The Kitchen Foundry
In part one I documented the construction of the Kentucky Fried Foundry, designed to melt small batches of aluminum using our Fluxeon® Roy induction heater. In this instalment I document the furnace’s use.
This shows the overall setup of the Kitchen Foundry. From left to right
- Power meter to measure the energy consumption
- Stuff to be melted
- The Hobo 4 channel thermocouple data logger
- The Roy 2500 induction heater with laptop on top
- Induction range with a batch of Southern iced tea brewing
- The furnace.
Notice that the furnace is sitting on a lightweight fire brick (LFB) supported with some kiln furniture. Notice that the kiln furniture provides plenty of air space between the LFB and the range top. That is very important, as I’ll show you later.
The Roy 2500 will operated from either 120 or 240 volts. This prototype has a 120 volt cord on it so I constructed a fully code compliant :-) 240 to 120 adaptor so that the Roy could be powered by 240 volts for this run.
The Onset Hobo data logger. This nifty and inexpensive logger has enough memory to take a sample every second for several days.
Something interesting to note here. All 4 thermocouples are at the same temperature and yet there’s almost a 6 deg disparity between readings. And note that the last plug is barely engaged. That’s because it is so out of tolerance that it would have broken the socket had I forced it in. The first thermocouple is a high quality high accuracy thermocouple from omega.com . The second one is made of thermocouple wire but has been abused by high temperature use. The third and forth ones are cheap ChiCom units from sleazebay. Note the blue spiral around the wire. The spiral designates thermocouple extension cable. This is a cheaper, low tolerance wire that is designed to carry the signal from the actual thermocouple to the readout device. Here they’ve used it to try to make actual thermocouples. Problem is, TC extension cable is typically spec’d at ±10 deg F while a quality TC wire will typically be better than ±1 degree. I used the cheap chicom ones because I was out of quality ones. Good enough for this experiment.
For this run I was also evaluating a new cooling system for the Roy 2500. So I was measuring the temperature at several strategic places inside the case plus the case pressure generated by the cooling fan.
This is a micromanometer. It is capable of measuring air pressure to 0.001″ of water. It is extremely handy for verifying the cooling air paths and pressures inside a sealed enclosure. Its operation is simple. In the rear is one AA cell. This cell is connected in series with the ON/OFF switch and a current limiting resistor. Then one lead connects to the water solution and the other to the micrometer head. The micrometer head has a tapered needle on the end.
With no pressure on the manometer, the micrometer is adjusted until some reading is obtained on the meter and noted. Then pressure is applied and the micrometer is adjusted upward until the meter returns to the same place as before. The pressure is the difference between the two micrometer readings.
This is a very expensive instrument but when designing compact, high power devices, it is invaluable for measuring pressures and flows at various points within the enclosure. This photo should allow any enterprising experimenter to make one of his own.
This is an important detail. The copper wire isolates the heat of the induction coil coming from the furnace from the Roy leads. Initially we had each one immersed in a container of water but we quickly realized that air cooling was sufficient.
This is what happens when the connector isn’t tightened well enough. The heat pretty much destroyed the connector.
Charging the furnace for the first melt.
Here is the first melt melting. From cold to 5 lbs of metal melted took about 15 minutes.
Unfortunately a crack opened in the Kast-O-Lite. It wasn’t supposed to do that but it did. We followed the curing and heat-up schedule to a tee but it still cracked. The good news is that it did not grow. In fact, after we put a large worm drive pipe clamp around the furnace, the crack practically closed. In any event, the crack did not affect the operation of the furnace.
Here’s the crucible almost full of aluminum.
And here it is with the lights off
So the furnace works but it needs several improvements.
The first improvement is the refractory. The Kast-O-Lite is heavy and relatively highly thermally conductive and with a large thermal mass. The outside of the furnace reached temperatures in the 500 deg F range. While I can pick up the furnace to make a pour using heavy welding gloves, I’d much rather the surface remain at near room temperature. And I’d like it to be much lighter than it is now.
We’re working with our supplier, Larkin Refractory to come up with a more suitable refractory. It will probably be one of the foamed refractories. These are lightweight yet strong, have low thermal conductivity and little thermal mass. That should cut the melt time by at least a third.
Not much here. Frankly I’m tired of the safety nannies splattering every thing and every activity with pages of warnings. If you don’t realize that red hot molten metal can be dangerous and can cause fires then you need to spend some time researching the subject on the web. I assume my readers have already done that.
Some folks might get their panties in a wad about doing this in the kitchen. In reality, the kitchen is one of the safest places to melt aluminum. The stove top is inherently fire-proof. If the worst were to happen – the crucible rupture and molten metal spill out – the metal would land on the porcelain, then run down through the burner openings to find more metal and fiberglass insulation. If one lacks a good outdoor place to do this, the stove or maybe the fireplace (if it’s real) are the best places to work.
We do the pouring on the floor. A sheet of stainless steel is placed on the vinyl and on top of that is a thick bat of FiberFrax. On top of that rests the mold.
Also not shown in the photographs are a large fire extinguisher and a bucket of water. The water is there to quench any metal that might escape the Frax and reach the floor. These precautions are just plain old common sense.
Next I want to illustrate something that is not immediately apparent. Refractory can store a lot of heat energy within its mass and if it is an insulating refractory, it will take a long time to work its way through the material. Let me demonstrate an example.
This was our very first attempt to melt aluminum with the Roy. A cut-off propane cylinder, some LFBs and under it all, a hunk of refractory hard board. This actually worked quite nicely.
Here is the crucible a couple of hours later. I had set a fan to blowing on the assembly so both the crucible and the top of the hard board were at room temperature.
My friend Matt went to bed but out of an abundance of caution, I decided to sit up awhile longer and kill some time on Youtube. You can see the orange spritzer bottle in the background. That became very important later.
A couple of hours into my Toobing, I smelled smoke and went into the lab to find my bench on fire. I used the spritzer bottle full of water to put out the fire and then investigated. This is what I found under the hardboard.
The heat wave had worked its way out the back side of the refractory and had set the bench on fire. This happened a full 4 hours after the heat was turned off and the fan applied.
Some air space between the hard board and the bench would have completely prevented this. As would have taking the crucible and refractory outdoors after we finished our experiments.
Let our experience be your lesson learned.
Posted by neonjohn on November 9th, 2014 under Cool Stuff, Electronics, Induction heating, Projects |
3 Comments »
Print This Post : Print This Post PermaLink
A major use for induction heating in industry is metal melting. Since I have an interest in metal casting, I decided to build a furnace suitable for melting small batches of aluminium for sand casting. I also wanted to see how well our Fluxeon Roy product, especially our soon-to-be-introduced 2500 watt unit would work at this task.
The first step was to build a small prototype furnace capable of melting 5 or 6 pounds at a time.
The main ingredients for this small furnace are shown here. The crucible is a squat propane gas bottle with its head cut off. The outer container is, of course, a KFC bucket, the large version (special thanks to the Athens, TN KFC for donating the bucket)
The cylinder is wrapped in FiberFrax spun alumina matting to electrically insulate the induction coil from the steel crucible.
Here my friend Matt is cutting from a blanket of Frax. Note the gloves! Frax is even more itchy than fiberglass!
The Frax is covered with fiberglass tape to hold it tight. This tape will burn off during the first heat but it is necessary during assembly. Other tape such as masking tape could probably be used but it would smoke and stink during the burn-out.
Winding the induction coil. Based on past experience, I judged 6 turns to be just about right. It turns out that 7 turns would be exactly appropriate for the Roy 2500 while 6 are just right for the Roy 1500.
The finished assembly. More fiberglass tape is used to hold the work coil in place. Note that the windings are concentrated toward the bottom of the tank. Heating the bottom most intensely greatly speeds getting that first melt started.
The next step is to mix the refractory. I used some Kast-O-Lite 3000 extreme temperature refractory because I had a spare bag laying around the lab. This is not the optimum refractory – it is heavy and has a relatively high thermal conductivity but it worked OK. I’ll be talking to my friends at Larkin Refractory concerning the optimum refractory and an update will follow.
When using Kast-O-Lite, it must be mixed very dry. Larkin recommends the consistency of oat meal. I did some test casts and found out that a stronger product resulted if less water was used. The photo shows what it should look like. Lumpy but the lumps should adhere to each other if squeezed in the hand.
Before mixing the refractory, the inside of the KFC bucket should be heavily greased to make the refractory properly release from the mold. Don’t be miserly here, try liquid cooking oil, PAM or anything like that. The refractory will soak it up and you’ll find yourself picking pieces of the bucket off the cured refractory.
This is what the refractory looks like dry. Like coarse cement mix.
The first thing to do is to lay down the base. This should be several inches thick, thick enough that the opening of the cylinder is level with the opening of the bucket.
Adding the refractory. Again, note how dry the mixture is. The foil cap makes it easy to add the refractory without spilling any into the crucible.
About every inch or two of refractory, take a stick or something similar and gently tamp the mixture. Not too hard or the paper bucket will bulge.
The cylinder will tend to work itself off center. If you see that happening, gently tamp on the thinner side. That will work the cylinder back toward the center of the bucket.
Finished. Notice how the top is nicely rounded. The instructions say NOT to trowel the mixture smooth. I finished the surface by gently patting it with my gloved hand.
The refractory is allowed to cure for 24 hours (don’t skimp) and then the assembly is gently heated to drive out the residual moisture.
In this photo the assembly is connected to a 2.2 model Roy and the heat is set to about 100 watts. At this point the bucket is stripped off to allow the moisture to escape out the sides.
Here Matt adds the secret ingredient :-).
Actually it’s just water. We filled the crucible up with water so that it would boil for awhile and hold the temperature at about 210 deg F. That let the heat slowly soak through the refractory.
The refractory instructions say to raise the temperature about 100 deg F per hour. We did that until the interior of the crucible was showing some color. Then we attached the 2500 watt Roy and fired for effect!
Note that if there is enough interest, Fluxeon will sell ready-made furnaces in a variety of sizes. This is a good size for small castings up to about 5 lbs (allowing a pound for the sprue). Next on my agenda is a 20 lb furnace built inside a 5 gallon steel pail. Stay tuned.
Next Instalment: Using the furnace.
Posted by neonjohn on November 1st, 2014 under Cool Stuff, Induction heating, Projects |
13 Comments »
Print This Post : Print This Post PermaLink
One of the banes in the lives of HVAC and maintenance mechanics is the removal of corroded, rusted in place fans, pulleys and other objects from the shafts they’re mounted on. Gear pullers often do irreparable damage and acetylene torches can often overheat and warp. Induction heat is perfect for this application.
I’m scrapping an old RV air conditioner so my friend Matt and I decided to demonstrate how the Fluxeon® Roy induction heater can be used to ease the job.
This photo shows just how bad the situation is. The AC had been on the RV for almost 30 years. Then it laid behind my cabin for several years exposed to the weather, until I could find someone to help me move it.
Not only is the shaft rusted to the shaft but the set-screw is also rusted up. I bent an Allen key trying to turn it.
Finally, the work coil. Almost anything will do. In this case I found a hunk of electric motor hookup wire. I put some fiberglass sleeving on it in case the rubber insulation burned off and then hand-formed it to make about 3 loops around the fan hub.
This photo shows the other side of the shaft. Not only is the shaft heavily rusted but the end is mushroomed where someone had previously tried to remove the fan by beating on the shaft with a hammer. Therefore before we started, I took a Dremel tool and ground down the mushrooming on the end of the shaftSetscrew is now finger-loose
After just a couple of minutes of heating, the set screw loosened such that I could turn it with just a couple of fingers.
Let the heating begin! Notice the smoke rising from the hub area. This is oid oil and rubber insulation being vaporized. I heated the hub for about 3 minutes to about 900 deg F. The heat was very smooth and even, with no hot spots. Only the hub was heated. The blades remained cool and un-blackened.
Shaft is out. After the 3 minutes of heating, I laid one fan blade on my bench, held the opposite one and gave the shaft a gentle tap with a brass hammer. It fell right out!
This is what the work coil looked like after the heating. The rubber smoked a little but was not melted. The sleeving proved to be un-necessary.
The shaft after the fan is removed. One can see just how badly it is corroded.
The fan hub after removal. Notice that it is still tight on the blade spider, is not warped or ben and with the application of a little oil, is ready for immediate re-use.
The dent marks were made by someone else in the past trying to hammer the fan off without the help of Roy
After removing the fan, just for fun we decided to remove the sleeve bearing which was also stuck.
This photo shows the work coil I whipped out from some #8 solid copper ground wire and some siliconized fiberglass sleeving.
Work coil in place.
Turn on the heat! The smoke is from the oil burning out of the porous sintered Oilite sleeve bearing.
The bearing really likes induction heating! At this point, it is loose on the shaft and can be lifted up with a pair of pliers.
Bearing ready to be removed
The bearing, ready to be lifted off. Notice that the shaft didn’t get very hot at all. Yet another benefit of induction heating.
Induction heating makes the miserable and time consuming job of removing stuck objects from shafts easy and painless. For more info on the Roy© line of induction heaters such as the Roy 2500 used here, go to http://fluxeon.com.Posted by neonjohn on October 1st, 2014 under Electronics, HVAC, Induction heating, Misc |
Comment now »
Print This Post : Print This Post PermaLink
Recently my business partner, Garett Churchill surprised me with a little goodie. A 4 channel DSO that will fit in my shirt pocket.
We’ve been very busy lately so I didn’t pay it much attention other than to charge it via the USB port. Today I decided to take a look.
The first thing I noticed on the back was the Creative Commons symbol. That really perked this open source advocate’s ears up! The first thing I did was go to their website www.minidso.com (through Google Translate – the site is in Chinese. The whole site is a BBS/forum format which I found quite un-useful.
A mini-optical disk came with the scope so I popped it in the drive. There were many jpg photos that Irfanview could not render, though the Linux document viewer could. And a manual. And source code. And a schematic! This this truly is open source! Is that amazing or what?
The first thing to do was to connect it to a signal generator and see what it can do. The photo above shows it displaying a 500kHz square wave. The fluff on the edges of the square waves are caused my my less-than-perfect connection
Don’t hardly think that’s 50 ohms across there!
Next I connected to my Rigol DSO and compared the amplitude and time interval measurements. Right on the money. For a little scope that cost around $169 (sleazebay), I’m getting more and more impressed.
Another thing that is impressive is how easy it is to upgrade the firmware. No special PC software needed. Just hold a button for a few seconds and it mounts like any other drive. Here are the complete instructions.:
If you want upgrade the firmware of DS203, PLS follow these steps:
1. connect the DSO to PC with a USB cable ,then press
“>II” key when power on.
2. After a few seconds , you can see a “DFU storage”
in your PC , copy your firware file “*.hex” to the
“DFU storage” . Note that only one file can be
copied each time. And waiting for auto-reboot.
3.Restart the DSO.
Simple enough, eh?
I next took the scope out to my truck and repeated some of the tests I outlined here. The results were the same.
Needless to say, I’m pleased! Here are some more photos of the instrument
I really hate to recommend anything where sleazebay is the only source so I went looking. Good old Alibaba.com came through. Go here and type in “DS203”. Several vendors will be reurned. Bear in mind that the posted prices are very soft! Feel free to haggle. Be sure you get the version with the two probes, a carrying case and the mini-CDROM.
—-Posted by neonjohn on August 11th, 2013 under Electronics, Neon and other lighting, Product Reviews |
3 Comments »