Diesel nozzles, EDM

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.
Small hole EDM uses hollow brass or copper tube as electrode and uses deionized water pumped thru the rotating electrode/tube at high pressure, 2000psi in order to preserve the electrode from rapid heat disintegration and flush out the swarf. Most shops commonly keep electrodes down to .25mm and specialty edm shops go as low as .1 mm regularly and the process is quite fast in hardened steel... perhaps 10- 15mm/min., a little slower and more troublesome in soft metals. If the part was fixtured at the correct angle and easily indexable, actual burn time would be a few minutes and use a fractional portion of one electrode that costs less than $5. The machines tables have small t-slots like a mill table for aligning fixturing and hold downs with t-nuts.
 
Thank you for the wealth of replies.

My thoughts on EDM is that it would allow for smaller holes then I can mechanically drill in to a hardened pintle style nozzle. If I make the nozzle out of free machining steel I could probably mechanically drill the jets but I want to make the nozzle out of something corrosion and wear resistant like a high chromium steel.

My goal is burning vegitable oil at high temperature, so corrosion is a consideration. Everything I consider suitable chemically and thermally, is too hard for me to believe I can drill sub mm holes reproducibly.

I'm months away from even starting cutting metal so there is no rush on a solution, and maybe a bigger drilled nozzle will work better.


Definitely the lack of info on miniature diesels is frustrating.
 
Small hole EDM uses hollow brass or copper tube as electrode and uses deionized water pumped thru the rotating electrode/tube at high pressure, 2000psi in order to preserve the electrode from rapid heat disintegration and flush out the swarf. Most shops commonly keep electrodes down to .25mm and specialty edm shops go as low as .1 mm regularly and the process is quite fast in hardened steel... perhaps 10- 15mm/min., a little slower and more troublesome in soft metals. If the part was fixtured at the correct angle and easily indexable, actual burn time would be a few minutes and use a fractional portion of one electrode that costs less than $5. The machines tables have small t-slots like a mill table for aligning fixturing and hold downs with t-nuts.

I need to look into if there is such a shop near me and get pricing. Thanks!
 
Thank you for the wealth of replies.

My thoughts on EDM is that it would allow for smaller holes then I can mechanically drill in to a hardened pintle style nozzle. If I make the nozzle out of free machining steel I could probably mechanically drill the jets but I want to make the nozzle out of something corrosion and wear resistant like a high chromium steel.

My goal is burning vegitable oil at high temperature, so corrosion is a consideration. Everything I consider suitable chemically and thermally, is too hard for me to believe I can drill sub mm holes reproducibly.

I'm months away from even starting cutting metal so there is no rush on a solution, and maybe a bigger drilled nozzle will work better.


Definitely the lack of info on miniature diesels is frustrating.
Have you considered the MAN M-system design? It quite deliberately doesn't atomise most of the fuel, instead the fuel is squirted onto the surface of the piston bowl and intense air swirl strips layers of vapourising fuel from the resulting film.

This is the method that was used in the multi-fuel diesels of the US Army's 2 1/2 ton trucks.
 
Have you considered the MAN M-system design? It quite deliberately doesn't atomise most of the fuel, instead the fuel is squirted onto the surface of the piston bowl and intense air swirl strips layers of vapourising fuel from the resulting film.

This is the method that was used in the multi-fuel diesels of the US Army's 2 1/2 ton trucks.
No, I actually had not heard of that system, that is very helpful, thank you. I'll go read up.
 
I need to look into if there is such a shop near me and get pricing. Thank
The key would be to provide a fast set up with a good fast indexing fixture that only takes a minute for the operator to set up. It must connect the workpiece electrically to the table. Find out the t-slot dims and make it fast to install the fixture and quick to index each burn location for them. The reference pick up is done by extending the electrode from its diamond guide a bit and touching spinning the electrode to the workpiece where conduction is made, an electric conduction edge finder. Should be a sub one hour job. Each burn will be less than a minute easy once set-up and picked-up.
 
The simple solution to me is a hole - sensibly small - with a smaller wire up the bore...Then the fluid passes through the annulus and the gap is incredibly small compared to the hole diameter... Will is work? - The "wire core" could be a part of the valve... perhaps? - I am sure I have seen this used commercially on something so it isn't anything original... (I don't do "clever").
K2
Yes it works, I tried it in my first diesel engine

75910-20210114.jpg
 
This is my latest needle style injector. The actual seal/orifice is between the tapered end of the needle and the edge of the 0.2mm bore. The seating angle at the end of the nozzle is greater than the angle of the needle in an attempt to get as close to a line seal as possible. Keeping everything concentric requires some thought. The 2mm bore in the nozzle and the seating are cut with a homemade D bit and finally the 0.2mm hole is drilled so it has to penetrate less than 0.5mm. Keeping control of the depths is also interesting.

New Needle Injector.jpg
 
Have you considered the MAN M-system design? It quite deliberately doesn't atomise most of the fuel, instead the fuel is squirted onto the surface of the piston bowl and intense air swirl strips layers of vapourising fuel from the resulting film.

This is the method that was used in the multi-fuel diesels of the US Army's 2 1/2 ton trucks.
I wish I had know about this sooner, it sounds perfect!

I wonder how well it scales.
 
I think a piston-burn chamber was useful for also keeping te piston crown cool, as it extracted heat in vaporising the fuel - and the "wet" fuel shielded the piston locally from flames....
But with the disadvantage of a THICK crown - and consequently much more mass to accelerate and decelerate, etc. = lost work.
Head pre-combustion chambers as easier to control for temperature (inside the water cooling of the head), and fixed, so permit pistons to be optimised to a much lower Mass, Balance, etc.
Just my ideas - for what they are worth?
K2
 
Is that the reason for your head burn chamber vs a piston burn chamber?
Partly. A precombustion chamber engine is also much easier to design than a direct injection one, I didn't have to worry about things like swirl ratios in the cylinder or the exact piston bowl geometry, just a spherical pre-chamber in the head and a passageway of appropriate size to connect it to the cylinder. Also I just wanted to be different from the engines other people had already built!

I think for making an engine that just runs it's all a bit immaterial, like Minh said his engine still runs well with several types of injector, and it has neither a bowl in the piston nor a prechamber. But if one wants to extract significant amounts of power without emitting excessive quantities of black smoke the combustion system needs some consideration. I myself don't have any particular plans on that front other than to drive something like a little generator (light some bulbs, charge a phone, etc) or a pump to demonstrate that the engine is 'capable of work.' But I enjoy the design engineering part of the project as much as building it, so I always will tend to want to do these sorts of things.

I think a piston-burn chamber was useful for also keeping te piston crown cool, as it extracted heat in vaporising the fuel - and the "wet" fuel shielded the piston locally from flames....
But with the disadvantage of a THICK crown - and consequently much more mass to accelerate and decelerate, etc. = lost work.
Head pre-combustion chambers as easier to control for temperature (inside the water cooling of the head), and fixed, so permit pistons to be optimised to a much lower Mass, Balance, etc.
Just my ideas - for what they are worth?
K2
I'm going to take an opportunity to 'Nerd out' a bit here. Diesels have one big fundamental problem: Only a mixture of gaseous fuel and air can burn, but the diesel injects liquid fuel into the cylinder at the very moment that we want ignition to occur. So we require some method to vapourise the fuel extremely rapidly, and we also need to very rapidly mix the resulting vapour with the air. All of this has to happen while combustion is already starting!

Most diesels solve the vapourisation problem by atomising the fuel into a very fine mist, the resulting high surface area to volume ratio allows the fuel to be very quickly heated above its boiling point by the air. The exception to this is the M-system I mentioned in a previous post, which instead relies on surface evaporation from the piston. Anyway, even if we were injecting gaseous fuel it can't burn if it all winds up in one little spot in the cylinder with no air around to react with. So there are basically two ways to get the fuel in contact with as much air as possible: the first is to spread the fuel mist around by firing jets of it in all directions, such that some fuel gets to every part of the combustion chamber and mixes with the air that is already there. This is the approach you might encounter in a very large low speed diesel, like in a ship. Unfortunately manufacturing constraints on the tiny holes in the injector make this infeasible for smaller engines, you just can't have enough holes of small enough size in the injector tip to have a jet aimed everywhere. This brings us to the second approach: if we can't bring the fuel to the air, let us bring the air to the fuel. By making the air move around the combustion chamber really fast, we can ensure that air goes to where the fuel is. So the opposite extreme is the precombustion chamber: there is only one hole in the injector, but we place it in a chamber in the head that is connected to the cylinder by a narrow orifice. During the compression stroke air is forced into the prechamber through this narrow orifice, resulting in it flowing around the prechamber at very high speed and mixing with the fuel which is only being introduced in one spot. Unfortunately, it isn't feasible to get all of the air into the prechamber, but on the other hand it won't take long for the pressure in the prechamber to start forcing already vapourised fuel out through the orifice at high speed to mix with the remaining air in the cylinder once combustion begins... Mercedes Benz notably leant into this fact by having multiple orifices aimed around the chamber in their precombustion chamber engines, but many others (including myself) use just a single hole offset to one side.

Most direct injected diesels use a bit of each method. So we have 3-5 holes in the injector, and a bowl shaped combustion space in the piston crown. (Why in the piston? Mostly because the valves are occupying the prime real estate in the head directly above the piston, if we were building a loop charged 2-stroke diesel we could put it in the head instead). The intake ports are carefully shaped (usually in a spiral) to induce a fast swirling motion of the air as it enters the cylinder, as the piston approaches TDC on the compression stroke the air is forced into the bowl and (thanks to conservation of momentum) the rotation becomes even faster. This motion is nowhere near as fast as what the prechamber method would produce, but we have multiple fuel jets so we don't need to get all the mixing to happen by air motion. Each jet from the injector reaches a sector of the rotating air mass, eventually leaving us with a roughly even mixture! Once again the M-system is the weird exception, it relies on very fast swirling motion (induced by its inlet ports and very deep and narrow bowl) to bring the air to the fuel film, honestly it's more like the prechamber method in that respect.

So why choose the direct injected 'mix of both' method rather than prechamber? After all the prechamber design has a simpler one hole injector, which could save us some difficulty, and has no need for funny shaped intake ports or heavy bowl shaped piston crowns. The main answer to this is efficiency. The prechamber design forces ~50% of the gases in the cylinder through a rather small hole (in my engine, a 4mm hole connects the prechamber to the cylinder. Compare that to the 16mm diameter of the inlet valve) on every compression and power stroke, which results in a lot of extra pumping work. It also has about 50% more surface area exposed to the compressed charge at TDC, resulting in greater heat losses to the cooling system. So the direct injected engine has a higher potential for fuel economy and (to a certain extent) power. Once we get to really big engines (marine diesels) we can even dispense with both, and just use an injector with a large number of holes- this is even more efficient, as the funky port shapes and fast mixture swirl of the 'mix of both' method are also a source of losses.
 
An excellent summary :) 👍

I initially chose a loop scavenged two stroke as I wanted to keep the combustion chamber as small as possible to minimize heat loss so I could compress the air to ignition temperature. The design was loosely based on the Petter Atomic diesel engine.

This had a number of problems, firstly the charging efficiency of the two stroke was not very good and I had to reduce the combustion chamber volume significantly to obtain sufficient pressure/temperature. I was able to get some firing by dripping ether into the air intake. Secondly the airflow/swirl in the cylindrical combustion chamber didn’t suit the injector spray pattern. As far as I can tell it only ran on the fuel that blew past the piston into the crankcase where it was then inducted into the cylinder and fired (a sort of carburettor). The third problem was the amount of unburnt fuel that came out of the exhaust and spread over a wide area.

It's rebuild into the current four stroke form was an immediate success although there is still a lot of unburnt fuel in the exhaust.
 
I find this fascinating. reinforces my existing knowledge base. Thanks Nerd for such a straightforward explanation. _ I simply didn't try and confuse matters with "Swirl"... But you covered it all really simply. One of the diesel engines I encountered (A Nissan car diesel, 2 ~2.5 litre 4-cylinder 4-stroke) had the intake to one side of the head and exhaust on the other - e.g. RHS-in, dog-leg, LHS-out - which was to generate more swirl around the combustion chamber on a small direct injection engine. (in- swirl left, spiral around the chamber - then double-back to exit out the exhaust. This was the only modern engine I encountered that didn't have a "straight-in, straight-out" layout of intake pipe, valve, Chamber, exhaust valve and pipe. All those rely upon other features to generate swirl. But I also encountered a petrol engine - 4 valves per cylinder, where 1 of the pair of inlet tracts had a butterfly valve that closed at low throttle settings, to induce more swirl when there was less air-fuel charge in the cylinder. Around 1/4 throttle the butterfly valve opened so it ran as a simple 4 valve combustion chamber. Symmetrical and opposite handed flows of gases through inlet-chamber-exhaust.
Cheers!
K2
 
I find this fascinating. reinforces my existing knowledge base. Thanks Nerd for such a straightforward explanation. _ I simply didn't try and confuse matters with "Swirl"... But you covered it all really simply. One of the diesel engines I encountered (A Nissan car diesel, 2 ~2.5 litre 4-cylinder 4-stroke) had the intake to one side of the head and exhaust on the other - e.g. RHS-in, dog-leg, LHS-out - which was to generate more swirl around the combustion chamber on a small direct injection engine. (in- swirl left, spiral around the chamber - then double-back to exit out the exhaust. This was the only modern engine I encountered that didn't have a "straight-in, straight-out" layout of intake pipe, valve, Chamber, exhaust valve and pipe. All those rely upon other features to generate swirl. But I also encountered a petrol engine - 4 valves per cylinder, where 1 of the pair of inlet tracts had a butterfly valve that closed at low throttle settings, to induce more swirl when there was less air-fuel charge in the cylinder. Around 1/4 throttle the butterfly valve opened so it ran as a simple 4 valve combustion chamber. Symmetrical and opposite handed flows of gases through inlet-chamber-exhaust.
Cheers!
K2
Those swirl flaps often appear on diesels too, if they are 4 valve types.
 
This is my latest needle style injector. The actual seal/orifice is between the tapered end of the needle and the edge of the 0.2mm bore. The seating angle at the end of the nozzle is greater than the angle of the needle in an attempt to get as close to a line seal as possible. Keeping everything concentric requires some thought. The 2mm bore in the nozzle and the seating are cut with a homemade D bit and finally the 0.2mm hole is drilled so it has to penetrate less than 0.5mm. Keeping control of the depths is also interesting.

View attachment 157867
Hello, these plans are complete?
 
Unfortunately I am blocked from that site for reasons that nobody can explain.
I have sent numerous inquires using various methods and email addresses, and for whatever reason, they don't respond.

The site is mostly blank for me other than text.

Wish I could see it, but perma-blocked apparently, and no way to get a response from anyone on that site.
As I mentioned before, its a closed club, and items posted there are not necessarily visible to outsiders.

.
 
Back
Top