I decided to build the machine shop out cast aluminum as described in the books. I had an advantage that I already had a good sized drill press and a small Unimat DB200 and the usual collection of hand tools along with a well equipped wood working shop.
One item I didn't have was a wood lathe, useful for making patterns
that would later be turned on the metal lathe so I borrowed ideas from
Fine Woodworking magazine and whipped one up over the Christmas holidays.
The spur drive and dead center were purchased and I had a friend thread
the tailstock.
In order to cast the bits needed for the lathe I decided that the Power
Hacksaw designed by David's son, Vince, would be a welcome addition to
my shop. I made an inventory of all the steel required for the projects
and bought them at my local Metal Supermarket. Some of the hacksaw frame
I gas welded and the rest I had a friend MIG weld. The result is a power
hacksaw that can cut unattended and leave me to play in the shop while
the metal bits are being created.
With the hacksaw complete I started cutting the various bits of hot
and cold rolled steel that I would need for the lathe, mill, shaper and
furnace frame. The furnace requires several round forms to hold the castable
refractory so the Slip Roll seemed like an appropriate project and this
took several years and a move before I completed it. Only after using a
metal cutting bandsaw was I able to finish cutting the 3/4" thick end plates.
Once again a friends milling machine helped make things exact but as it
turned out the precision wasn't as critical as I thought as the rolling
former worked well even with the small alignment errors. The same friend
did the MIG welding and with the completed Slip Roll was able to create
the forms for the furnace seen behind the roll.
I dusted off the hacksaw at this point and cut the raw materials for the foundry frame. After a quick trip to the welding supply house for a refill on my gas welding supplies and I welded the base frame together. My MIG equipped friend welded the bits that were unworkable with Gas due to the furnace design. Three bags of 2800 degree Ladlecast refractable and a few hot days doing the preliminary baking in the household oven finished off the furnace.
At this point I realized that the almost 150 lbs of furnace and fan would be too difficult to carry around so out came the hacksaw again. I borrowed a small 125A MIG welder, picked up a helmet and taught myself to blob filler material onto Hot Rolled Steel. The frame with casters and the burner assembly are the result. With a width of lest than 28 inches it will fit through most standard doors and will ultimately sit underneath the molding bench.
Click on the photo above to see a larger photo of the burner. You can just see the small Micro-processor based controller by the FAN. The micro is based on a CANDIP (www.candip.com) which uses an ATMEL 8515 uC and has RS232 and CAN outputs. There are 3 power transistors: one for the Spark plug coil drive; one for the Gas Solenoid and one for the PWM motor control.
The furnace is started by closing a contact which starts the motor turning at slow speed; enough to just create airflow into the furnace body. A flow sensor checks for air movement and if it's there then starts the spark plug device generating a spark every 10ms. A short time later the gas solenoid is turned on and if the flame detector sees UV light it then stops the spark and speeds up the motor to burn speed. Then entire operation during testing would take almost 15 seconds but the latest tests have it igniting within about 5 seconds. If the UV detector sees the flame vanish then the gas solenoid is closed and the FAN speed increased to a high rate purge cycle for 10 seconds. Then the fan speed is dropped back down to ignition cycle and the processor tries to relight. Three strikes and no flame and a shutdown is forced.
With the squirrel cage fan a reasonable blue flame was created inside
the furnace but the air resistance with the plug for the lid installed
caused the furnace to extinguish regularly. 
The
theory at the moment is that the junction between the Burner Pipe and the
Venturi section causes turbulence and as the furnace heats up and combustion
backpressure increases the flame front moves into the burner pipe. It appears
that the squirrel cage is not powerful enough to sustain combustion and
propane pressure cannot be increased above 4.5 PSI.
So why build a machine shop? Two reasons. The education received while fabricating the tools is immeasurable. Tools that make repairing things like this:
There's more to come... stay tuned.
