell.

    Edit:
    I think when the furnace is cool, I will reach inside while the lid is closed, and use a feeler gauge at the hot face/lid joint, to make sure the entire lid is seating completely.
    If the lid is not totally seating, that could be the cause of the leakage.

    rImg_1980.jpg


    rImg_1988.jpg


    rImg_1993.jpg

    rImg_1995.jpg


    rImg_1996.jpg


    rImg_1997.jpg


    rImg_1999.jpg


    rImg_2007.jpg


    rImg_2009.jpg


    rImg_2010.jpg
     
    Last edited: Oct 3, 2018
  8. PatJ

    PatJ Silver

    Here are photos of the casting in sunlight.

    I think there was a little too much water in the sand, and I think the pour was far above optimum pour temperature.

    rImg_1969.jpg


    rImg_1973.jpg


    rImg_1974.jpg


    rImg_1976.jpg
     
  9. PatJ

    PatJ Silver

    By comparison, this is a similar flywheel cast in resin-bonded sand, in iron.
    The sand used was commercial foundry sand, OK85.

    Resin-bound sand basically gives you a high quality finish like Petrobond.
    I did try Petrobond with iron, but the iron temperature eroded the sand badly, and when I opened the mold too soon, the whole affair burst into a big fireball.

    I have been told by the Metal Museum folks that I can get almost as good a finish with sodium silicate bound sand as with resin-bound sand, but I have not tried ss with iron yet.

    rIMG_9867.jpg
     
  10. Melterskelter

    Melterskelter Gold Banner Member

    The resin-bound sand molded flywheel looks to have a very fine finish. But, before you give up on green sand having a good finish, try adding 4% coal dust or ground coal. That will result in a much better finish than you got. I think that even with very hot iron the green sand with coal would have produced a pretty nice finish with no burn-on. The resin bound sand must itself must act as a gas producer much as coal produces gas. The manufacturers of the resin have, over the years, surely tweaked the formulation with that in mind.

    Main thing is that now that you are confident in being able to melt iron, you can start to refine your results by making adjustments to your sand, alloy, and pouring conditions.

    One good indication that you are getting to mid-2500F melt temperature (2550) is that you will see that the iron does not skin over but rather remains shiny after you skim it. At lower temps the iron will, over the course of a few seconds, form a filmy thin layer of matt-finish "scum" starting from the area first skimmed and progressing to the area last skimmed. When you get to about 1550 the pool of molten metal retains a mirror surface and I also see not boiling as in bubbling but a roiling turbulence in the pool. You have to look for the turbulence as it is not totally obvious.

    Denis
     
  11. Congratulations! I never questioned you.:rolleyes:

    You might try a roll of ceramic fiber all around your lid, away from the hot face, to act as a gasket. After the lid has set it, spray on rigidizer. It should be cool enough there to last if it seals the gas from escaping.
     
  12. PatJ

    PatJ Silver

    LOL, well I questioned me (more than once), but I got over it ("I got better" as they say in the Monty Python movie).

    After watching porositymaster melt iron over and over again at the Soule show last year, I knew it was just a matter of time before I figured it out.
    He made it look so easy and very routine, but he was using a rather exotic looking Ursutz burner, so I thought perhaps that was the magic that was making it so easy for him.

    Its not magic; it basically boils down to having the right refractory and crucible, and knowing how to tune a burner (regardless of what type burner is used).

    If you don't know how to tune a burner for maximum heat (I didn't), then you are probably not going to melt iron, or in my case are not going to consistently melt iron.

    .
     
    Last edited: Oct 4, 2018
  13. PatJ

    PatJ Silver

    How to Melt and Pour Iron with Consistent and Predictable Results:

    I am going to try and summarize the items that kept me from melting iron consistently.
    I had a few successful iron melts in previous years, but they were basically accidental successes caused by me blundering randomly upon the right combination of correct inputs.
    The problem I had in the past was that I could not repeat the iron melts in a predictable fashion, and never knew if my iron would reach pour temperature or not.

    1. Learn how to tune your burner, whether it be propane, siphon-nozzle, drip-style, Urtuz, etc.
    You will need a combustion air blower to melt iron, regardless of whether you use propane or oil. The combustion air blower does not need more CFM output than a standard shop vac.
    The least amount of combustion air possible should be introduced into the furnace to produce the desired amount of heat, since an excessive combustion gas swirl in the furnace lifts slag out of the crucible and splatters it on the walls and interior lid surface.

    Learning how to correctly tune your burner is #1 in importance in my opinion, since nothing else really matters if your burner does not produce enough heat to melt the iron (and you are trying to melt iron).
    I tried fuel (diesel) flow rates from 2 to 6 (+) gallons/hr, and had made the assumption that more fuel and air must be better.
    I believe that fuel flow rates higher than 3 gal/hr in a 13" dia X 14" tall (interior dimensions) furnace actually cool the furnace.

    How did I figure out how to tune my burner?
    I ran it at night and tried a variety of fuel flows, and carefully observed how intense the radiant heat was.
    Note that large flames are very misleading, and don't necessarily indicated maximum heat.
    The maximum heat is produced when there are little or no flames produced inside or outside of the furnace.
    I did allow for slight flames out the lid of the furnace (a slightly rich mixture, or slightly excess amount of fuel), as you can see from the videos above, but the night time video makes the flames look much larger and more intense than they would appear in the daylight.

    A siphon nozzle style burner which uses a commercial spray nozzle such as a Delavan or Hago, and using a fuel tank with approximately 12 psi pressure (with a 30 psi safety valve), will start easily at any temperature on diesel only, and will not require adjustment during the melt (assuming the burner is constructed correctly).
    I use a ball valve for quick on-off control, and a separate needle valve for fine control.
    I initially calibrate the needle valve using a measuring cup and stop watch, and using fuel tank pressure only, before the first melt, so I can start at a known fuel flow rate.
    I do not change the needle valve setting after this initial calibration, but rather make slight adjustments to the combustion air flow valve to produce a slightly rich burn (with some flames out the lid, but not excessive flames out the lid).

    I use clean diesel as a fuel, and run it through an in-line automotive fuel filter.
    A siphon nozzle burner will not tolerate any foreign material in the fuel flow.
    If waste oil is used, it must be screened and filtered to produce fuel that is free from any foreign material, and it must not contain water.
    I have heard reports that hydraulic cylinder oil should never be used with an oil burner.
    One or more spin-on automotive fuel filters can be used if the waste oil needs more filtering that a small fuel filter can provide.

    2. Keep your burner mass as low as possible/practical.
    My first furnace had approximately 500 lbs of refractory in it, and I used the same mentality on the furnace as I did the burner, ie: bigger must be better.
    A heavy mass furnace will actually melt iron, but the melt time to reach pour temperature is increased by potentially 30 minutes or more, since the mass has to reach a high temperature before the iron will begin to melt.
    There is a compromise between melt speed and durability.
    A heavier furnace will probably be more durable if it is constructed correctly with the correct materials.
    A lighter furnace will reach pour temperature faster.

    3. Elevate the crucible in the furnace as high as possible.
    I can't prove that this is important with melting iron, but I have observed several successful iron guys elevating their crucibles on fairly tall plinths, and mentioning that this is important for hot iron.
    I have also heard reports of tall pinths failing in mid-melt, and letting the crucible fall over against the inside of the furnace (a very precarious situation), so use very high quality material when making plinths, and discard them if they begin to crack.
    I elevated my #10 crucible to within perhaps 1" or 1.5" of the bottom of the lid.
    This elevated position also helps when you are trying to add additional metal to the crucible, as well as helping when you are skimming slag (if you skim slag with the lid closed).

    4. Don't try to melt thick pieces of iron until you have the iron-melting procedure mastered.
    One of my mistakes was to use very thick pieces in the crucible, perhaps 3" thick, and this makes for a very slow melt (especially the initial starting melt), and a slow melt causes a lot of metal oxidation and associated slag production.
    If you use thinner pieces of scrap, pehaps 1/4" to 1/2" thick maximum, then you can establish a molten pool not too long after the burner is started, and then each additional scrap piece can be added and pushed through the slag cover and submersed under the pool of molten iron.
    Thin pieces of scrap iron have a lot more surface area than thick pieces of iron, and they melt much faster than thicker pieces of iron.
    If you use heavier pieces of scrap, add them later in the melt when a substantial molten pool has been formed.

    5. Don't skim the slag off of the iron until the end of the melt.
    The slag provides a cover over the iron that protects the metal and minimizes oxidation.
    If you do not add additional metal during the melt (and in the process break through the slag cover with each additional piece), then the slag on top the melt will thicken and eventually get very hard, and it can be difficult to break through the slag cap, and very difficult if not practically impossible to skim at the end of the melt.

    One of the biggest problems for me with iron was understanding what was happening with the slag.
    Sometimes, I would not add additional metal during the melt, and the slag would get hard on top the melt, and I would make the false assumption that the melt was going cold, and then start adjusting the burner, and inadvertently adjust the burner from the correct setting to the wrong setting.
    What was actually happening was that the molten metal was hidden beneath the slag cover, and it was at pour temperature, but I did not realize what was happening.
    Once the slag cover hardens, it gets dark in color, mimicking the look of a melt that has turned solid.

    It is essential to get an initial pool of metal started as soon as possible after starting the melt, and then pressing each additional added piece completely through the slag and beneath the surface of the molten metal.

    One online source mentions using borax as a flux to help break up the slag.
    Borax is not necessary at all if the above method is used to add metal during the melt.
    The slag at the end of the melt (using the above procedure) is not hard, and is relatively easy to skim, although it is much thicker than the slag seen on the top of aluminum melts.
    Borax also attacks the surface of a clay-graphite crucible, and can seriously reduce crucible life, so I don't recommend using it; it is not necessary.

    6. Seal any cracks in the furnace walls or lid as soon as possible with a high temperature sealant (ITC 100 or 200 is a high temperature material that can be used), or a fine refractory/water mix.
    Be sure to dry your refractory out completely using an appropriate dry-out schedule.
    Do not spray on a coat of ITC100 before your refractory is completely dry.
    Fire each layer of ITC100 (if you use this or similar material) before adding a second coat (if you choose to use two layers of ITC).

    The interior of the furnace is pressurized, and any leaks will produce a jet stream of hot gases that can degrade or melt exterior shells, burner tubes, etc.
    Be sure the burner tube fits snugly into the tuyere opening, so that hot gases to not blow past the burner tube and overheat it.
    Introducing excess combustion air into a furnace will exacerbate any leaks, so run your burner and blower using as little combustion air as possible/practical.
     
    Last edited: Oct 4, 2018
  14. PatJ

    PatJ Silver

    How to Melt and Pour Iron with Consistent and Predictable Results (continued):

    7. If you need to machine the iron that you cast, you will probably want to add ferrosilicon.
    The typical ferrosilicon I see for sale is 75%.
    The appropriate amount of ferrosilicon to add to a melt appears to be about 2% or slightly less (example: add 8 oz of ferrosilicon to 25 lbs of iron).
    Adding excessive ferrosilicon will weaken the casting.

    The ferrosilicon must be added just before pouring the metal, since its effects only last a few minutes.
    The ferrosilicon seems to keep the graphite in solution while the metal is hardening, thus preventing hard spots from forming in thin sections of the casting.
    For thicker castings, perhaps 1.5" or thicker, ferrosilicon may not be needed.

    Ferrosilicon also seems to work as a flux with iron, and the fluidity appears to increase significantly after ferrosilicate is added.
    I do not envelope the ferrosilicon in aluminum foil since this contaminates the melt, and also prevents the ferrosilicon from mixing with the iron.
    I stir the ferrosilicon into the iron with a graphite stirring rod.
    The ferrosilicon and graphite rod must be free of any moisture.

    8. I use a Morgan "Salamander-Super" clay graphite crucible, which is rated for about 2,900 F.
    There are multiple Morgan crucible types with the "Salamander" label, so pay attention to the "clay graphite" detail.
    The Morgan silicon carbide crucibles have a much lower temperature rating, and often this temperature rating is lower than the pour temperature of iron, thus some silicon carbide crucibles may have a short life if used with iron.

    Tap lightly on the crucible before each pour, and make sure it has a ringing sound. A dull thud sound can indicate a cracked crucible, which should be immediately discarded.

    Be sure to temper the crucible before first use according to the manufacturer's written tempering instructions. Note that the instructions vary with crucible types.

    Generally a crucible should be contacted with the lifting tongs on the lower half of the crucible.
    Don't try to lift a crucible from the top half.
    I use a mechanical stop mechanism on my lifting tongs to positively limit the closing position of the tongs and prevent placing excessive force on the sides of the crucible.
    Don't stack crucibles during storage, and keep them very dry.

    9. When initially charging the unheated crucible, don't let the initial metal charge protrude above the top of the crucible, or if so, only slightly, to prevent oxidation of the iron.

    Don't jam the metal into the crucible when adding an initial charge, since the crucible can crack when the metal expands.

    I use a layer of heavy cardboard under the crucible to prevent it from sticking to the plinth.
    For my last melt, I also added a layer of powdered graphite to the top of the plinth.
    Any slag spills that run down the side of the crucible and contact the plinth will cause the crucible to adhere to the plinth.

    When adding iron after a melt has begun, the scrap should be preheated to drive off any moisture that may be in the metal.
    I do this by suspending each piece of metal in the exhaust stream for approximately 15 seconds, or until any moisture is driven off. Adding metal that has even a slight (non-visible) amount of moister in it to a molten iron bath can cause an explosion.

    If possible, clean iron should be used for a melt, but I have seen extremely rusty iron used for melts with no bad effects. I use iron that has a thick coat of paint on two surfaces, and that also does not seem to have any bad effects, and does not cause excessive slag formation.

    10. Class 20 and 30 gray iron has a lower melting point than the higher Classes of iron, and a lower melting point than steel or alloyed cast iron.
    I use Class 40 gray cast iron, and did not have any problems with it reaching pour temperature last night.

    11. Glasses that are certified to provide IR and UV protection with high temperature metals should be used at all times around iron melts due to the high amount of radiant energy produced, and a good face shield/hardhat combination should also be use, along with full leathers, and leather boots.
    Leather gloves work ok with iron, but can easily overheat if you do not have a heat shield on your pouring shank.
    Metalized foundry gloves provide more protection to the hands than leather gloves.
    Slip-on metalized heat shields can be added over leather gloves to give additional heat protection.
     
    Last edited: Oct 4, 2018
  15. PatJ

    PatJ Silver

    How to Melt and Pour Iron with Consistent and Predictable Results (continued):


    12. A pouring shank for iron is best if it includes a heat shield to protect the gloved hand that is nearest the crucible.
    It is absolutely essential that the shank include a functional crucible retainer mechanism that works well with crucibles that very in size even for the same crucible number. The retainer should hold the crucible in the pouring shank even when the crucible is upside down and shaken vigorously.
    A crucible full or iron will place a great deal of force on the retainer mechanism, so it should be strong enough to retain the crucible when the crucible contains it maximum rated amount of iron.

    I prefer a pouring shank with a straight shaft.
    A bent-shaft pouring shank can add a moment (a twisting force) which can make it difficult to handle and pour iron.
    I also try to keep the center of mass of the crucible slightly lower than the centerline of the pouring shank, so the weight is balanced and the forces neutral throughout the pour, and a twisting force is not produced as the crucible is emptied.

    I do not use combination lifting tongs and pouring shank with iron because I don't feel they are reliable or easy to use with the high mass of iron (applies to perhaps a #10 crucible and larger).

    I prefer a handle mounted on the pouring shank that is within about 6" of the crucible (with a sheet metal heat shield), and I weld the handle in place vertically on the bottom of the shank shaft so that the wrist is in a comfortable and strong position throughout the pour.

    The shank handle should not require you to put your hand, wrist or arm into an awkward, elevated or difficult position during the melt, since it is important to have a steady and consistent pour that keeps the sprue full at all times (to prevent entraining air into the molten metal and into the mold).

    .
     
    Last edited: Oct 4, 2018
  16. PatJ

    PatJ Silver

    View looking into the lid opening for the 2nd iron pour with the new furnace:



    This video shows the burner running slightly rich at 2.75 gal/hr.
    The heat is causing the entire interior of the furnace to glow red, including the entire interior of the chimney.

    A setting that does not cause the entire furnace interior (and chimney if you have one) to glow bright red is not the correct burner setting.

    .
     
    Last edited: Nov 30, 2018
  17. PatJ

    PatJ Silver

    Here is a test of the cement mixer/muller.

    I departed from the typical mixer/muller build which generally discard the top half of the drum and use one or more wheels, and used a 70 lb iron weight wrapped in plastic instead.

    I wanted to use the entire drum to maximize the mulling capacity, and so I retained the top half of the drum.
    I think I had about 200 lbs of sand in the mixer (it was three 5 gallon buckets of mulled sand).
    I did not install the interior tines, and covered the tine mounting holes with aluminum tape.

    I was not sure if I would get any mulling action with this setup or not, but it seemed to work pretty well.
    I temporarily used a boat paddle as a scraper, and as kludgy as a paddle would seem, it was a simple and half decent solution.

    I think I got the sand slightly too wet, but it did have some good green strength.

    The mixer/muller test video is here:




    .
     
    Last edited: Nov 14, 2018
    joe yard and Tobho Mott like this.
  18. Melterskelter

    Melterskelter Gold Banner Member

    Comments to throw into the mix:
    1). This is a very nice summary of a lot of information gleaned from many sources. Well done, PatJ. It should help a lot of people getting started to have success far sooner than if they were trying to sort all this on their own.

    2). I will add information I have gotten through my own limited experience and from a very experienced commercial iron foundry owner/operator. I generally agree with this comment but would temper it some:"Thin pieces of scrap iron have a lot more surface area than thick pieces of iron, and they melt much faster than thicker pieces of iron.
    If you use heavier pieces of scrap, add them later in the melt when a substantial molten pool has been formed." For some iron melts I have used a large proportion of fins and other small thin bits as charge stock. Unfortunately they produce a lot of slag and not much good metal. I have not noted significantly shorter melt times with thin parts. My preferred charge is as dense as possible with say 1 to 3 inch square or diameter scrap set into the crucible and the spaces filled with whatever can be fit in---thin, short whatever.

    3). Ferrosilicon is depleted in iron when it is melted and should be replaced if you want to avoid "chilled" iron and poor flow characteristics. I found this out the hard way by very time consuming and frustrating experience and then found research articles speaking to the issue.

    http://lejpt.academicdirect.org/A26/059_064.pdf

    Finally, yesterday, I had the opportunity to measure Si content of samples of melts and remelts (and some 100-year-old Stanley plane bodies) using a spectrometer. Those measurements clearly showed diminished Si and Manganese in remelted iron. And, even though the foundrymen at Stanley had never heard the word "spectrometer," they knew enough to add it to their very thin and beautifully cast planes bodies as the concentration in bit the plane bodies cast 50 years apart had levels of Si just above the spec for 30 grey iron. How much to add is a good question, but Pat's suggestion of 2% is about what I have used14 ounces of 75% Si in 50 pounds of CI.

    4). I also have found the recommendation made by experienced oil burner operators to tune the burner so that it is just a bit leaner than smoking and yet with reddish flame a few inches out the chimney is best. It seems counter intuitive as I'd have thought a leaned-out burn would be hottest. But the cooling effect of sufficient air to lean out the mix also cools the furnace too much. I seem to do best if my furnace is not burning so lean that it is free of carbon-just a little dirty around the edges- when melting. I suspect that mix also is a bit of a reducing flame and reduces slag formation.

    FWIW. Once again, hats off to PatJ for the best iron melting summary I know of.

    Denis
     
  19. PatJ

    PatJ Silver

    And hats off to folks like ironman, scavenger (justin in tx), porositymaster (in MS), davidf, melterskelter (dug up and provided much useful info on iron), and many others for the tricks and tips about iron and foundry construction in general.
    My furnace is basically a hybrid of the ideas from people here on the forum and justin on the old forum.

    Its not really my ideas about iron, but rather a conglomeration of what all these folks have told me.
    It is a good example of what can be learned from a foundry group.
    The shared knowledge and experience is very helpful, and often downright inspiring.
    And such a nice forum to share it on, with no photo issues.

    It does not get much better.

    .
     
  20. Tobho Mott

    Tobho Mott Gold Banner Member

    Congrats Pat, the new furnace seems to work great, a thing of beauty! The how-to section above is nice too, and would have made a valuable standalone thread in its own right.

    Your mixer seems to work well - mine (a cut off cement mixer converted to a wheel type muller as described above) only takes about 2/3 of a 5 gallon bucket at a time, tops, for what it's worth. Maybe be it would have had more capacity if it had a pulled wheel instead of a pushed wheel. :(

    So I'm with Denis - don't give up on your re-usable greensand yet. The old timers who made the machines you want to recreate made it work with water and clay, and so can you! If it was too wet this time, just try it a little dryer next time. Your sand looks nice and strong in the mixer video.

    Jeff
     

Share This Page