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

The Mizzou instructions recommend 5 pints of water for 55 lbs of refractory, but also state that "minor adjustments to the amount of water are permissible to achieve the desired flow".
The instructions also say "do not exceed 11% water under any circumstances".

I think 5 pints of water to 55 lbs of refractory equates to about 9% water.
A pint of water is 16 fluid ounces.

I ended up adding a few ounces of water per 10 lbs of refractory, so it works out to 9.31 % water, which is well within the recommended range.

What is odd about refractory is that tiny amounts of additional water make a big difference in how the refractory flows.
Exactly 9% water makes for a dry mix, and when the needles are added, the mixer basically cannot mix it.
Using 2 extra ounces of water per 10 lbs of refractory changes the mix to a rubbery consistency that is still very stiff, but it does mix completely, and it will flow slightly.

So I guess the thing to note is if you add additional water to your refractory, add very tiny amounts, since a very small amount of water has a very big effect on how wet the mix is.

And as Kelly mentions, if you add too much water, you ruin your refractory.
I used an accurate scale to measure both the refractory and the water (two separate scales).

.

 

I am getting ready to strip off the forms, so we can make some comparisons with non-vibrated and vibrated refractory.
Its a non-issue for me since I am very careful about ramming the refractory, but we will see what the visual results show.

 

I stripped off the forms, and the interior lid surface looks very good with minor spalling, but no air bubbles at all.

The top of the refractory has quite a few bubbles, but that is cosmetic, and will not affect the function of the refractory.

I put the lit and the plinth in plastic bags and sprayed water on the surfaces, and will let them cure for a few more days.

Looks like I have a new domed lid, and it seems to have turned out as I wanted, so that is good.





























Edit:
I welded two bosses onto the lid shell so I would have an option to add a lid lifter.
With this lid turning out a little heavier than expected, I will have to start contemplating a lid lifter design, but for now I will hand lift it while the weather is good.
The lifter for this lid will have to raise it 2" (+) before it pivots, and so my typical hinged design will not work; I will have to lift up first, and then pivot.
I don't like the lift and pivot designs because the last one I build got stuck in the partially open position, ruining a #30 crucible.
I will have to make a more robust design this time, with some sort of backup option to quickly remove the lid should the lifter fail.

 

Looking pretty good!

 

That should make a fine furnace! Well done.

Denis

 

Thanks guys for the support and feedback; its very useful information.
Many of your ideas and suggestions have been directly incorporated into this build.

This furnace and burner are starting to get fun, and I love it when things start to come together and work as planned (for a change).
I am excited and think this furnace will work well with iron.

.

 

The dreaded fall monsoons have set in, and it has been raining every day now for several days, and there is rain every day in the forecast for several more days.

I envy those guys who have a big open shed, who can work and cast metal in any weather.

Probably just as well, I can let my lid refractory fully cure and catch up on a few other items.

Edit:
Just checked the weather forcast.
A little sunshine Wed/Thurs, with more rain after that.
Maybe I can dry out the lid and try it before the next wave of rain hits.

 

I decided to go ahead with the multi-flow burner controls, which started as a post here:
http://forums.thehomefoundry.org/index.php?threads/multiflow-burner-controls.32/

I have kicked the multi-flow control idea around for a while, and debated the merits of it.
On the one hand multiple-flow valves seem like an unnecessary complication when you consider everyone else manually adjusts their single fuel valve, and it all works just fine.
But on the other hand, most people are not melting iron, and more importantly not trying to learn how to melt iron, and not trying to optimize their iron melts.

And nobody has been able to definitively say what the optimum fuel flow rate for melting iron is, or whether that value varies with crucible and melt sizes.
My furnace is set up to use either a #10, 16 or 20 size crucible. I think I could possibly use a #30 crucible if I add an extension ring to the top of the furnace, should the need arise.
So if there is an optimum fuel setting for each crucible size, it would be nice to know what that setting is in order to minimize fuel usage.
For those using waste oil, fuel usage may not be an issue, but why burn more fuel than necessary, and why not have faster melt times if that is possible with a different fuel flow?

It would also be nice to be able to instantly change between multiple settings, such as 2,3,4,5 and 6 gal/hr, without having to blindly guess at a fuel setting.

Looking at the data that is provided for a Morgan commercial furnace, we can see that for that furnace, the fuel flow rate does vary with melt/crucible size.
The Morgan data would also seem to suggest that they use a higher fuel flow rate than most are using for iron; more like 5 gal/hr instead of 3 gal/hr for a furnace that is similar in size to mine.
And the Morgan melt times are faster than what I have seen most achieve in a backyard setting.

I could just try my siphon nozzle and calibrate it to different settings, but it can be time consuming to re-calibrate.
I looked at using a micro-flow meter, such as this one, but I really can't justify the cost, and I have not been able to get an answer about whether the meter will work with diesel.
https://www.omega.com/pptst/FTB300_SERIES.html

I do recall at least one person on the forums using a flow meter for their fuel, but I can't recall who that was, but the idea makes perfect sense if you can find an inexpensive flow meter.

So bottom line is I will never find out the optimum fuel settings for melting iron without trying different fuel flow rates, and the easiest way for me to do that is with a multi-flow valve arrangement.

I almost have it all piped up, and can perhaps run an operational test today, weather permitting.
I may wait until tonight were I can video tape the different settings, and when the darkness helps determine which settings produces the hottest furnace glow.

I need to make a stand for my two air regulators (one for siphon air, and one for fuel tank pressurization), and combine it with the fuel valves.

Photos when it gets light outside.



Edit:
Ironsides reminded me yesterday of my mantra "2018 - The Year of Iron", and so the pressure is on to put up or shut up.
Talk is cheap, and I have done more than my fair share of talking about melting iron, and so I need some solid results, and I think I will have them soon.
.

 

Nice work!

From a thermodynamics standpoint, to melt iron you have to have the atmosphere around the iron at higher than melting temperature for long enough to let the iron melt. That's obvious, of course, but when you know the mass and specific heats of the furnace, crucible, and charge, you have to add that amount of heat to get up to temperature. Then as the furnace warms up you have to add the heat lost through the furnace walls through convection and radiation. Other than heating the contents, the variation in energy is what is lot through the walls. The slower you fire, the more energy gets lost through the furnace walls.

Then there is the energy going out the flue. The more fuel you are firing the more energy you lose up the flue. The slower you fire, the less energy goes up the flue. But firing slowly loses more energy through the walls. There is a lot more percentage wise going up the flue but if you are going to use three gallons of fuel in an hour you find that you have to lose energy up the flue to get the fuel to burn. Assuming you are keeping the air/fuel mixture right, there is very little flue energy savings by firing slowly.

I believe firing as hard as you can (while burning efficiently, and not destroying your crucible or furnace from too much intense heat) is inherently the most efficient way to melt. The energy transferred to the charge to be melted is based on the temperature difference between the atmosphere in the furnace and radiating temperature of the furnace walls to the temperature of the charge. If the furnace is only 100F higher than the melting point of the charge, the charge will be slow to melt. But it will eventually melt. If the furnace is 1,000F higher than the charge, then it will melt much quicker. Heating the furnace to 3,500F might be ideal but it will likely not last an acceptable length of time.

So my goal in melting is to bring the furnace to temperature as quickly as possible then moderating the fuel and air to hold the furnace at that temperature until the charge is melted and up to temperature. That is the most efficient heating cycle fuel-wise and should also be the shortest melt time. lowering the furnace temperature as the charge gets up to temperature helps avoid overheating the charge but is less efficient in terms of time and fuel.

 

As I mentioned previously, thermo was never my strongpoint, but I do understand wind chill.
I use to try riding my motorcycle in the winter, and I did not make it very far before I discovered the disadvantages of wind chill.

So I wonder if a higher fuel flow will have the wind chill effect, or whether it will actually cool the furnace due to insufficient time for the fuel to combust.
And granted, much of the crucible heating is probably due the radiant energy from the furnace walls.

Only one way to find out, and that is to try a few different fuel flow rates and see what happens.

As far as damaging the crucible and furnace from too much heat, my hot face shell is replaceable, so at this point I will sacrifice it and a #10 crucible if necessary to find out what the limits are for melting iron.

Here is the "octopus". Lots of hoses and fittings.
Will it work?
Place your bets now.

I have it set up with a dual-fuel selector switch, in case I want to start on diesel and then transfer to waste oil.

I went with five ball and needle valves, which will have fuel flows of 2,3,4,5 and 6 gal/hr.
I could have used fewer valves, but I did not want to be trying to add/subtract things during a melt.

There are two separate blue manifolds, and they secured together with some hose clamps.
I put a ball valve on one fuel inlet, in case I wanted to operate it off of one fuel source only.

 

Nice manifold work!

As far as wind chill goes, if you keep your air/fuel mixture right, and you are burning most of the fuel, the cold air will be at flame temperature as it exists the flame. My statements above were predicated with "from a thermodynamics point". It has be to assumed that your furnace design allows combustion to take place in the furnace. All the oil and air has to be heated to flame temperature but that is true at 1/2 gallon per hour or 12 gph. If your burner tube is too small air velocity may have to be high enough you do get some chilling low in the furnace. That would be burning too much fuel for your furnace design. In my furnace with the burner tilted down the plinth gets hotter than the crucible at high flow rates even though there is flame exiting the flue.

edit: I did want to comment on wind chill, since it is so misunderstood. Wind Chill is defined as the apparent temperature bare skin sees during a wind. Wind does remove heat from an object faster than still air, but it cannot remove heat to a temperature lower than the wind. You can carry a jug of water on your motorcycle on a 33F day and the water will never freeze. You skin will feel the same as if it were a 10F day in still air, but even your skin will not fall below 33F. A strong wind on a 70F day makes us feel cold because we are trying to maintain 98F body temperature. Your motorcycle on a 100F day will never make you feel cold.

 

I agree with Oldironfarmer about having a high rate of burn even if it's fuel inefficient as commercially used furnaces tend to reflect this: fuel is cheaper than wages for smaller furnaces. I used to have a textbook on designing efficient switchmode power supplies and the section on cooling power supplies was interesting. Heatsinks get rid of waste heat from active parts and airflow improves this cooling effect: above a certain airflow speed, the cooling actually gets worse due to boundary layer effects. So for best cooling performance, a large heatsink with low speed airflow performs better than a small heatsink with high speed airflow.

This high speed boundary layer might work in our favour with a furnace as the heat energy is radiantly coupled to the crucible but the boundary layer makes it harder for the combustion gas to carry heat away from the crucible.

Pat: before you posted your idea of multiple fixed flow nozzles, I was thinking that four nozzles could be arranged as a four bit binary system: a set of four nozzles: 0.5, 1.0, 2.0 and 4.0 GPH would give you 0-7.5 gallons per hour in 0.5 GPH increments. Using it would involve 16 combinations of tap opening and closing so it's not practical unless you had some cheap solenoid valves and a rotary binary switch to drive them in sequence.

 

That is an interesting point about the boundary layer effect.

16 combinations is too much even for me.
I will try the 5 valves and see how that works.

The wind chill discussion is interesting too.

I made a stand to hold the compressed air regulators and valves, and used the remains of a keg to make a tree-stand for the flow control valves.

Edit:
Don't use Cornelius kegs for fuel tanks.
The rubber gaskets do not stand up to diesel.

I have a 5 gallon "Cornelius" keg, and will use that as a local fuel tank so I can more accurately track fuel usage by weighing the tank.
I bought a hanging scale to suspend the Cornelius keg.

My wife came outside, took one look, and said rather skeptically "We are going to have to get several people over here with cell phones when you use that thing".
God bless her, she means well, but translated, what she said is "I don't have a clue as to what all those gauges, valves and tubing are for, but it looks really scary. With enough cell phones, we can call the fire department and transport what is left of you to the hospital."
She is actually pretty technically savvy, but obviously it looks like there is a lot of stuff in the driveway; that is probably a fair statement.

The keg tops with the pipe attached make a pretty good temporary hanger, and I suspended the burner from one while calibrating the five needle valves.
Word to the wise; start with calibrating the highest flow first, since if your fuel tank pressure is not high enough, you will have to increase it to meet the highest flow rate, and that will change all the lower flow rates.

Next step is to try the multi-valves out.

 

I'm sure your wife was saying "it's such an interesting, complex operation we need several guys filming with their cell phones so we don't miss anything".

Including the accident.

 

There was one guy who was using a pressurized fuel tank, but he did not have a safety valve.
His pressure regulator failed, and the pressure blew his fuel line off, and he had a mini-inferno going.
I have a safety valve on the fuel tank that vents at 30 psi.

Diesel is not that volatile at room temperature.

Here we go; I will try to video it all.

 

Yep, I have done that before trying to get a wet wood fire going by dumping diesel on it (years ago). Whoooooooosh. It all went at once.

The tests went well today.
I am still uploading video.

In summary, it seems like the 2 and 3 gallon per hour settings seemed to work best.
At rates higher than 3 gal/hr, there was a lot of flame, and much more blower noise and air movement, but it did not really seem to get as hot as a setting that was between 2-3 gal/hr.

The maximum heat setting appears to be between 2 and 3 gallons per hour, and there seems to be a spot where you can introduce a certain amount of fuel and air, and there is a significant and noticeable chemical reaction that produces a lot of heat.
The setting has to be pretty rich, and the very hot part of the reaction has to extend all the way over the top of the crucible and out the lid.
The combustion air flow is relatively low when the hottest flame is being produced.

My guess is that too much fuel and air is moving the mixture though the furnace without it ever having a chance to properly mix and burn.

The multi-valves worked well.
Once I get use to the 2 and 3 gallon settings, I think I will be able to use a single needle valve.
I guess I will have to do two melts, and compare the 2 and 3 gallon settings to get fuel usage and melt time comparisons.

I put the plinth in the oven in the kitchen.
The wife asks "What are you cooking?".
"A plinth" I responded.
She did not ask what a plinth was, luckily.
She probably did not notice that I borrowed her decorative potholders either.
Its hard to hold onto a 400 F plinth.


Here are a few photos.
After the tests, I cooled off the furnace with the blower for a while, and then put the new lid in place and let it start drying out.
I will leave the new lid on overnight and let it continue drying.

I should be able to finish drying the lid out tomorrow, and can perhaps completely assemble the furnace and lid, including the IFB's, etc.


Video when it finishes loading.

 

Nice! I'm assuming you had the furnace nice and hot before you started testing various flow rates.

Have you calibrated your air bypass valve? Knowing air flow is just as important as fuel flow.

Since you've got your manifold, why not leave the 1 gph setting and add a 1/4 gph and 1/2 gph. You may find you like 2-3/4 gph best, or 2-1/4, and you may like 1 gph or 1-1/2 gph for soaking after you get to temperature. With a 1/4 and 1/2 with 1 gph you could run the full range of 1/4 gph to 3-3/4 gph with ease. When you get really comfortable with your setup you should find different fuel flow are better for different conditions.

 

It may be that your combustion air velocity at high fuel flow is too fast for flame propagation to allow full combustion to occur in the furnace. It might be that two burners at 2 or 3 gal each would produce higher furnace temps than one at 4-6 gal. I know you are aware that MIFCO uses a two-burner setup on some of their furnaces. (They ain't fools). Perhaps that is why they do two burners.

Denis