Pressure and Siphon-Nozzle Style Foundry Oil Burners

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Well-Known Member
Staff member
HMEM Supporting Member
Global Moderator
Jul 2, 2021
Reaction score
MidSouth, USA
A question came up about oil burners for foundries, and while this information I think is all contained somewhere in my foundry thread, it is a bit difficult to dig out.

I will pick out the oil burner info, and post it here.

Use extreme caution if you construct an oil burner, and wear any and all required safety gear to ensure your own safety.

My post from a recent oil burner thread:

As 100model mentions, you can use a drip-style oil burner, and you don't need a siphon or pressure nozzle at all.

I built a drip-style burner just like 100model's burner, and after many attempts to make it work correctly, I could never get good control over it, and it caused fuel puddling in the bottom of my furnace. I gave it a very intense effort to make it work, but I cannot get a drip-style burner to work correctly with my furnace, and I know a lot about oil burners.

So why do some folks use pressure or siphon nozzles when a drip-style burner will work?

1. A drip-style burner requires a propane start, and a siphon or pressure nozzle burner running on diesel does not.

2. A siphon or pressure nozzle burner are very controlable, and have a wide turn-down ratio, ie: they will operate over at least a 4:1 fuel ratio.
A nozzle rated for 1 gal/hr will also operate well at 4 gal/hr while still achieving good atomization.
I am aware of one individual who uses an Andrino controller, with a thermocouple and control loop, to control the output of two siphon-nozzle burners for his aluminum scrapping furnace. This arrangement gives him automatic and precise temperature control of his melt.
You can't do this with drip-style burners.

3. My siphon nozzle burner is extremely stable in operation.
I use 10 psi pressure on the fuel tank (with a 30 psi safety relief valve), and this gives rock solid operation.
I literally never adjust my burner; not during a melt, and not between melts.
There is much to be said for never having to adjust a burner.

4. I have started my siphon nozzle burner on diesel down to 30F, and it starts and operates instantly, with no propane warmup required, and no preheating of the fuel (the diesel tank and the diesel in it are at 30F).
A siphon and I think a pressure nozzle burner running on diesel are impervious to a wide outdoor air temperature range.

There is a white paper out there that was written by Delevan's lead engineer, and he goes into great detail about the various types of oil burners; how they work; and how they differ.

Keep in mind that there is no "best" burner type.
Each burner type has its own pro's and con's.
One has to select the style of burner and fuel that best suits their particular situation/application.

Last edited:
Basically, with a siphon-nozzle burner, you feed compressed air in the center pipe, and fuel in the side of the adapter via an elbow fitting.

There are two parts to a nozzle (referring to a siphon or pressure nozzle arrangement), which is the adapter body, and the actual burner tip.

Below are the internals of a Delavan siphon nozzle tip.

Below a is schematic diagram for a Delavan siphon nozzle burner.

I use 10 psi compressed air pressure to pressurize the fuel tank (be sure your tank will hold pressure if you do this), with a 30 psi safety valve.
The 10 psi gives a consistent diesel fuel flow at all times regardless of fuel tank level or any other variable.

I use an automotive inline fuel filter in the fuel line just before the burner.

I use about 30 psi compressed air to the siphon nozzle (atomizing air; this is labeled "AIR IN" in the schematic diagram below).
Some folks use slightly less compressed air, but I have found you can have fuel puddling if you get much below 30 psi atomizing air.

I use a #30609-11 nozzle, and a #17147 adapter with the siphon nozzle burner.
The chart below lists the #30609-11 at a nominal 1 gal/hr fuel flow rate, but it actually operates over a wide range of fuel flows, depending on your compressed air pressure, and fuel tank pressure if you are using a pressurized fuel tank.

I have seen folks melt iron with diesel, kerosene, cooking oil, waste oil, etc., and they all seem to work well as long as they are clean and well filtered, and will flow to and through the nozzle without clogging.

Some hydraulic oils are reported to be toxic when burned, so use caution with what you are using for fuel.

Waste oil can be contaminated with sludge, water, antifreeze, heavy metals, etc., which is why I avoid using it.

Note that the Delavan siphon nozzle is called a "Siphon Type SNA" by Delavan, not to be confused with a "Pressure Nozzle", or other types of nozzle styles.

Also note that the Delavan siphon nozzle and the Delavan pressure nozzle both screw into the Delavan #17147 adapter, however, siphon nozzle and pressure nozzles are piped up differently.

A siphon nozzle uses a siphoned oil input line, and a compressed air line.

A pressure nozzle uses a pressurized oil input line (at about 100 psi), and generally a return oil line to the fuel tank.

I am not aware of any spray angles offered on a siphon nozzle, but the pressure nozzles are offered in a range of spray angles, and can also be purchased with solid or hollow spray cones.




Last edited:
Here is the Delavan "VARIFLOW" pressure nozzle, and it looks very much like the siphon nozzle, but it operates differently.

A pressure nozzle operates exactly like a pump perfume bottle, ie: a pump provides pressure to a liquid, and the liquid is discharged under pressure into the air in an atomized spray.

As I mentioned before, the siphon and pressure nozzles both use the same adapter, but these nozzle types are piped up differently.

This pressure nozzle operates at about 100 psi fuel pressure, with no compressed air.
The fuel pressure is what provides atomization as the fuel flows out the nozzle.

Typically a gear pump is used to provide the pressure for a pressure nozzle, but as 100model has shown in some of his reference videos, some are using automotive fuel pumps instead of a gear pump.

I have been using a Delavan siphon nozzle for a long time, and I am pleased with how it operates, however, it does require compressed air to operate, and I like to take my furnace to various iron-pour events, and do demonstration pours, to show the art-iron folks how to melt iron with diesel instead of coke.

Some exhibition sites do not have compressed air, and sometimes they don't have much in the way of electrical services either, and so I am in the process of building a Delavan pressure nozzle style burner.
The intent is to eliminate the need for an air compressor, and just use a very small fractional horsepower motor to power a gearpump.

A pressure nozzle burner with a gearpump would be easy to transport, and would only require a tiny 120 volt electrical circuit, which could be provided with a small generator if necessary.

This motor is 1/3 horsepower, and is oversized, but it should last forever even if run continuously due to the very small loading on it.
If I had to do this build over again, I would use a slightly smaller motor.

One thing that 100model mentioned is that a gear pump will not tolerate any trash in the fuel at all, and so I will use the spin-on inline fuel filter ahead of the gearpump, and will probably use another fuel filter of some type just ahead of the burner.

Note that any fuel filter downstream of this gear pump must be rated for 100 psi or more, since that is the output pressure of this gear pump.

This gear pump is generally used in a commercial fuel-oil burning heating unit.

It will operate at 1725 rpm, 100 psi, 7 gal/hr maximum output.
I will use it at 2.6 gal/hr, and return the balance of the 7 gal/hr to the fuel tank.

I used the lowest available rpm gear pump, to help with long term reliability.

I will operate this pump at 7 gph, 100 psi, 1725 rpm, which is one of its standard output ratings.

My oil burner only burns 2.6 gal/hr, so I will feed the excess oil back to the fuel tank.

Note that the fuel line and any filters on the output side of this pump must be rated for continuous use at over 100 psi.


Last edited:
As far as the spray angle that I will use with the pressure nozzle burner, there are limited angles that I have been able to find, so I just purchased the most narrow pressure nozzle that was available.

Solid cone seems to be the best for a furnace application.
I will dig out the statement from the Delavan white paper about when to use a solid and hollow cone pressure nozzle.

Here is more information about the Delavan pressure nozzles, and the ones I purchased.

Looking at my purchase sheet, it looks like I purchased all Type "B" solid cone nozzles.
Some had a 60 degree spray angle, and some had a 45 degree spray angle.
And the rated fuel flow on some of the nozzles was 2.5 gal/hr, and on others was 3.0 gal/hr.

I basically purchased a variety of pressure nozzles, with some to be used as spares.
I will start initially with a 45 degree 2.5 gal/hr nozzle, and I think that will work well for my furnace.

This is the stainless burner tube that I have built for my pressure nozzle burner.

I experimented with the fins cut into the end of the burner tube (as shown in the photos below), but ultimately I cut the end of the burner tube off, and I don't use any fins internal or external to the burner tube.

This particular tube is thinwall stainless, 2.5 inch diameter.
I have also used 2.5" steel muffler pipe, and it works well too, but cannot be left out in the rain without rusting.

I used a Lincoln tombstone-style AC welder with nickle rods for the welding on this tube.

When I weld muffler pipe, I turn the welder to the lowest setting (40 amps), and 6013 rod, with an in-out motion to control the heat and prevent melting through the thin wall of the muffler pipe.

The end cap on the burner tube above was originally set up for a drip-style burner, as an experiment.

I will modify this end cap to accommodate my pressure nozzle and its associated fuel feed pipe.

The cap is sealed to the burner tube with a stainless plumbing pipe seal, which has the rubber removed.
The seal between the cap and the burner tube does not have to be airtight, since all you are sealing is low pressure combustion air from the leaf blower, and any minor leaks to not affect the burner performance.

Sometimes I use a piece of HVAC aluminized tape to seal the fuel pipe to the burner cap fitting, but I have used a compression gland in the past for 1/4" tubing.
Again, if this seal is not perfect, you are just leaking a bit of low pressure combustion air from the leaf blower.

The control circuit that I plan on using is as shown below.

Note that I drew my control circuit over a commercial packaged oil burner diagram, so only pay attention to the plumbing, not the other items shown in the diagram.

I am not 100% positive I have the fuel flow circuit correct, but I think it will work.

The design intent is that if the needle valve is fully closed, then 100% of the fuel will flow to the nozzle (7 gal/hr), and none will flow back to the fuel tank.

If the needle valve is fully opened, most of the fuel will flow back to the fuel tank, and little or no fuel will flow out the nozzle.

By adjusting the needle valve, I hope to get my desired 2.6 gal/hr fuel flow, with the excess fuel flowing back to the fuel tank.

As I understand it, from James Olson's white paper titled "Fuel Nozzles for Oil Burners", a hollow cone pressure nozzle is generally used in applications that use 2.0 gal/hr or less in flow, while being discharged into a relatively wide combustion chamber. (see pages 8 and 9).

For fuel flows higher than 2.0 gal/hr, and for a more narrow discharge into a narrow combustion chamber, a solid cone (and often a more narrrow discharge angle) pressure nozzle is used.

I interpret a narrow cone as being better for use with a foundry furnace, since you don't want fuel or flame impinging on the side of the crucible.

In general, the thing that is desired is to have a burner (regardless of the type) that fully combusts all of the fuel inside the furnace, with no smoke, and no soot deposits.

It is also desirable (for me) to have a burner style that is adjustable over a wide range of fuel flows.

And it should be noted that I have used my siphon nozzle Delavan burner with and without a combustion air blower (without a combustion air blower is called a naturally aspirating setup).
A siphon nozzle oil burner does work well in a naturally aspirating arrangement, but use in this configuration is limited to melting metals that have a lower melting point than iron.

Here is a link to James Olson's white paper about oil burners.

What would the benifits be with a set up like this if only melting aluminium?
I have watched a 25kg cast iron pour where the iron was melted with propane. Very large chunks of brake rotor were added to the melted iron and they melted rapidly. The man who had this set up has now rebuilt his foundry to burn oil. He wasn't having much luck at the time I visited him. I will send him a link to this post as it could help him.
The only benefit of using an oil burner to melt aluminum is speed of melt.

I sometimes use aluminum 356 to make permanent patterns, and I often use my naturally aspirated propane burner for smaller aluminum melts, since it is a very simple arrangement, and easy to setup and takedown.

For larger aluminum melts (which I generally don't do), or if you are just in a big hurry, then one can use the oil burner.
I have brought 20 lbs of aluminum to pour temperature (about 1,350 F) in 12 minutes, from a cold start.
Or melting aluminum in cold weather where you may have problems with the propane tank getting too cold.

For routine aluminum melts for most folks, a naturally aspirated propane burner is more than sufficient.

Gray cast iron is not that much more difficult to melt than aluminum, IF you have your burner and combustion air blower set up correctly.

Molten gray iron is a LOT hotter than an aluminum melt, and one has to wear a lot of leathers, use heat shields in strategic places, and protect the eyes from the intense infrared by using shaded welding goggles.

GreenTwin, this is a wonderfully useful collection of information - so helpful in understanding the confusing array of oil nozzle choices. I do still have one question: am I right in thinking that the nozzles you have ordered in post #9 require a body to screw into, where the body has the feed and return lines?
For a Delavan, and perhaps some of the other manufacturer's nozzles, both the siphon nozzle and the pressure nozzle screw into the same adapter, which is shown below. It is a Delavan #17147 adapter.

If you purchase a siphon nozzle, you bring compressed air into the open end of the adapter, and oil into the side of the adapter.

If you purchase a pressure nozzle, you bring the input oil line into the open end of the adapter, and bring the return oil line out the side of the adapter.

Here is a video about a Delavan siphon nozzle burner.

Note that I don't use the propane option that is shown in the video below, since it is not needed when using diesel.
I don't like dumping propane into an oil burner tube, since it can backfire in a dangerous way if you are not really careful.
I don't use propane with any of my oil burners.

The spin vane that he shows in the video below is not needed, and I don't use one on any of my oil burners.

I would highly recommend not using any PVC or rubber on an oil burner.
These items will melt sooner or later.


Last edited:
Hey Green,

Do you know anything about the possible effects of pulsed fuel flow on a pressure fed burner?

My furnace is too small for a commercial gear pump. The gear pump I made hardly builds any pressure (clearances too large), so I'm inclined to try a piston pump which is far more manageable in terms of the tolerances I have to maintain. The downside is of course that a single cylinder piston pump will only deliver fuel 1/2 the time.

My current thinking is that it should be OK so long as the residence time of the burning fuel in the furnace is significantly longer than the pause between squirts from the pump, but I'd appreciate any other opinions or advice.
Hey Green,

Do you know anything about the possible effects of pulsed fuel flow on a pressure fed burner?

My furnace is too small for a commercial gear pump. The gear pump I made hardly builds any pressure (clearances too large), so I'm inclined to try a piston pump which is far more manageable in terms of the tolerances I have to maintain. The downside is of course that a single cylinder piston pump will only deliver fuel 1/2 the time.

My current thinking is that it should be OK so long as the residence time of the burning fuel in the furnace is significantly longer than the pause between squirts from the pump, but I'd appreciate any other opinions or advice.
I made my own gear pump using small gears from an epicyclic gearbox and is only 0.18cc/rev and works a treat.
You can see it at Free CAD Designs, Files & 3D Models | The GrabCAD Community Library
There is a link to my YTube vid there too.
If you do a piston pump then its easy to make a double acting unit but the downside is more check valves,