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

After all of your efforts I wish you success, or at least a pathway to it.
However, that device looks like something that would go nicely inside a .410 shotgun shell.

Pete

 

My vote for naming it is the "Troculator".

 

Thanks, I appreciate the thought.

My experience with foundry stuff is that generally holds true.
And tonight was no exception.

Edit: This turned out to be a bit of a War and Peace novel in length, so if you want to save yourself an hour, you can read this sentence and immediately skip to the next post; the bottom line is "the burner did not work".

I guess I could say "it was a good learning experience", but that would be a euphemism for "it did not work worth a *%&$#@!.

The burn seemed to start out well enough.
Lots of flames, relatively poor atomization, but the furnace seemed to be getting hot enough to melt iron, it seemed pretty adjustable, but then the tuyere area started smoking.
I was hoping that perhaps the smoke was steam from some water that could have leaked under the tarp, but I held a lighter to the "steam" and it lit, so it was vaporized diesel.
The atomization was not good, and the burner was puddling back down the tuyere and out the furnace.

At this point Murphy entered the room, and things got ugly.
The hot face shell heated faster than the lid, and so jammed the lid shut.
I tried to lightly pry the lid off, and the hot face shell cracked in half around its circumfrence about half way down the furnace.
I noticed what was happening and quickly released the prybar pressure, and the crack mostly closed up.

I let the lid heat up and finally got it open.
Looks like the lip protruding down from the lid is a very bad idea, because either the hot face and/or the lid refractory are shifting/moving around just enough to jam the lid.
I think the solution for the lid problem will be to use a diamond blade in an angle grinder and cut a curve in hard fire bricks, and lay them down flat round the top of the shell refractory to form a flange.
A flange would not be necessary if the lid did not have a lip, but I don't have a good way to get rid of the lid lip at this point, and any material used to fill in the lip would eventually crack and fall out.

The cracked in the refractory shell is clean, so like the other cracks that have already formed in it, this crack will be sealed up with ITC200.
Lesson learned.

I suspected that the new burner may puddle with the drip being back in the tube, so I cut the fins off the end of the burner tube, and allowed the end of the vane "thing" to protrude beyond the end of the burner tube about 1/2".
This helped slightly, but I could look into the furnace while it was running and see a dribble of fuel coming off the end of the thing, with basically no atomization at all.

I decided to go to Plan "C", which was to default back to an "ironsides" style drip burner, remove the thing, and use a spacer to center the bare fuel tube in the burner tube.
My burner tube has a tight fit in the tuyere which prevents the burner from sliding all the way into the tuyere, so I moved the drip tube so that it protruded about 1" beyond the end of the burner tube, and very clearly was at the inside surface of the hot face shell.

I started the burner again, and it ran for a minute, and then spit and sputtered for a second, and then went completely out.
That could only be an empty fuel tank, and that turned out to be true, so I refilled the Cornelius keg.
A side note on the Cornelius keg; one of the kegs I bought had a rubber seal that did not like diesel, and the seal swelled so badly that it will no longer seal the tank.
The other keg fuel tank seems to have a seal that is compatible with diesel (so far), but I will have to find a better fuel tank, or find a workaround for the seal.

With a freshly filled fuel tank, I started up the burner again using the bare drip tube (ironman-style).
For whatever reason (I have no idea why) the burner refused to regulate, and fluctuated badly.
If I got it rich enough, then it would lean out in about a minute without me adjusting the needle valve.
I tried everything, but there was no way to adjust the burner and hold any given setting.

At this point it was time to cut the losses and fold.

Some observations about drip-style burners:

1. Obviously ironside's drip-style burner works well for him, and last configuration I tried is basically a carbon copy of his drip-style burner, but I suspect that ironsides motives for making and posting videos is as much to drive amateur backyard casting guys insane trying to copy his success as it is about actually casting metal. That is just my hunch from personal experience, not to say that his videos aren't extremely helpful (they definitely are).

2. Observing my furnace tonight, the drip-style burner that is tuned correctly does appear to get hot enough to melt iron, assuming I did not have all the other problems such as puddling, etc.

3. By looking into the furnace while it was running, I could see that the "thing" and the open drip tube both did the same thing, which is to dribble fuel into the combustion air stream.
The fuel travels across the furnace about 6 inches and then impinges on the wall and floor of the furnace where it flashes to vapor, and then burns.
This is not an entirely clean process, and the setting that produced enough heat to melt iron was a noticeably dirtier burn, with some soot visible out the lid opening.
After striking the far side of the furnace, the fuel breaks up in a rough fashion, with some heavy yellow flames. Heavy yellow flames are a sign of poor atomization.

A properly tuned siphon nozzle burner will burn perfectly clean, with no soot or smoke at all, and the combustion is visibly more complete inside the furnace, with no heavy yellow flames.

It is clear that I am not going to get a drip-style burner to work like I want it to work.
I do appreciate the siphon nozzle burner more and more as I try other various burner styles.
I have not tried the pumped nozzle burner yet, but I consider the siphon nozzle burner to be far superior to a drip-style burner as far as completeness of combustion and controllability.
I can light my siphon nozzle burner, turn on the leaf blower, fine tune the fuel needle valve (all within 60 seconds or less), and then literally never tough the needle valve again for the duration of the melt.
The downside to the siphon nozzle burner is that it requires an air compressor, and so if you don't have an air compressor, or don't have a good air compressor, then your options are propane/natural gas, drip-style oil burner, or pumped-style oil burner.

Since I am not going to try to use a drip-style burner, the next test will be to go back to the siphon nozzle burner, and try it with a small air compressor, and see just how small of an air compressor I can use (how many CFM are required) in order to maintain 30 psi compressed air on the nozzle.

I have several small air compressors, and they do not use very much 120V power, so the plan right now when/if I attend some shows in the spring is to take two or three small air compressors, and use a siphon nozzle burner.
At least the siphon nozzle configuration would be a sure thing since the siphon nozzle I use is really a rock-solid performer when set up for 3 gal/hr, 30 psi atomization air, and 15 psi pressure on the fuel tank.

Another thing that may or may not be obvious to those reading along is that backyard casting is not really a good hobby for those individuals who demand immediate gratification with minimal problems.
Unless you are very lucky, there are many things that need to be tried, many failures and setbacks, and a lot of trial and error, with an emphasis on "error".
If you can't work through the problems, and come up with effective solutions to the problems that you experience, then your casting days will be short.

I am in it for the long run, and I am still super thrilled to have figured out how to get iron to a high pour temperature.


So there we have it.
The "thing" (that is what I am going to call it from here on out) does not work.
No big surprise that it does not work.
I only gave it a 25% chance of working before I started testing.
It is really turning out to be a bit of an art to get full atomization and and optimal burn.

Time to polish it up and call it art-deco, or steam punk, or something and put a big price tag on it.
I can't wait to find a prospective art buyer and ask them what feelings the "thing" piece of art evokes in them emotionally (you have to use "art-speak" if you are going to sell art; after all art=emotions, and without emotions you have no sale).....

Edit:
I have a video and a few photos; I will upload and process in a minute.

Edit02:
On the plus side, I only have about 4 hours time in the "thing", and that is why I don't spend much time when making prototype equipment because more often that not it do not work correctly.
I can still use the stainless steel burner tube since it is generic in design.

.

 

I tried to add a bit to the above post, and got a message "Error, you cannot post messages larger than 10,000 characters".
Gosh, who can function under such sever limitations; what the heck ?

Just kidding, 10,000 words is enough, and probably 9,990 words more than anyone is going to read anyway.
You have to be a real hardcore backyard person to wade through a 10,00o word single-post book about a drip-style oil burner.

.

 

I ordered some new resin-binder the other day, and it is a water-based binder, and as I understand it, sort of a mix between non-water-based resin binder and sodium silicate, and will make molds that are tougher than sodium silicate, but not quite as tough as non-water-based resin binder (according to the sales rep).
The fumes are suppose to be greatly reduced, although a respirator will still be required since there are solvents/materials that can be dissolved in water (just like water-based paint).
The water-based binder I bought is suppose to work well with iron and steel, and I will be using it with OK85 (85 mess silica sand) which is mined in Oklahoma, and the mines can actually be seen on the map satellite view. OK85 is very dry sand (probably oven-baked), and that is required when using resin binder.

I have had great success in the past when using resin-bound sand for molds and cores, with both aluminum and iron. It allows some very creative mold making, with things such as cementing multiple mold parts together, carving/drilling/cutting the molds after they set to create sprues, runners, gates, basings, etc.
Resin bound sand also makes great cores.

The art-casting work seems to use resin-bound sand exclusively with iron, as far as what I have seen online and locally, and I think the reason is that art iron pours generally offer scratch blocks for sale. Scratch blocks are square blocks of resin sand with a recess in the open face. The purchaser uses various tools to carve shapes and contours into the face of the mold, and the mold is then poured with iron as an 0pen-faced mold.

.

 

Here are the photos from the drip-style burner test.

Here is the burner tube end cap, made from a piece of steel bar stock cut in a circle, welded onto a piece of 2.5" diameter muffler pipe, with a pipe fitting welded in place.








The cap is secured to the burner tube with a stainless plumbing collar.




I used aluminized tape to temporarily seal things, and that turned out to be a good idea since I had to move the drip tube twice.











The burn seemed to go pretty well at first, and then the dreaded "fuel puddle" appeared below the burner tube, and heavy smoking followed by a fire started on the outside of the tuyere.



Round #2, I cut the fins off the end of the burner tube and moved the burner interior towards the end of the tube.
This did not improve the operation of the burner.



Round #3, I removed the vane assembly altogether and tried an open drip-tube (aka ironsides style).
Strike three; this did not regulate well for some reason, and I could not get a clean burn no matter what I did.
This was the moment when the burner test started to get tedious.









.

 

Seems a bit like plowing frozen ground, and the plow keeps breaking.
No pain, no gain I guess.

At least I know the furnace works well at melting iron when using a siphon nozzle burner.

 

I got a duplicate of the video here somehow.

 

Here is the video of the three drip-style burner configurations.
At 1:00 the fuel tank ran empty, and the 15 psi created sort of a jet stream of fuel into the furnace until the air pressure bled down and the last amount of fuel was emptied from the tank.
I was a bit surprised, but have seen a number of flare-ups and so not really alarmed other than to step over quickly and turn off the blower and fuel.

This is the polar opposite of how the furnace acts with my siphon nozzle burner.

 

As a test, tomorrow I am going to plug my siphon nozzle burner back in (I have quick-connect fuel connectors), and test the furnace again without changing any fuel settings.
This test would rule out any problems in the fuel delivery system that may be causing the drip-style burner to act so erratically.

 

I feel your pain, I really do, I just had an iron melting fail with fuel leaks/flames towards the end. At least your stuff melts on occasion .

 

I have tried a lot of oil burner types, and while many/most will melt aluminum, there appears to be a narrow range of combinations and methods that work well with iron.

Ironsides uses the drip-style burner seemingly effortlessly to melt iron.
Porositymaster uses the Ursutz-type burner and also effortlessly melts iron.

I have experimented twice with the Ursutz, and did not feel like I was getting anywhere as far as making a reliable burner that would easily melt iron.
Pretty much ditto with the drip-style burner (so far).

I had a lot of trouble with the siphon-nozzle burner for a while, and I think most of that came from not knowing what a hot burn looked like, not realizing that the slag on top of an iron melt had to be broken up during the melt, and then trying to get the burner to run hotter by forcing more fuel and air through it, and so went from a burner that was working ok in the early stages (on occasion) to a burner that did not work at all.
Thinking back on it, there were several melts where the burner worked perfectly, but I let a hard slag cap form on top the melt, and so I wrongly assumed that the burner was not working correctly.

Once I zeroed in on the 3 gal/hr burn rate, and learned to add scrap and push it through the slag cap and under the molten iron surface, I have had perfect iron melts ever since, consistently every time.
I can understand why some people complain about siphon nozzle burner since they have a narrow range in which they will operate correctly with a furnace to melt iron.
It took me a while to find that range, and I think some people never discover the correct fuel flow, but if you find that spot, the siphon nozzle is a great iron-melting burner.

If you have decent refractory (regardless of the weight), then the secret of successfully melting and pouring iron lies with the all important burner.
Iron melts are just not very tolerant of burners that operate very far off from an optimum burn with nearly complete combustion of all the fuel that enters the furnace.

Luckily I now know exactly what the furnace looks like when the burner is operating correctly, and that is very helpful when comparing other burner types.

Of all the types I have tried, the siphon nozzle is far and away the best performing (at 3 gal/hr), easiest to control, most stable, responsive, and one that does not require adjustment during the melt.
But you need an air compressor.
There is no free lunch in the foundry world (or not much anyway).

.

 

It is truly a matter of sticking with it, and trying things until you discover what works well.
But it may be a rocky road for some; it has been for me.

.

 

Comparing the video from tonight's drip-style burner, when it was obviously operating hot enough to melt iron (at 0:26)




with the siphon-nozzle burn from the other day, you can see how clean and clear the siphon nozzle burns.



and this video at 1:20, you can see how clean the siphon nozzle burns when tuned correctly.
Absolutely no smoke or soot, which means pretty close to 100% combustion of the fuel.


.

 

I think I might have an idea as to why your drip nozzle failed.
When I first started working in maintenance for the local packing house. On occasion the sanitation workers would ask us to make a suction nozzle. It was used to spray cooking oil at a flow rate of around 1 G.P.M. The air was supplied by 1/4 air hose.

We built them out of a 1/4 inch pipe T, 1 brass hose barb, and a piece of 3/16 or 1/4 OD. . copper tubing.
The hose barb is drilled to the size copper you chose then the tubing is soldered in the hose barb. I hope the picture explains this wall enough.
The trick to getting the things to work properly was to cut the copper at 45 degrees and then make sure the cut had sharp edges. It was tuned by using lock tight on the hose barb and simply tightening until the copper was centered in the T with the high side of the copper facing the air in and the low side of the copper angle facing material out.

I never tested one but the vacuum was considerable. They sprayed this oil on equipment in a fine mist to seel the surface from oxygen prior to use.

I think several might have been the cause of yours not to work properly. You were dumping the oil out the end of a long straight thick tube. As the air flowed down this tube it is slow in a turbulent flow. The air would normally be at its fastest in a laminar flow in the center of the tube if it was not for the oil delivery tube causing drag. If the delivery tube end has sha rp edges with no place to puddle, approaches from a right angle. The faster air in the center of the burner tube will cause the oil to shear in the smallest droplet possible for the conditions. Another potential problem might be that with the delivery tube inside of a guide tube. There is a gap and a low pressure spot where one tube exits the other. Oil could puddle and be dumping.
The nozzle pictured I threw together quickly when curing my furnace. The air source was a hair dryer. You could feel a light vacuum on the fuel line. I put my fuel tank up 5 foot on a ladder although I think it probably would have fed all the fuel I would have wanted if the tank was level with the burner. Just using the minimal vacuum being produced.

The brute burner benefits by the delivery tube coming in from the side, being small in comparison to the burner and it is close to the inside of the furnace heating the oil.

Joe

 

I have heard that the oil delivery tube will deliver better atomization if the opening is at 90 degrees to the airstream.
I guess I could bend a 90 degree angle on the end of the fuel delivery tube and try that, but at this point I am ready to kick the drip-style burner to the curb and go back to the siphon nozzle burner.
There is only so much of this playing around I can take, and I am really at my limit right now.
But definitely an idea I will try one day, or perhaps someone else here will give it a try.

.

 

I noticed last night in one of the comments in ironsides video that he mentions he has a 1/16" hole drilled in the end of his fuel delivery pipe to aid in pressurization, and a pressurized fuel tank.
I was not aware of the hole in the end of his fuel pipe or the fuel tank pressurization.
So that is one explanation of why his drip-style burner works and mine does not.
Its the little details that matter sometimes.

For me, I am going to proceed with a gear pump design to feed a pressure-type Delavan spray nozzle.
I really like some sort of atomization, either by mechanical means or compressed air, since it gives a smooth burn and excellent control.


I plan on building a pumped-style oil burner for everyday use, and use the siphon-nozzle burner as a backup in case the pump stops working.
The pump motor is 1/3 hp so the power required is very small, and I should be able to easily use it if I travel to shows.
I started on the pressurized pump build in post #404, but got off that idea to try the drip-style burner.
I am proceeding with the pump build at this point as follows:

Here is the motor I purchased for the pumped unit.
Most of the newer pumps run at 3450 rpm, but I wanted a slower unit to reduce wear and noise, so this motor runs at 1725 rpm.







And here is the 1725 rpm pump unit.
It is rated for anywhere from 7 gph @ 100 psi to 3 gph @ 150 psi.



I seriously considered using the belt and pulley arrangement to slow the pump down.
As I understand it, I only need 100 psi for good atomization, so at 100 psi, I would have an excess of 4 gph.
Slowing the pump down would drop the pressure below 100 psi.

If I turn the adjustment screw on the side of the pump, I can crank it up to 150 psi, and get 3 gph, but I don't want to run at 150 psi because the hose I bought is rated for 100 psi, and atomization may or may not be better at 150 psi.
150 psi would also put more wear and tear on the pump for no reason.

So I will have a supply and a return from the fuel tank, and I will use a needle valve in the return line, and adjust the needle valve to get my 3 gph at 100 psi to the nozzle, and return 4 gph to the fuel tank. 4 gph is not really a large amount of flow in my opinion,and my current siphon nozzle burner tube runs cool to the touch, so I don't anticipate a fuel heating problem.
If the fuel is heated a bit by the return pumping, I think my 20 gallon fuel tank will be large enough to handle it.

I am going to add a spin-on fuel filter (from Northern Tool) to make sure I don't get any trash in the nozzle tip.




I will go with a Lovejoy coupling to avoid any alignment issues.
The two pieces have to be ordered to match the motor and pump shaft sizes, and in my case those shaft sizes differ.

 

The pumped nozzle will look and work like this.
I am going to try and get close to the spray angle of a siphon-nozzle tip, which I think is 55 degrees.
I ordered a 2.5 gal/hr type B solid cone nozzle with 60 degree spray angle, and also a 3.0 gal/hr type B solid cone nozzle with a 45 degree angle.
I could not find either size in a 55 degree angle, but I think either the 60 degree or the 45 degree angle will work.











These nozzles screw into the standard Delavan adapter and will be connected as a tw0-pipe system similar to the Variflo diagram below. (Note: I am not using the Variflow-type nozzle, but it was the only tw0-pipe diagram I could find).

 

Here is the layout I will use, and will have to fabricate the parts.

I will use quick-disconnects for the lines to the fuel tank nozzle.