Yet another keg furnace

Nice!

Jet A is actually kerosene. Quite a bit lighter than #2 Diesel, and still lighter than #1 Diesel, but not by much.

 

It's been a while since I looked up the energy content of it, just that it was very close to diesel. If I end up running a lot of it in the furnace, I'll blend some oil in to lubricate the iron gear pump (just in case).

 

Mark have you tried veg oil? Where I live I can get as much as I want.

 

I did with the rotary cup experiments and was able to light it from a cold start, due to the fine atomization. There's a local fish and chip shop with 400 litres in 20 litre drums if I want it. It would make sense to start on Kero/diesel and then once the furnace is hot switch to cooking oil. It wasn't that hard to filter: just used an old cotton sheet pegged across a 200L drum opening to filter the crumbs and sludge overnight.

 

I tried veg oil in my furnace and it started without adding diesel. I cannot do that with waste oil as it is too thick.

 

I'll have to run some more fuel experiments with the current setup, I haven't had the time to try many fuels. The spinning cup worked on just about everything I could throw at it, but then again they are used with bunker oil on ships.

 

No. 6 bunker fuel is some heavy stuff. It has to be heated to flow down a pipeline.

 

Here's my efforts from a few days ago casting brass, I may try a larger blower than the jumping castle units now that I have the combustion working the way I want it to. I have a bolted flange with sealant on the tuyere to fix a major oil vapour leak and if you look closely in the video when the furnace lid is open, there's now vapour and flames on the top left of the furnace as vapour works it's way between the stainless keg and the refractory: there seems to be an oil vapour leak near the tuyere pipe, with vapours traveling to the top of the furnace now that the tuyere pipe seals properly. Apart from the leak it all runs fine and I'm very pleased with it's operation.

 

Thanks! it's optimized for resin bound sand and aluminium, I should have modified the runner to be a bit fatter. When used with aluminium it develops a shrink crack that lets you snap off the runner from the casting: pure chance!.

Here's the first time the pattern made a mold:


Being cast for the first time:

 

I forgot to mention the secret trick to getting such a resin mold to seal properly: as soon as the sand resin mix is hard enough to tip out ~ around 5 minutes or so, it's slightly floppy and a bit soft: you assemble the mold on the bench and the two halves conform to each other and fully cure hard. This ensures a good seal against the molten metal. When you are making a batch of resin bound molds they all get assembled when just strong enough to do so, for this reason.

 

Resin sand molds are the cat's meow as far as making clean and strong molds.
Unfortunately it can be tricky to buy small amounts of the resin/hardener/catalyst.

The resin molds do move slightly after strip time.
I generally leave them in the cope and drag until they fully set, and place them on a flat surface during that time.

Combining the cope and drag before final set is also a good idea (I have not tried that) as long as the bottom of the drag mold is perfectly flat, and is sitting on a perfectly flat surface, else they will settle slightly uneven.

I use furnace cement to glue the cope to the drag, and apply it along the edge after the mold is closed.
I have gotten a good seal between the cope and drag as long as the molds are kept on flat surfaces until the final set.

That is a great looking part.

 

I'm really pleased with how it came out, maybe increased runner size and a graphite coat would improve it.

Come to think of it I did make the pattern edges flat and true, a quick sand on a flat surface would work, the two molds wouldn't need to be parallel, just flat enough to avoid twisting when assembled on a flat surface. Buying resin is getting really hard, there is a product called Desmodur RFE which is a cross-linking additive: it drastically boosts cure speed, strength and heat resistance of quite a few glues especially urethane. The sanding belt joiners use it and quite a few other industries. It has a very short shelf life and is sensitive to light and moisture. While it's toxic, it breaks down to CO2 and water and is probably the reason no bake resins have such a short shelf life.

It may allow the use of normal urethane glues for casting purposes, it only has to hold together long enough for the metal to solidify then disintegrate to allow easy removal.

 

I've just been machining the spindle casting for a couple of 6204 ball bearings. So far it looks great, there's been a few minor blemishes but no massive flaws caused by two flows meeting and not mixing. The first photo is with the camera xenon flash, the second is with a cool white LED worklamp: you can see the strong yellow component of the blue-yellow illumination. Third is the outer housing after I went berserk with some files, I'll finish up with bead blasting and clear lacquer.

 

Lately the tropical monsoon weather has solidly rained me in, so there's no chances to run the furnace at all. The last few times the various 1/2 to 2/3rd Hp jumping castle air blowers have all run out of puff when trying to melt iron. I've had trouble getting a high pressure centrifugal blower locally, the type that are relatively thin but have a larger diameter. So at this point I was offered a suitable centrifugal impellor that came from a vacuum pneumatic pipe system used in shops to move those tiny capsules with money and documents around the shop. The original pump used a 7.5Hp 2880 rpm motor with three impellors on the one shaft all connected in series, I think to get the necessary vacuum. At any rate it's a well built sheet aluminium impellor that would take ages to make....all I have to do is make the housing, easy right?. Most of the blower materials are free items lying around, just need some fasteners and the extrusion used for the spacer.

It measures 65 cm across or about 26", it's too big for my needs but the impellor needs about 2.5Hp at 2800 rpm to function.

 

Mark
The casting looks very nice. I hope mine come out someplace close to that when I get things up and running.
That is a BIG blower impeller. I have a nice 2 HP blower that I used as a dust collector and vacuum some years back but have not since found a use for it. It may go back into service for what I was using it for but as it was setup. The collector had a footprint the same size as a full size pallet. I don’t know that I want to deal with something that size again
We missed all the rain but got 4 -5 inches of snow followed by frigid temperatures. Today high only made around +12F, tomorrow forecast calls for a high of 36F. Unfortunately by Wednesday the high is predicted for -10F. I HATE WINTER! Our summers seldom reach 100F though.
My shop is around 15 miles from the house we are now living in so I don’t make it out as much in bad weather. I did make it out yesterday for a couple of hours to work on the fuel tank and pump.
In Australia your cooler weather is a ways off. I do hope you do not see the type of cold we do.
Joe

 

Thanks Joe, that casting ended up being usable but had some defects where the flow didn't join up properly. They were completely hidden until it was bead blasted and the surface bulged a bit from the thin brass stretching from the bead impacts....nothing that can't be fixed or filled over. The impellor is about 40mm or 1 & 9/16" thick so the housing looks wide to compensate for the thinness: the diameter should result in a relatively higher pressure blower compared to a small diameter, wider unit. It should be able to supply a respectable level of combustion if it goes well. The metric 40mm channel spacer is too narrow so I had the brainwave of adding an extra sheet of that 4mm or 3/16" aluminium to cut holes in to recess the impellor hub and the four mounting bolts for the motor flange.

My little gear oil pump off a car engine has popped the overload breaker three times now, it looks like the 1440 RPM, 1/2Hp motor is not enough power when the fuel regulator is set to 55 PSI. It'll run for 15 minutes before popping so that may give you some idea of motor power requirements for your fuel pump. I wouldn't mind experiencing the cold for a short while, late last year the backyard reached 113F/45 deg C and according to the news the last few days have been the wettest in 130 years and I can remember they said the same for 2010 too. A nice temperate workshop climate would be hard to beat, even late at night it's like a sauna I have several fans mounted on the ceiling now to help a little. The coolest it's ever been here on the coast is 50 deg F so I compensate by wearing Tee shirts with thicker fabric.

 

Mark
I think you should slow the pump down by 2 or 3 to 1. The ratio now looks to be around 1/1 so the pump is turning at 1440 RPM. This type of pump setup would be well suited for moving a large volume of oil in an engine but the volume is not needed for a fuel furnace.
You could cut your pump RPM to less than 500 RPM and still have an excessive amount of volume available without over taxing your motor.
1 Hp will develop 1200 PSI at 1 gallon a minute or 1200 PSI at 60 gallons an hour.
The Hp to move the amount of fuel used in a furnace is very small. You can get 120 PSI at 6 gallons an hour with .01 HP, + any inefficiencies
I have moved away from the pump drive scheme I had originally started. The Rube Goldberg machine and have decided to go with a 12VDC fuel pump that puts out at or just under 10 PSI maximum.
I will be using this pump to supply fuel only to a siphon nozzle. Compressed air will be used for atomization.
Last night it was - 17F and today the high was -7. Not a lot is going on until the weather breaks.
With any luck it will be up in the 40s in a couple of weeks but you can never tell for sure.
Best of luck with your pump. The setup looks good it is just spinning a bit fast.

Joe

 

Hi Joe, I have tried it at 720 RPM with a 1:2 pulley and it ran fine, but I could not exceed 40 PSI with the pump. This is probably due to pumping low viscosity Kerosene as the pressure gets much higher when pumping used cooking oil (65 PSI +) at 720 RPM. I'm in two minds as to what's happening as the setup ran fine at 40-45 PSI/1440 RPM for hours on end but you can hear the motor change sound when adjusted to higher pressure of 50-55 PSI. The pump gets reasonably warm at higher pressures so I suspect there's some churning/slippage in the pump when run at higher pressures. It's a compromise as a smaller gear pump would easily supply the flow rate and pressure with less return via the regulator than a worn car oil pump.

If I can find a small pump for kerosene at 55 PSI then I'd be happy, maybe a modern, smaller 4 cylinder engine oil pump that uses a positive displacement 5 lobe rotor in a 6 lobe stator:

 

It's a very close fit for the impellor due to the 40mm channel thickness: I had to cut a circular hole at the rear for the hub to go through and then cut a second cover sheet which was glued and screwed to the first to cover the hole. I made a steel sleeve which was shrunk onto the motor shaft as much as possible, the shaft has a step and a mangled left hand thread so that area is filled with epoxy to try and give some more strength. The impellor has a 35mm split bore with two clamping bolts to hold it on. After checking for any wobbles or damage, it all seemed fine and I gave it a quick short power up: very loud and seems to function like an air raid siren!! . Will cut a hole for the impellor intake and assemble it and see how well it works as a siren. I'll run it with the impellor vertical and stay behind it in case it explosively defenestrates.