Injected Diesel 56cc 2 Stroke, Will it ever work?"

[Lloyd-ss]

Diesel Injector Dilemma

After much effort and many "almost works," prototypes of a unit injector, I am about ready to admit defeat and compromise on the issue by marrying an off the shelf injector nozzle with an off the shelf injector pump. The parts will need to be modified to remove excess size and weight.

I think I will be going with the GM (Detroit Diesel) HV6 injector nozzle, and the Yanmar injector pump from their line of LA single cylinder engines (199 to 406cc).

The HV6 nozzle has much smaller holes and the internal spring can be changed to achieve different injection pressures.

Here is a pic of the 2 nozzles (GM HV6 and Yanmar LA series.) I listed some comparison specs from both stylesof injector nozzles.


What baffles me is that the Yanmar is for a much smaller cylinder volume, and operates at over 2,000psi, but has fairly large nozzle holes.
The GM operates at below 1,000 psi, has has a much larger cylinder volume, but its nozzle holes are much smaller. So much to learn, LOL.


-----------------------------

Here is a pic of the HV6 nozzle and the Yanmar injector married together. I hope they are compatible. The design of the nozzle and the pump are fairly straight-forward (famous last words, Ha)
There is still a LOT of excess weight and material on the Yanmar pump, but I will eventually get around to putting it on a diet to make it suitable for the model I plan on making.



Testing is next. At first glance, it seems to offer resistance against the plunger from the pressure build-up, just like it should.
This is just a proof of concept prototype, unless it works, that is.
Lloyd


x

 

[ajoeiam]

Diesel Injector Dilemma

After much effort and many "almost works," prototypes of a unit injector, I am about ready to admit defeat and compromise on the issue by marrying an off the shelf injector nozzle with an off the shelf injector pump. The parts will need to be modified to remove excess size and weight.

I think I will be going with the GM (Detroit Diesel) HV6 injector nozzle, and the Yanmar injector pump from their line of LA single cylinder engines (199 to 406cc).

The HV6 nozzle has much smaller holes and the internal spring can be changed to achieve different injection pressures.

Here is a pic of the 2 nozzles (GM HV6 and Yanmar LA series.) I listed some comparison specs from both stylesof injector nozzles.
View attachment 139918

What baffles me is that the Yanmar is for a much smaller cylinder volume, and operates at over 2,000psi, but has fairly large nozzle holes.
The GM operates at below 1,000 psi, has has a much larger cylinder volume, but its nozzle holes are much smaller. So much to learn, LOL.


-----------------------------

Here is a pic of the HV6 nozzle and the Yanmar injector married together. I hope they are compatible. The design of the nozzle and the pump are fairly straight-forward (famous last words, Ha)
There is still a LOT of excess weight and material on the Yanmar pump, but I will eventually get around to putting it on a diet to make it suitable for the model I plan on making.



Testing is next. At first glance, it seems to offer resistance against the plunger from the pressure build-up, just like it should.
This is just a proof of concept prototype, unless it works, that is.
Lloyd


x

Very interesting!!!!!!!

I'd say you got some good progress going!!

Any chance you would get a measurement on the holes in the nozzles?

TIA

 

[Lloyd-ss]

Very interesting!!!!!!!

I'd say you got some good progress going!!

Any chance you would get a measurement on the holes in the nozzles?

TIA

Joe,
Here is a picture of the Yanmar nozzle with a bare copper wire .006" in dia, and a silver plated copper wire of.008" dia. The .006 fit in without much trouble, but the .008 took some effort. Both wires came back out of the holes without getting stuck.If you look at the picture in post #121, you can see that the GM HV6 nozzle has holes that are significantly smaller. I did not attempt to fit a wire into those holes.
Lloyd

 

[Lloyd-ss]

GM nozzle and Yanmar pump. A marriage not made in heaven. Test Video.

This video is from a few days ago after I got this proof-of-concept set-up built. Proof-of-concept is a handy euphemism for putting something together quickly and enthusiastically without too much attention to detail because that would just take too darn long!

It is still too big and bulky, but it works better than the last build, the stroke length and timing need to be worked with (plenty of adjustability built into this version), as does the cam profile. And the excess fuel volume and air agitation at higher rpm needs to be addressed. So little time and so much to do! Right now I feel like I am just throwing stuff at the wall to see what will stick, but that is fine and I am enjoying it. I hope it is entertaining, or laughable, too.

 

[Lloyd-ss]

A UNIT INJECTOR THAT FINALLY WORKS.
With Video

It's been 3 months since I posted in this thread, but a lot has been going on. Especially the number of so-so results.
But I might finally have something. It is a diesel "unit injector", where the pump and injection volume control, and the injector nozzle are all in a single unit, like those used in the old Detroit Diesel 2 stroke engines. It is pumping, and injecting, and there is a plan in place for the fuel volume control, which will be managed by a governor.

I did cheat a little and use the Yanmar nozzle housing (with the holes), but not the needle. All the other parts are mine. It is set up with an adjustable rocker so that the length of stroke and its timing relationship to the fuel inlet port in the pump/plunger body can be fine tuned. It is fairly fussy about the timing and how much stroke remains after the port is closed. This information should be helpful when it is integrated into the engine.

The test set-up is fairly awkward and the unit injector is only a tiny portion of it. The injection seems to be working well. The nozzle has 4 holes although it only looks like two in the video.

 

[Steamchick]

Video didn't play for me, but "we'll done" on your success.
I understand modern common rail injectors follow a similar principle but with a feed at a few bar pressure, with the injector pumping the high pressure from an electrically powered piezo pump in the body, to give precise metering and timing.
The second part you will need to develop will be an equivalent injector mounted powered pump perhaps?
K2

 

[Lloyd-ss]

The video plays for me. I will try putting in as a YouTube link.

Video added and it seems to work ok.

The injector pressure needs to be increased and shimming the internal spring should take care of that.

 

[Lloyd-ss]

This picture shows a few more details of the unit injector. A low pressure supply pump might be needed, but right now, the gravity feed is sufficient.

 

[minh-thanh]

Hi Lloyd-ss !
Looks great !
Please update ...

 

[Lloyd-ss]

I tried adding a .010" shim to the internal spring, but I think that used up all of the travel and put the spring into coil bind. Time to take it apart, take some measurements and see if I need to make a new, stronger spring that still has enough travel.

 

[Lloyd-ss]

This is the spring I had to replace. The new spring had to fit around a .078" tube and inside a .148" hole, and used .033 dia wire x .240 long. I still had to grind the O.D. just a tad to get smooth action. AND, it fixed the problem. I will get a picture of the internals next time I take it apart.

The internal parts are already in place to control the volume of each stroke. Adding the external parts (rocker and adjustment and drive pin) will be next. As it is now, it is mainly a single stroke pump with a differential pressure discharge poppet,

 

[Lloyd-ss]

Unit Injector. I think this version is going to work!

Finally! The junk box is full of progressively better iterations of the unit injector, that is high pressure pump and injector nozzle in a single unit.

As I said previously, I used 3 parts that were factory-made. One is the actual GM injector nozzle (just the housing, not the pintle), and, the Yanmar piston and housing (both of which required extensive modification and "body sculpting."

I have a few bugs to work out, but this time my confidence level is at 90%.
It has a low pressure fuel feed line, and a return line. I might need a low pressure pump on the feed (20psi or so), but for the moment, it is a suction feed. The pump high pressure piston has a modified angled notch in it for the injection volume control, and a small lever that will be governor controlled, to twist the pump piston and notch as needed. It is not tiny. I will call it robust.
Lloyd

 

[Lloyd-ss]

Video of the Unit Injector in operation

Finally, I am pretty happy with the way it is turning out. Still plenty to do, but no big gotcha's.
Lloyd

 

[Nerd1000]

Hey Lloyd,

That's pretty exciting. How does the atomisation look?

 

[Lloyd-ss]

Hey Lloyd,

That's pretty exciting. How does the atomisation look?

Nerd,
From what I can see at the moment, the atmomization looks pretty darn good. It is hard to see in the video, but it squirts in 4 directions in an almost horizontal fan shape, maybe 15 degrees down from horizontal. The first several strokes are clearly a "fog", but then the fuel starts condensing on the jar and it starts to drip down the sides. I will take it outside where I can let it spray in free air without stinking up my shop. A flame test, although probably not necessary, will be fun.

The stroke volume control lever which rotates the plunger and its helix, works well, but I need to quantify the delivery rates. That will mean hooking up the electronic sewing machine motor I have with a digital RPM display

https://www.homemodelenginemachinis...st-a-3-4-hp-sewing-machine-motor-video.34494/
Check out the video at reply number 30 of that thread.

The volume of fuel needs to be between zero and maybe 4 cubic mm per stroke. Right now, I can't tell how completely the fuel is shutting off when I move the lever to the off position. I know I will need to install a fuel shut-off somewhere in the supply line. I have also noticed that the system seems to be self-priming which might mean that no low-pressure supply pump will be needed. That would certainly be a bonus.

Lloyd

 

[Roger B]

Flame tests are fun

How does the spray pattern match your combustion chamber?

 

[Lloyd-ss]

Flame tests are fun

How does the spray pattern match your combustion chamber?

Well, seeing that the only "real" parts I have so far are the blower and injector, and a blank sheet of paper, I have the luxury of designing the perfect combustion chamber around the spray pattern, LOL. This is becoming a very new-agey, organic process.

I will get a better picture of the spray pattern when I do a paper test, and a flame test. But the variable speed test motor needs to be incorporated into the test set-up before I can get any meaningful empirical data.

 

[Lloyd-ss]

I couldn't wait any longer. Its a pretty day, perfect for a flame test.
No idea what it proves, but yes, it was fun.

 

[Steamchick]

Hi Lloyd, I understand the desire to develop combustion chamber designs from scratch, but you could try some very interesting research - by Mr Ricardo.... I fact he was so good after years of research and study that his basic designs reached the conclusions that led the world for half a century or more.
I like to "stand on the shoulders of giants", rather than re-invent the same thing - poorly, if at all.
But good luck in your endeavours. It is a Hobby, so ENJOY whatever you do.
Incidentally, the spray pattern you see is but a poor representation of what happens in the combustion air as adiabatically heated, swirling, at a pressure of maybe 25~28bar, and even much higher as the combustion heats it at explosive (sonic) speeds. - A bit like comparing paper planes to a Mach 2 Concord?
Simply: Because of the technology available to Ricardo to pump fuel, the pressure was low (by today's standards) and he developed the pre-combustion chamber, that is either in the cylinder head or piston crown, and this is a zone of very rich burning fuel and smallish amount of air. The "flames" of very hoy ionised fuel, air and exhaust gas developed in the pre-combustion chamber then blast into the "main" volume of air to complete their combustion, and heating that air to the pressures that power the piston.
Ricardo's papers make interesting reading...
K2

Some odd bits you may enjoy?
https://en.wikipedia.org/wiki/Harry_RicardoBy Bartłomiej Bulicz - &lt;span class=&quot;int-own-work&quot; lang=&quot;en&quot;&gt;Own work&lt;/span&gt;, <a href="Creative Commons — Attribution-ShareAlike 3.0 Unported — CC BY-SA 3.0" title="Creative Commons Attribution-Share Alike 3.0">CC BY-SA 3.0</a>, <a href="File:Side valve engine with Ricardo's turbulent head 02.png - Wikimedia Commons">Link</a>

https://www.dailysportscar.com/archive/subscribers/technical/ricardodiesel.htm
https://www.dailysportscar.com/archive/subscribers/technical/ricardodiesel.htm

 

[Steamchick]

Aha!
Found the article I was looking for...!
k2
https://oldmachinepress.com/2013/01/02/inside-the-cylinder-of-a-diesel-engine-by-harry-ricardo/