Ball Hopper Monitor - Casting Project

[GreenTwin]

More Rob Wilson photos.

.

 

[GreenTwin]

More Rob Wilson photos.

.

 

[stanstocker]

I force-dried the coating with a heater, and sanded it with 400 grit sandpaper.
It is doing what I want it to do.
The black spots are air bubbles, apparently, but don't seem to make a big difference.
There will be a top coat of shellac.

I will probaby use a finer sandpaper, more like 600 or 800 grit, if there is such a thing.
This is a good method, so I will start spraying the hopper parts.
.

The foam backed fine sanding pads available from model suppliers, hobby shops, and amazon are wonderful for smoothing complex shapes, They can be washed after use, but do not require water in use unless you want to wet sand. Up to 5000 grit in case you want a mirror finish in mud

A search for "softback sanding sponge" on Amazon will take you down a small rabbit hole for a while.

Have fun!

 

[Lloyd-ss]

Photo pc240091 in thread #580 (above) gave me a bit of a start. That huge radial (centrifugal?) blower on the furnace looks like it could shoot flames thru the roof.

 

[GreenTwin]

The blower you refer to is shown in this photo.
Some folks use a VFD on their blower.
This may be a low rpm motor.




Here is what looks like an analogous commercial unit made by Mifco, and the combustion air blower is quite large.



A Mifco "B-301" is about the size of my first furnace, which is probably on the large side for hobby work.

I experimented for several years with different burner types and blower configurations.

Member "ArtB" is the individual who finally explained combustion engineering to me as relates to a foundry furnace.
Per ArtB, there is a fixed surface area inside of a furnace, say "X" square inches, and so there is some maximum gallons per hour of oil that can be forced into the furnace with an oil burner "Y" that can be completely combusted inside the furnace.

Once you get beyond Y gallons per hour, then the furnace can no longer fully combust all of the fuel that is being injected into the furnace, and any fuel added after "Y" does not burn until it exits the furnace via the opening in the lid.

When I started building a foundry, I assumed that with burners, bigger was better, and I also assumed that a higher fuel flow rate would burn hotter than a lower fuel flow rate.
As with many assumptions I made in the beginning, I was wrong about the fuel flow vs temperature.

I made a series of valves (a valve tree), so I could quickly test various fuel flow rates.
Each valve was calibrated with its own needle valve.
I could turn on 1, 2, 3, 4, 5, 6 etc. gallons per hour, and I experimented with flow rates up to 10 gallons per hour.



After much experimentation, I found that for a 13" diameter, 14" tall furnace interior, a diesel fuel flow rate of about 2.7 gallons per hour produced the hottest furnace interior, and produced the fastest iron melt.
I ended up calibrating my siphon nozzle (cold) to 2.7 gallons per hour, and then I adjusted my combustion air blower (which is a Toro variable speed leaf blower) to give about 4" of flame out the lid when operating (operating rich or reducing, to minimize oxidation of the iron).
This happens to be the air produced at the lowest speed of a Toro leaf blower.

Many large blowers have an intake damper, to allow adjustment of cfm into the furnace.
A large blower will outlast a smaller blower, due to the commercial motor.
I initially used the output of a shop vacuum, but shop vacs will wear out over time.
.

 

[GreenTwin]

I purchased a Grizzly dust collector motor, and made a dual siphon nozzle burner from it, assuming if one nozzle was good, then two were better.
This burner produced some serious flames, but it forced so much air into the furnace that it actually cooled the furnace.

At some point I am going to install this blower in my shop, and run a duct out to my furnace, so that I don't have the noise from a leaf blower right next to my furnace.
This blower will outlast my leaf blower.
.

 

[GreenTwin]

This was a dual 180 degree siphon nozzle burner arrangement that I tried.

A single burner tends to have a high velocity (the same velocity as what comes out of a leaf blower), and so that fuel/air flow tends to climb up the back of the furnace, instead of swirling around inside the furnace.
You can counteract this somewhat by angling your single burner tube down a bit.

Two burners reduce the combustion air flow to 1/2 of what a single tube burner sees for the same fuel flow, and so you get a much more evenly distributed ring of flame low in the furnace when using two burners at 180 degrees.

This burner would have worked well, had I known how to tune it correctly.
I was forcing too much air into the furnace, and that caused this arrangement to run too cool.
At some point I will go back to using a dual burner system, since they do work very well, and produce a faster melt.

.

 

[GreenTwin]

This is the Mifco data for their furnaces, with blower motor horsepower and cfm listed.

Mifco lists melt capacities for aluminum and brass/bronze, but you can also melt iron in a Mifco furnace.
The key to having a furnace last when operating at iron temperatures is to use a high quality refractory, such as Mizzou, which works well with iron.
There are some plastic refractories that are rated for 3,800 F, and those work better as a furnace hot face than Mizzou.

I am not sure what continuous temperature the interior refractory of a Mifco is rated, but brass/bronze melting temperatures are not much different than iron melting temperatures.
.

 

[GreenTwin]

The black spots in the sprayed-on wall joint compound are air bubbles.
I don't think they will be a problem, since they are small relative to the lines in the 3D print.
Sanding minimizes them.
I think the final coat of shellac will close these holes.

The good part about spraying on the wall joint compound is that it produces a very even coating on the surface of the 3D print that is thick enough to bridge over the print lines, but thin enough to not obscure the fine details of the print.
This is exactly what I was trying to achieve.

The problem I have had with using paint and shellac to try and fill the print lines is that these materials are so fluid that they just follow the print lines, and don't bridge them, so you end up getting taller peaks, and still have valleys and visible lines
.

 

[vk7krj]

Hi Pat, I know I have suggested this before and had the idea dismissed, but seeing the large amount of work you are doing to smooth these prints I thought I would try again.
Please spend a few minutes and watch this video- one of many on the subject, some much more complicated than they need to be-



It may save you many many hours of sanding.

Ken.

 

[GreenTwin]

Thanks for the tip Tasman.
I already have to deal with the resin-binder fumes, and I do that out in my shed, with either a chemical respirator, or a powered respirator.
But I really dislike the fumes; they get into everything, clothing, etc., and so I would be reluctant to add yet another high-fume process to the mix.

The PLA fumes from 3D printing have started bothering me, and so I am building a ventilated enclosure that vents outdoors.
I can't really hand any more fumes from anything.

It should be noted though that there is more to preparing these patterns than just smoothing the surface.
For instance, the logo on the gas tank has to be filled with wall patch, and then wiped down with a damp cloth, to create the fillets around the letters.

Other parts need more fillets than I can get Solidworks to add.
Solidworks is very particular about fillets, and for complex surfaces with a lot of curves, the fillets often do not work.
Any fillet that I can't add in Solidworks, I must add manually with a filler.
Sometimes it is not become completely obvious what needs a fillet, or perhaps it becomes obvious that a pattern needs a bit more fillet after printing, and so again those are added manually.

The slow-speed ceramic sponge goes a long way in smoothing the surface out before the filling process starts, and the sponge is a clean process that is easy to do.
The process is to add any required fillets with Fastpatch, and then spray on the sheetrock compound.
The sprayed-on sheetrock joint compound is working pretty well, and unlike paint, it does not run if you spray it on too thick.
Light sanding overall, and then a final cover of a few coats of shellac.

My process seems like a lot of work, but really the process is actually pretty fast, and it has no fumes, other than perhaps the shellac, which is not too bad.
There is no avoiding filling some areas with Fastpatch though.

Thanks for the feedback.
Perhaps I will try it one day on a more simple print.
.
Edit:
I use a space heater to dry the Fastpatch and joint compound, and so that speeds up the process a lot.
I can fill one pattern while the other is drying.

Edit2:
The sanding sponge in the slow speed drill also removes the sharp edges from the print.
.

 

[GreenTwin]

Buffed off the gas tank, filled the letters with fastpatch, and sprayed on joint compound.

I will probably sand it, and then spray it again.
.

 

[GreenTwin]

I have a feeling that as some have mentioned, flaking may be a problem.
I could mix the joint compound with shellac, since we know we can spray slurry, or just build up coats of shellac.
I will play around with the gas tank halves, and then make a decision.
Shellac builds up pretty fast, and can be recoated in as fast as 15 minutes.
For the spots that require fastpatch, that material will have to remain regardless.
.

 

[GreenTwin]

Here are the gas tank pattern halves after drying.

The sprayed on joint compound is as thick as perhaps four layers of shellac, and the coverage over the ridges is very good.
Sanding the compound only requires a very light touch with 400 grit sandpaper.

The difference between filling with shellac or a high-fill paint and filling with joint compound is that the compound sands very easily back to the defining edges of the 3D print, and so it is easy to maintain the dimensions and varioius shapes of the 3D printed pattern.

I think there is no choice other than to keep going with the sprayed-on compound.
One big benefit is that it only takes a few minutes to sand out each gas tank pattern half, with just extremely light sanding, so that really helps with hand/wrist fatigue.

I will spray on one more coat of joint compound, perhaps a little thinner than the first coat.
The final overall coating will be one or two layers of shellac.

If the filler cracks off when the pattern is pulled from the sand mold, that will not matter, since its work will be done, and it can be discarded as a temporary pattern.

I will go through the entire process with the joint compound/shellac/molding, and finally casting in aluminum, for the gas tank, to proof the entire process, before moving on to another pattern, such as the water hopper.

.

 

[GreenTwin]

For the water hopper, there were some compelling reasons to make patterns and coreboxes.

For this gas tank, there is no reason to make separate coreboxes, and so we will use these pattern halves as coreboxes.
Normally one would use coreprints, but we will dispense with those for this pattern.

The two mold halves will be rammed with bound sand, with some steel dowl pins inserted perpendicular to the parting line, and then the mold halves will be joined and clamped while the core sand is setting.

I will cast the holes in the pattern halves, and then use graphite rods to hold the core in place, one rod at the top, bottom, and either end.
One rod will be inserted for the gas passage into the tank from the boss on the bottom of the tank.
Some consideration will have to be made, since the interior of the tank will not be accessible after casting, so perhaps the gas passage core needs to be water-soluble, or perhaps it will break down from the heat since it is small.
A curved wire could probably be forced into the tank passage from the outside to break up that core.

Speaking of breaking up the core, the entire core inside of the gas tank will have to be removed, so all of that needs to be water-soluable.
A sodium silicate core will break up pretty easily using water if the percentage is kept relatively low (I forget the exact percentage, but I seem to recall 3%-5%).
People often increase the level of sodium silicate in their sand, because they are over-gassing, and that makes a ss core very difficult to break up after casting.
The bosses on either end of the tank are for core support and also for core removal.

Runners/gating on these pattern halves will be similar to what was used on the water hopper, with a "U" shaped runner, spin traps on each end, and two long knife gates either side.

Edit:
I need to remember to vent the core down the center, and perhaps vent out the larger fill boss, since the smaller bosses on the ends will have small graphite rods in them. The larger boss will also have a graphite rod, but it will be larger, and I can probably drill a vent hole through it.
.

 

[GreenTwin]

Thinking out loud............I remembered that I have an unused refrigerator in my shed, and I could put several 5 gallon buckets of sand in that, with a few light bulbs.
I am not sure I want to put the resin in their, since the refrig is still functional and my be used one day for food.

I also recall that I have about 50 feet of industrial heat trace cable, and so I could find a container (such as a metal garbage can) and put the resin in there with the heat trace.
The heat trace is 120V self regulating, so that would be easy.
Now that I think about it, I salvaged a hot water heater, and it is out in the yard, so I can cut the top off of that and use that to store things.
It is insulated too, so that is a plus.
I save hot water heaters to use the metal as the outer shell for foundry furnaces.

I am getting close to needing to make some molds, and so I have to get everything ready for that.
I really pays to keep the foundry stuff together all in one place, and I have not been very good about doing that, but will make sure I do it this time.
I will head out to the shed tomorrow on a mission of discovery, to find everything I need to run the furnace, and the items I need to make sand molds.

I am going to melt with propane.
I don't feel like dragging out the oil burner rig, and I don't need diesel to melt aluminum.
My propane burner is self-aspirating, and so all I need is the burner, a hose, regulator, and 20lb propane tank.
The propane rig is easy to drag out as one unit.

I guess I will use my large furnace, and although it is oversized for this purpose, it is well insulated.
The outside of the furnace remains cool to the touch when the furnace interior is at iron temperatures.

I need to find my molding table.
Hopefully I did not get rid of that during the last shed cleanup.
.

 

[GreenTwin]

I am going out to the shed, and will attempt to fire up the furnace today.
I have to find everything.

The gas tank pattern halves are almost ready for an aluminum casting attempt.
I have sprayed on the 2nd coat of joint compound, and they are looking good.
I had one chip-off spot, but I think there was a wet spot on the table that stuck to the undried coating, and it chipped off when I lifted it.
No problem, I can touch up that spot.

I will sand these off, and if I can find my shellac today, I will start that coating process.

I am still working on an exact joint compound mixing/spraying technique.
I changed to a taller container, and that is working better.
And I am working on calibrating an exact mix of compound and water every time.
It is better to mix up the slurry before putting it in the container.

I have some slight sag in a few places, but that sands out easily, so not like a sag in paint, which can be tricky to sand out.

Its time to fire the furnace and melt some stuff.

.

 

[GreenTwin]

Living in a forested community is very scenic, however the falling leaves are rather a pain.
I just cleaned up what may be a million, or perhaps a billion leaves on my driveway.
Luckily my combustion air blower is a Toro leaf blower.
I can see concrete driveway now, so that is a start.

Located most of the foundry equipment, and moved it out of the shed.
I still need to find my small scale.

Fired up the furnace and ran it for a while empty, to dry it out.
Some water has gotten into the outer ceramic blanket layer, and that is giving off clouds of steam.
I may have to remove the ceramic blanket to dry it.
I need to build a doghouse to completely cover the furance, to keep it dry, and the doghouse needs to be all metal so that it can be moved over the hot furnace after a melt.

I experimented with reducing the fuel flow, since 2.6 gal/hr melts aluminum too fast; ie: you don't have time to use the pyrometer.
I went down to perhaps 1 gal/hr (not sure exactly), and the burner was still stable.
I will use the oil burner for melting aluminum and iron.
I have plenty of diesel, and a finite amount of propane, so no sense using propane for the aluminum melts.

I did not get any molding done, and I did not find my molding table.
I can use a piece of plywood.

I am still gathering everything that is needed for a melt/pour.
.

 

[GreenTwin]

Interior.
.

 

[GreenTwin]

I finished sanding the gas tank pattern halves.
I applied a coat of shellac, and unfortunately it seems that shellac disolves most of the joint compound.
I am not sure about the fastpatch material, it seems to be holding its own.
No big loss, I can work with shellac only.

I will probaby start using Durham's to fill thicker areas.
I am pretty sure Durham's will stand up to shellac.

I used fastpatch only on the water hopper, so it should be ok with shellac.

I think I will go through the entrire process of casting the aluminum pattern halves for the gas tank, and then cast the tank in iron, to proof-of-concept the entire process for one item.
Perhaps that will save some time on the remainder of the patterns, should any problems occur that we need to be aware of.

.