New oil burning furnace build

Discussion in 'Furnaces and their construction' started by Mark's castings, Jun 6, 2024.

  1. After my existing furnace went underwater during the December 2023 floods, I thought it was time to start work on a truly iron capable furnace. The existing furnace reaches over 1400 degrees C only under special circumstances: with a faster burning fuel or with a smaller crucible that increases the combustion volume.

    Both methods do work but for normal fuels like sump oil, diesel and Jet A1, the furnace struggles to get hot enough with an A25 crucible in the furnace. The assumption is the furnace chamber volume is too small to burn fuel fast enough to overcome the heat losses via the chamber walls. Geometry wise the shape that encloses the most volume with the least surface area is a sphere, followed by a cylinder which is what I'll use.

    Plugging some numbers into a spreadsheet and using the existing furnace as a starting point, you can see that increasing the bore diameter allows the volume to rapidly increase with modest increases to the surface area. So with an A25 crucible, increasing the bore lets the chamber volume that supports combustion to outpace the surface area (heat loss).

    I'm hoping that the larger volume increases the efficiency of harvesting the heat by allowing the burning gasses to "dwell" for longer in the chamber before exiting: the bottom right figure shows 3.37 times the existing volume so the combustion heat should spend longer in the furnace as a result. The surface area increases by 85 percent so heat losses will increase too but not as much as the hoped for gains.

    furnace bore calcs 1.jpg
    Last edited: Jun 6, 2024
    Tobho Mott and Charlie2Crows like this.
  2. I'll cast an inverted "Top hat" liner with the dense 1600 deg C refractory concrete I've bought and make it around 50mm/2" thick. I'm debating whether to cast it into a stainless steel container like a modified keg as that will contain refractory as it inevitably cracks and also prevent oil vapour leaking into the outer layer of fibre refractory which I'll have on the outside of the dense refractory. I've also noticed the fuel spray cools the refractory where it hits and vaporizes as the refractory is black and not glowing yellow so I'm thinking of a a 150mm/6" bore refractory lined tuyere for a short distance of say 20cm/8" or so. That way the fuel spray can be arranged to hit the mouth of the tuyere and vaporize there rather than in the chamber.
  3. HT1

    HT1 Gold Banner Member

    alot i want to say, but there are fellas here better edumacted, that can probably articulate it better , but just basically, some of your numbers are very suspect Diameter of an A25( Now I know modern Crucibles are all over the place mostly because of the metric system and an A25 and a 25 are different animals so maybe that's the issue),

    Also A Keg will not be big enough shell for an A-25 furnace at 16.125 inch OD best you could get would be about 1.25 inch of refractory with nothing behind it for your 35 Cm bore , a 30 gallon metal drum is a better option at 18.5 inch diameter, but step up to a 55 gallon and two full inched of Kaowool easily ???

    ill leave the heat loss stuff to the College edumacated fellas, but just from experience it is a larger issue then some think

    V/r HT1
  4. Hi Henry, I went and measured a Morgan AHM 25 I have in the shed: it's basically a silicon carbide with extra glaze that makes it more flux resistant or something. Regarding the keg size: you're right, I built an 11" bore in a keg and later grafted half a keg on top of that to get my existing furnace which is the first row of measurements in that chart and the 11" bore is too narrow for the 25 crucible with less than an inch gap each side.... I'd been thinking of cutting up two kegs and joining them to get something closer to a 55 gallon drum in diameter. I guess the point I was trying to prove to myself with the chart is that for modest increases in bore size I can fit a lot more fuel and air mix in the chamber at once.

    I'm using the Carroll Shelby approach: "There ain't no substitute for cubic inches"
    Last edited: Jun 7, 2024
  5. I've added a second section with "Freedom units" to match the metric section, the crucible is assumed to be a cylinder for volume calculations rather than a doubly truncated ellipsoid as I'm lazy. The main number to pay attention to is the last column on the right: going from an 11 bore to almost 20" increases the free chamber volume 3.37 times while the surface area only goes up 1.85 times.

    furnace bore calcs 2.jpg
  6. HT1

    HT1 Gold Banner Member

  7. That agrees with what I have within 10mm for the AHM25. Here's a photo of a Salamander Super A25 from the workshop so 21.2 cm wide by 28.7 cm tall.

    A25 crucible.jpg
  8. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    The first thing I would say is figure out a way to reduce your refractory mass from 150kg. Some years ago, there was a (German fella I believe?) who built a beautiful furnace in a 55 gallon drum. It was solid refractory and weighed > 250 kg. The idea was for it to be indestructible but even with a very high energy burner, he wasn’t getting anywhere near iron temps in 3 ½ hrs. Refractory is more conductive than you think, and it takes one heck of a lot of energy to raise the temp of that refractory mass and until you do, it’s robbing heat from the furnace interior.

    I think you’d be far better off casting a simple cylindrical hot face, say maybe 1” thick, that sits loosely captured in a stepped base and upper annular ring to contain it, then insulate the rest with wool. The simple refractory shapes would be easy to mold and easy to replace.

    I’ve never jumped into much of the combustion melting discussion but it is interesting to me.

    That dwell time part of the discussion was always of interest but I thought it could be reduced to mostly a matter of volumetric flow, active furnace volume, and flame propagation speed. When I say active furnace volume, I really just mean the volume between the crucible and furnace wall. What happens above the crucible is of lesser value for actually getting heat into the melt but in the end everything needs to be heated and a large amount of heat is still lost out the vent.

    If you know that volume, and you know the volumetric flow, you know the dwell time, but the volumetric flow inside the furnace needs to be adjusted for the dramatic expansion caused by combustion. If you adjust the volumetric flow for the expansion that occurs during combustion, it’s easy to approximate the dwell time. If dwell isn’t well above the flame propagation speed x flame travel distance, it won’t be hot. Instead of all the science and guestimating, you already know from examples as to approximately what fuel flow rate makes for a hot burn for a given furnace volume.

    As far as dwell time discussion, doesn’t all that become a bit moot in the case of Ursutz type (external) burners? The issue there is heat loss outside the furnace. So why don’t you make a false floor in the furnace under which will essentially be the firebox ala Ursutz, and make what is traditionally a very cold region of the furnace (the floor), a very hot region?.......and, by design with an insulated furnace, all the heat (except what goes out the vent of course) must go into the furnace and preferably the melt……isn’t that the objective?

    I know we have discussed and you've fiddled with this a bit before but it only requires a disc sitting on top a plinth and you can make holes (or annular gap at OD) in the at disc so it serves as a tunable baffle for the fire box.

    Furnace Isolated Fire Box.jpg

    HT1 likes this.
  9. Hi Kelly, those are all good points: I plan to minimize the refractory from the current 65mm/2.6" down to say 1.5" and back it up with some silica wool based insulation. Time to melt iron was 70 minutes plus and most of that would be due to heating the refractory mass.

    Experiments with the thermocouples showed the refractory disc didn't improve the maximum temperature and slightly slowed down heating. It does give a beautiful evenly dispersed flame up the furnace walls so maybe a washer shaped disc would give even flames up the crucible sides. The disc was present for the successful iron runs but I've chalked it up to the mystery fuel that softened the plastic buckets. I'll have enough chamber height for a refractory disc on top of the plinth in case I need that, it's only a few extra inches and extra height eliminates the annoying cyclone spin on top of the molten metal.

    I like your drawing of the furnace sections, I'll probably cast a cylinder tube that fits a socketed base like you've drawn. The ring round the top would be recessed/socketed to fit the cylinder tube. I'll whip up a drawing and add it to this post.

    furnace 1.jpg
    Last edited: Jun 8, 2024
  10. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    If you think the secret sauce is dwell time, I wouldn't give up on the idea of a firebox for the reasons stated. After you heat the mass of the furnace, if you put heat into a well insulated box, it's either going into the melt or out the vent.

    There is one more part of the discussion and that is mode of heat transfer. Conduction is only relevant at the crucible/plinth and after the metal melts then it transfers much better from the crucible to the melt, but it must be transferring to the crucible first.
    All that's left is radiant and convective transfer for that.

    Velocity will dramatically affect convective transfer, but don't underestimate radiative. In my electric furnace, it's the only meaningful mode of heat transfer and I learned the value of maximizing it. If your hotface can readily conduct heat out of the furnace it's a big loss.

    There are a couple inescapable truths.......iron is commonly melted in conventional tangential Tuyere furnaces, and when well-tuned even with propane.....There is a limit to how much air/fuel can be burned in a given volume and time.

    For a given furnace geometry, I'd imagine the vertical position of the crucible in the furnace could be a significant factor in the characteristic burn and heat transfer. Large head space above the crucible just seems very undesirable.

  11. I think dwell time helps with heat transfer and compensates for slow burning oil, certainly the extra volume can accommodate more fuel burnt without blowing it out the top unburnt. I'll keep the option of the firebox/toroidal/donut racetrack that can be added later as it improves the vaporization of fuel spray and greatly evens out the flame distribution. There's a good chance a larger chamber will make the flame distribution uneven and the toroidal path cures that: I'll just make a disc with a hole in the middle so the flame go up the crucible sides this time and not the furnace bore sides.

    I think my existing furnace with less than an inch of gap between the bore and the crucible was an edge case of not enough combustion happening near the crucible to overcome losses. Turning up the fuel and airflow just blew more flames out the top and supercharging was not an option with leaks. With forced air cooling and even water cooling of equipment there comes a point where the flow speed is too high and things get actually worse due to boundary layers preventing conduction transfer.

    Before the extra half a keg extension, the current furnace had only 2.5cm/1" clearance between the crucible and the lid which gave rise to the slag and molten aluminium tornadoes, at worst I aim to have a lid gap equal to the bore air gap and a generous lid opening.

    So in short the existing furnace chamber is too narrow....adding extra length doesn't fix things and having a shape that gives the most volume for the least surface area should help with conduction/radiation losses.
  12. I'm digging out the Morgan furnace brochure for rough dimensions to use....of course their smallest furnace weighs 500 kilos so I aim to be below 1/4 of that weight without trying hard. I think they give a rough idea of furnace sizes to aim for. There's no mention of the chamber sizes in their brochure unfortunately. Looking at the furnace drawing there's about 2/3rd insulation and 1/3rd air dimension A of 640mm - 197 for an A20 crucible gives 443mm or 17.44 inches.
    Divide that by two for each side and you have 221mm, 2/3rd for insulation is 147mm and 74mm for air gap, That's close enough to 5.8" of insulation and 2.9" airgap around the mouth of an A20 crucible.

    Morgan furnace dimensions.jpg

    Morgan furnace dimensions 2.jpg

    Morgan furnace dimensions 3.jpg
    Last edited: Jun 8, 2024
  13. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    To the extent the schematic is the 5.8" would include all furnace wall hotface and insulation. How would you intend to divide that thickness into the refractory/insulation? Also to the extent you can rely on the diagram, the furnace bore is shown to narrow near the top of the crucible. -That's sort of interesting.....

    1" does seem a little skinny. Also of note the Morgan lit remarks the central vent helps preheat the crucible if other venting schemes were common. As to scale, the vent does seem to be about half the crucible diameter which if so would be 4" and on par or maybe slightly low side with many builds here.

    In any case, you will have a fun build and a new furnace for your efforts.

  14. I just plan to have to refractory concrete as thin as practical: when it cracks I'd like it to hold together say 1.5" or 40mm thick and the rest 10cm or 4" will be insulating wool. I'd like a 4" air gap all round the crucible mouth. I can always neck it down but can't widen it later.
  15. HT1

    HT1 Gold Banner Member

    Mark's castings likes this.
  16. I see the B series will melt some cast irons with gas.....gas seems to make life much easier with furnace design and would be the fuel of choice if it was cheap over here. The B301 would be what I'm aiming for with maybe an extra inch or two on the bore diameter for oil use. It's 13" bore would give a minimum 2.5" air gap at the top rim of an A20 crucible vs the estimated 2.9" of the Morgan oil burner.

    Mifco Gas furnaces 1.jpg
    Last edited: Jun 9, 2024
  17. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    B301 looks to have somewhat smaller crucible to wall clearance (maybe ~1.5") and <2400F and some grey iron duty, presumably because it is designed for and rated using NG or LP as delivered. The Morgan data only shows fuel consumption and melt times for iron using oil as fuel.

  18. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Same goes for Diesel.

  19. Gas is easier to clean up a spill, oil is just messy :rolleyes:. That B702 would have an air gap of 4" for a 40 size crucible down to 2.75" for a 70 crucible.

    I'll cast the refractory bore cylinder first, I have some 12" diameter PVC pipe I can slit and expand to say 16" bore and fit some inner wood circles and massage with a heat gun to the expanded diameter. If I use a keg slit lengthwise, I can weld in some extra from another keg and bring it 19" diameter and that would cast the refractory to 1.5" wall thickness. The same wooden form can cast the socketed furnace base and the socketed ring for the top to cap off the fibre insulation.
    Last edited: Jun 9, 2024
  20. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I was referring to an A20 in B301

    I do similar but just buy cardboard concrete tube near desired size and section as necessary. If the OD will be the wetted side, I wrap it in packing tape. Here's my latest plug mold for my larger face and flanges with coil shelves. 15" Bore x 18.5" tall. Since it's a lift off, it's big enough for IC 60.

    Large Furnace Plug Mold.JPG

    509Maker and Mark's castings like this.

Share This Page