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. oldironfarmer

    oldironfarmer Silver Banner Member

    Be careful with the welding hood. You don't have peripheral vision and can be a source of stumbling because you can't see down very well.

    From what I've read polycarbonate blocks essentially all the UV a, b, and c. I shy away form shields which don't specify polycarbonate for that reason.
  2. PatJ

    PatJ Silver Banner Member

    I hear what you are saying about stumbling; that is certainly not something you want to do with a crucible full of hot iron.
    As it is, I can't see much of anything with two layers of No.5 shield, and so I end up leaning my head back and peering under the outer face shield.

    I am use to having things on the ground, and I keep a mental picture of what is where, and I remember where the "walk zones" are.
    I always put my crucible stand, flasks, tools, burner and valves, fuel tanks, etc. in exactly the same spot every time, and when I have my facemask down, I generally try to take short steps with the assumption that I need to stay on my feet no matter what.
    I also look down before planting a foot, when leaning over and peering into the furnace.

    I have not heard that about polycarb, but I know it does not hold up well in sunlight over time such as in a light fixture lens application, and I spec borosilicate glass for that.
    For a facemask, I think something what won't shatter is a must.

  3. Rule one of Metal Casting Club: don't trip over.

    I think borosilicate passes more IR and UV than normal lime soda glass.
  4. PatJ

    PatJ Silver Banner Member

    I have been asked to demo the new furnace for some folks, and I would have to transport the furnace, which is no problem since it probably weighs in the 250 lb range (total weight of frame, furnace, etc.).

    But a greater issue is with the availability of compressed air for the siphon nozzle burner.
    One location has plenty of compressed air, and another has none.

    To date, I have not been able to make a drip-style burner work successfully, and so while the drip-style burner is not out of the question, I will have to figure out how to make one work if I want to use one.

    I could transport a pancake compressor, but they are noisy and tend to get overworked with a siphon nozzle burner, although at 30 psi pressure to the nozzle, I think a pancake compressor would work, assuming there was enough 120v power for it.

    A number of people I have seen in the past have used a pumped-style spray-nozzle burner, and I have some pump-style nozzles that I purchased accidentally before I knew the difference between a pumped-style nozzle and a siphon-nozzle.

    So I decided to build a pumped-style spray nozzle burner, and ordered a pump and motor that seem to be the ones typically used for the combo oil burner units (the ones with the blower, pump, motor and spray nozzle all combined into a single package).
    I bought this motor. It is probably a bit oversized, but better a little on the large side than too small, and it will not use any more horsepower than is needed to pump the fuel, regardless of its motor hp rating.


    And I bought this pump.
    I am planning on running it between 2.5 and 3.0 gal/hr, and will need a bypass circuit to do that.


    I selected a 1725 rpm motor and pump since I figured they would last longer by running slower, and thus justify the increased cost.
    I also figure that a pump will save wear and tear on my air compressor, and a pump is much cheaper than my air compressor.

    I considered purchasing an entire pump/blower/burner assembly, but the outlet is quite large (perhaps 3" diameter), and there is no way I am changing the tuyere on my new furnace.
    I also don't want the bulk of a burner assembly since I am planning on transporting it.

    So the task that remains is to mate the pump to the motor.
    I found this photo online (not my photo), and it shows a flange/coupling affair, which is what I assume I will have to fabricate.


    I will have to check the pumped nozzles I have and see what the flow rate and spray angle are.
    With pumped nozzles, you get into solid and hollow cone spray designs, as well as spray angles, and I will need to figure that out.
    The siphon-nozzle burner tip I currently use has a very narrow spray pattern, which seems ideal for furnace use.
    I will have to make sure I don't get a pumped nozzle that has such a wide spray pattern that it impinges on the burner tube, furnace walls, or crucible.

    And my backup plan is to make a drip-style nozzle (when all other burner types fail).
    In the past I did not extend the drip tube to the end of the burner tube, so perhaps that is why my drip-style burner did not work.

    Has anyone gone down this path before?
    Any tips/tricks to getting it all set up and working?

  5. DavidF

    DavidF Administrator Staff Member Banner Member

    If only you had asked a week ago...
    Have the foundry packed up right now and still about 3 weeks out on the new shop construction.
    Your better off running a pair of them. You can turn one off and just feed air in the tyre as/if needed. Gives you a little bit of control that way once up to temp...

    This is with a twin tyre ser up. Running a 1.2 gpm nozzle in one and a 2.5 in the other..
    Two separate pumps, with 2 separate blowers.
    Last edited: Nov 12, 2018
  6. Melterskelter

    Melterskelter Silver Banner Member


    This is an very interesting tack and one that I will be interested to see in action. I like the idea of getting away from a compressor. Mifco must be using something like this in their burners.

    There is a good chance you already have allowed for this but I will say it out loud anyway:
    One thing you will likely need to incorporate in your connection of the pump to the motor is a "Lovejoy" style coupler. Getting a perfectly concentric and co-linear configuration of motor and pump is very difficult. So, a lovejoy is used to make up for minor deviation from ideal motor/pump orientation. Without the lovejoy the motor and pump bearings are in a tug-of-war to see which will outlast the other.


    How did you determine motor HP as a full HP seems like a lot? Edit:Maybe the motor is actually 1/3 HP --- it is very hard to read in the picture.

    Last edited: Nov 12, 2018
  7. Errm ...yes? I'm glad I used A section belts and pulleys: you can vary the pump speed with the pulleys and it's simpler to build (at the expense of wear and tear). Already I've reduced the pump speed from 1440 RPM down to 720 RPM as my pump bypassing too much oil through the regulator making the pump and plumbing quite warm (oil shear maybe). You're not going to get the excellent atomisation of the Delevan nozzles you're using and that may require changes to your furnace nozzle position of bringing it into the tuyere a short distance.
  8. PatJ

    PatJ Silver Banner Member

    To the best of my knowledge (check me on this), Mifco only offers burners that run on either propane or natural gas.

    The coupler is probably a good idea, although I think I can machine it accurately enough for a solid coupling (but maybe not).

    Here is a better picture of the nameplate:


    Most of the combo burner units used a motor smaller than 1/3 hp, and generally they use a 3450 rpm motor/pump.
    I erred on the side of caution and went with a slightly larger motor, and went with a slower motor to reduce wear and tear.

    Yes, this is exactly the sort of feedback I need.
    No sense repeating problems that others have already solved.
    The pulley idea seems to make perfect sense, and eliminates the alignment problem too.
    And I think the wear and tear on the pump would be much less if it were slowed down, and there is absolutely nothing to be gained by pumping fuel around in a circle, and as you say you can heat the fuel by pumping it in a circle.

    I have seen bypass systems overheat the fuel oil in situations with multiple 1 MW gensets, and a large chiller had to be installed to keep the fuel cool.
    The oil is actually cooling the nozzle along with the combustion air (in siphon and pumped nozzle oil burners).

    I think the pumped nozzle atomization would work ok give the fact that a drip-style burner uses no atomization at all, and if you look at ironman's videos, he uses the drip-style burner quite successfully.
    The main reason I would like to stay with some type of pressurized nozzle is to avoid having to use propane for startup.
    Basically I don't want to have to transport a propane bottle.

    I did try cranking the atomization air down to zero with the siphon-nozzle burner (during my burner tests), but I started getting a lot of black smoke and fuel puddling in the furnace.
    Clearly the drip-style burner is dispersing the fuel into the air stream better than the concentrated stream that comes out of a siphon nozzle when only fuel tank pressure is used to move the fuel through the nozzle.

    I know for a fact that others here have used the combo/blower/burner units for foundry work.
    Nudge is one who comes to mind.
    See post #43:

  9. I'm using a 1/2 Hp motor as it was handy, in operation I can hear a change in sound as the pumping load comes on and off and it will effortlessly power the pump to 90 PSI if I let it.

    I don't run the belt with guitar string tension, there's a slight bit of slack and the belt wedges tight only when running. I'm using a cheap Ebay car fuel regulator too: it runs perfectly on kerosene and diesel but runs too high a pressure with cooking oil. This seems to be dues to oil viscosity, so a dedicated oil pressure regulator would likely solve this problem. Possibly an even slower pump speed would let the regulator keep up with the flow.

    All I can suggest is run it on Kerosene or diesel for the away game to make it significantly easier to ignite. The sliding fuel tube and nozzle/combustor can assembly is a game changer as far as I'm concerned as I can tune the combustion so it occurs down low in the furnace rather than having spectacular flames out the top. The 4" diameter tuyere pipe runs relatively cool which was unexpected: I was going to line it with refractory but it's not needed. This may not be possible with smaller diameter tuyere pipe. The combustor can is showing serious heat discolouration: I'd love to fit a 1" quartz window to see or maybe get a cheap USB camera in the pipe to see what's happening.
  10. PatJ

    PatJ Silver Banner Member

    Running the return fuel though a piece of radiator, such as a piece of coil from a refrigerator, could solve the heating problem, but I am thinking slow the pump down to the point where there is very little return fuel required and a radiator would probably not be required.

    While I can't see what is going on inside the burner tube itself, I know that there is no combustion going on inside the tube because when that does happen, it overheats the burner tube and causes it to get bright red.

    I can look into the furnace with the lid open and get a good idea of what is happening with the combustion, and where it is happening.
    Looking at the photo below, it can be seen that the area in front of the burner tube (inside the furnace) is cool, and consists of unburned fuel.
    My burner tube fits tight to the tuyere, and it runs cool all the time without any heating or degradation of the end of the tube.

    The end of my nozzle is almost at the end of the burner tube, so there is no danger of the spray impinging on the burner tube.

    The green arrow is pointing at the cool part in front of the burner tube.


  11. PatJ

    PatJ Silver Banner Member

    Once the furnace reached its operating temperature, the temperature differential smooths out as can be seen in the photo below.
    There is still a cold spot in front of the burner tube even when the furnace is at operating temperature because you are introducing ambient-temperature air into the furnace, and it cannot immediately heat up to the temperature of the furnace interior.

  12. The oil pump plumbing would have been warm to touch maybe 40 deg C or 104 deg F: this would be useful in colder climates to preheat the oil to a lower viscosity and make atomization easier.

    Most of my original experiments had combustion patterns are identical to your first photo(post #410), I'm attaching a link to the first combustor can test, of the furnace with an open combustor can around and behind the nozzle. As far as I can tell it's just a form of pre-ignition that allows the fuel to be burning before it enters the chamber and really begins to get hot. At the video start you'll see the 1/2" copper tube entering the PVC elbow: the white cable tie on the copper is an indicator of how far the combustor can and nozzle are back in the tuyere. In this video it's about 4 inches back in the pipe. This is the only time I've had the flames circle the chamber twice, once around the plinth and then once round the wall before exiting the furnace and after a short while the plinth was glowing orange. All other earlier experiments has the lining melting at the top of the chamber and cool carbon soot all over the plinth and bottom half of the crucible, so it was great to get the heat down low for once. Sliding the nozzle forwards until level with the chamber bought the flames back out the top of the furnace.

    Last edited: Nov 14, 2018
  13. PatJ

    PatJ Silver Banner Member

    I was reviewing videos of the new furnace and notices a smoke ring coming off to the right side of the furnace at 0:06.
    I have never seen that before, and don't know what caused it.
    I guess I am easily amused by foundry/burner stuff, but I thought it looked pretty cool.

  14. Melterskelter

    Melterskelter Silver Banner Member

    I am pretty sure that smoke was from back flow up the tuyere as the flame got started. Once you readjusted the position of the burn tube and good flow down the burn tube was well established, it was gone. I have seen similar with my furnace.

  15. PatJ

    PatJ Silver Banner Member

    I am also surprised at how hot my 1" hotface gets on its outside surface.
    Somehow I expected that the outside of my hotface refractory would not glow red because I never noticed that on my 3" thick furnace refractory exterior.

    But the outside of this hotface shell gets bright red, as you can see in the video, with is a bit scary since I did not know you could operate refractory like that.
    I guess time will tell how long a 1" Mizzou hotface will last, but so far it is holding its own with little signs of wear, but I do have an interior coat if ITC100, and that seems to really help protect the refractory, plus I used 3% stainless refractory needles, which are commonly used in industry in refractory to retain it as a solid shape even if it cracks.

  16. PatJ

    PatJ Silver Banner Member

    For the oil pump I bought should arrive tomorrow.
    I may just attach a variable speed drill to it, and find the approximate speed at which I can reduce the bypass flow to almost zero while still producing the 3 gal/hr flow rate.

    If I can get a good estimate of that speed, then I could order some pulleys.
    Mark's castings likes this.
  17. PatJ

    PatJ Silver Banner Member

    I received the fuel pump and 1/3 hp electric motor.
    The motor is huge, and no doubt far larger than what I need, but better a little too large than too small and overheating in mid-melt.

    I have considered the fan belt/pulley arrangement, but I am not positive that such a setup is necessary.

    The specs on the pump are confusing.
    7 gph @ 100 psi
    3 gph @ 150 psi

    I need to read up on pressure nozzles again.
    I think a higher pressure gives better atomization.
    But the flip side is that if 3 gph @ 100 psi is sufficient, then the pulley/belt arrangement would remove much of the load from the pump and probably allow it to last longer.

    I need to figure out how to plumb it and the correct fittings to use.
    The instructions mention not using teflon tape or compression fittings, so I guess that means the fittings must be the flare type.
    I did find a pretty good website about fuel oil pumps.
    I need to read a bit more.

    I know it has been mentioned that returning oil to the tank can overheat it, but then my burner tube runs cool to the touch when I am using it to melt iron, and so I am not convinced my setup would overheat my fuel tank.

    It would probably be easier to connect the motor to the pump using the lovejoy coupling that has been mentioned, and let the excess return to the fuel tank.
    I guess I will play around with the pump using a battery powered electric drill and see if I can get some flow rates measured through a pressure nozzle, and try to get a feel for fuel flow vs rpm.

    Here is an Audel's book on oil burners.
    Audel's books tend to be light on the technical/mathematical side, but good on the graphic illustrations and straightforward explanation of how things work. Sometimes the wording is rather elementary, but you get the idea.
    These two chapters seem relevant to what I am trying to do.
    https://inspectapedia.com/heat/Audel Oil Burner Guide Ch 14e.pdf
    https://inspectapedia.com/heat/Audel Oil Burner Guide Ch 15e.pdf

    Edit 02:
    This chapter also has some good information on oil burners:
    https://inspectapedia.com/heat/Audel Oil Burner Guide Ch 31e.pdf
    Last edited: Nov 14, 2018
  18. If you wanted to go the route of a direct drive pump with a variable speed DC motor, then the KB speed controllers are very common and have closed loop feedback to give a fairly constant speed.

    That would give you a flexible pump system that may not need a pressure regulator, just a pressure gauge to adjust for the particular fuel being used that day. You'd have to fork out some cash for a 90V DC motor or salvaged treadmill motor.
  19. oldironfarmer

    oldironfarmer Silver Banner Member

    Nothing wrong with running Mizzou at a red heat. It's actually better when the exterior face is insulated and the refractory does not have such a large temperature drop across it. That causes differential expansion and cracks. But cracks are OK so long as pieces don't fall out. The needles should prevent that.
  20. PatJ

    PatJ Silver Banner Member

    I am hoping to demonstrate the new furnace at a couple of iron-art shows next spring, but at least one location will not have compressed air.
    So the idea was to use the pumped spray nozzle, with the motor and pump above.

    I am a little short on time, so instead of starting on the pumped burner design, I decided to make another attempt at a drip-style burner.

    I tried to make a drip-style burner a few years ago, but I uses a fuel tube that entered the burner tube perhaps 8" back from the end.
    This was a mistake (the fuel tube should have gone down to the end of the burner tube), and so the burner did not work.

    I looked at a burner type in a book I stumbled across called "Fire Assaying" or something like that.
    Below is a diagram I found for an oil burner with some interesting features.
    It basically seems to be a large drip-style burner, with an adjustable cone.

    Last edited: Nov 18, 2018

Share This Page