Hydraulic Injection Pump And Injector
-leading to-
The Eventual Design and Building of A True Diesel Engine.
Go here for Part Three, Finally Making Some Engine Parts.
Go Back Home. 
25 August 2017:
Getting started on the injector parts, I made what I -think- will work for a pintle (teensy poppet valve).
     The little collet and the drill rod blank.                                Grinding the small diameter pin for the head.
Center punching for the head bore.                                               The (hopefully) finished pintle.
All of this took the whole morning.  After I was done, I quit for the day with a bit of eyestrain.

Since the chuck in my lathe will not close down smaller than 0.100", I had to make a funky collet.  I used an end of 1/2" bar stock and drilled it to accept the 0.069" drill rod.  I then slotted it so it would squeeze in the chuck to hold the workpiece.  This worked pretty well.

The first thing I did was to use the Dremel "toolpost grinder" to grind a smaller diameter on the end of the rod for the head to slip over.  A piece of 1/8" drill rod was used for the valve head.  I made some executive decisions on some of the dimensions to be able to actually build the parts.

After facing the 1/8" drill rod with the toolpost grinder, I used a centering punch in the tailstock to mash a center point in the rod.  Then it was drilled with a #55 drill to fit the smaller diameter on the stem.

Once the head was semi-pressed onto the end of the stem, the protruding smaller diameter was peined over to lock it into place.  

Lastly, the spring retainer clip groove was ground in the stem and the stem was cut to length.

In the injector, I am relying on clearance between the stem and the injector body to allow fuel to flow to the tip.  Also, the clearance will allow the tiny poppet to float to find it's best seat.

At least, that's how I think it will work.
26 August 2017:
The lower part of the injector body is done.

Injector lower body with pintle.
I didn't get as much done today as I thought I would.  While hogging the diameter of the lower injector body, the primary cog belt in the lathe broke.  While replacing it, I found a couple of issues relating to that part of the lathe that had to be addressed.  The part did get done.

Because the pintle was a relatively tight fit in the body, I used some fine alumina lapping compound to get the fit better and take a first shot at lapping the valve.  Working in these small dimensions is not easy.  First, the runout in the lathe chuck is about 0.0005".  For normal work in larger fractions of an inch, this fades into insignificance.  For dimensions in a few thousandths, it has to be contended with.  I hope I can chase this to a point where the parts will fit well enough to work.

I hope to finish the injector tomorrow.
29 August 2017:
I've finished the injector and started testing it.
Itsy-bitsy retainer washer.                                                                Funky first effort spring.                                                                        Under test.        
I took a wild guess as to the strength of the cracking pressure spring and have it quite a bit too strong.  I can peg the gauge and it doesn't do a thing.  Tomorrow, some changes will be made and further tests done.  I think the valves in my pump need a little more work, as the pump bleeds down to a couple hundred psi after a few seconds.

Making these itty-bitty parts takes more time than making the bigger ones.

Oh, yes.  My friend Denis Basson from Australia brought to my attention that I am not properly describing the stem thingy.  It is NOT a pintle thingy but is a poppet valve thingy, a totally different animal.  I will be much better at the terminolgy after I get farther along in my apprenticeship. :-)
30 August 2017:
After lapping the injector poppet for a couple of hours, I finally got it to sort-of spray but it still needs more work.  Then, the bolt that holds the banjo fitting on the injector broke.  I'm calling it a day.

El Drato!
The banjo bolt is an 8-32 machine screw.  Since I have to drill it for a fuel passage, it was weakened to the point that, after several tightenings for bleeding and testing, it finally gave-up.  Tomorrow, I think I will change it to a 10-32 to give more strength.
31 August 2017:
Today, I upped the size of the banjo bolt to 10-32, re-drilled the injector top and tapped it, Drilled the top bolt axially with a #55 drill and then used the same size drill to drill the radial hole to meet it.  Drilled out the banjo for clearance.  Made two new copper sealing washers and put it back together.  Leaked like a sieve.

After lapping washers and generally fiddling around with it, it still leaked just enough so I couldn't get to anywhere near the cracking pressure I was shooting for.  Finally, I just kept tightening the bolt with less and less leakage until it started twisting off.

I quit for the day.  Maybe I need to go with something like 1/4-28.  It will look sorta klunky but will be the same size as the banjo bolt on the pump.  
6 January 2018:
Wow!  It's been a while since I visited this project. In the meantime, I worked on The Scotch Yoke Engine, The Upside Down Engine and did a few Honeydew projects.

After thinking about it a lot, I've decided to change the injector to a true pintle type with a needle and seat at the nozzle.  It will use differential pressure between the force holding the needle seated and the hydraulic pressure trying to force the pintle up in it's guide.  the fit of the pintle in the injector itself will be very close.  Any leakage (and there will be some) will be drained back to the fuel tank or pump reservoir.

The latest injector design.

Because of the necessity of having to drill a fuel passage beside the pintle bore, I had to enlarge everything including increasing the bore from 1.000" to 1.125".  This will allow the engine to be easier to build and will give roof for valve springs.

I may start on dimensioned drawings tomorrow if nothing ultra-stupid comes to mind.
7 January 2018:
Well, naturally, I found something ultra stupid in yesterday's work.  There was no way to access the cracking pressure spring for assembly.  I made the required changes and am back to where I was yesterday.

This is a little better!
16 January 2018:
Not a whole lot to report.  I do have the injector drawn-up and dimensioned and am now killing time getting up the nerve before trying to build it.  Dimensions are on the small side.  The injector pintle is 1/8" in diameter and the spray hole is 0.028" in diameter.  It will be a challenge to machine everything to fit correctly!

While I was wasting time, I started drawing up the rest of the engine.  What you see here isn't cleaned-up for presentation but it should give you a pretty good idea of what I am contemplating building.

It should look a little like this when done.
To give you some idea of the size of it, here are some dimensions:
- Bore: 1.125"
- Stroke: 2.5"
- Swept volume: 2.48 cubic inches
- Compresion ratio: 20.8:1.  Variable downward to 11.3:1 by removing up to 0.100" from shims on the rod.
- Total height: 18.5"
- Flywheel diameter: 10"
24 January 2018:
This project seems to be dragging so I guess I should try to get back to it.  A couple of days ago, David from Texas sent me part of an injection pump and an injector tip from an International Harvester WD-40 engine.  

In the past he has sent some very helpful information including his idea of a single valve hit and miss Diesel.  This would solve several problems with sizes of the parts.  First, having only a single valve would give more room for the injector.  Second, fuel volume would be constant so there would not have to be the complication of varying the cut-off timing.  Adding to the simplicity would be the fact that I could combine the pump and injector function into one unit, pushrod operated.  More thinking is required.

There are also some helpful thoughts coming from SMOKSTAK.COM.  I started a thread there -  LINK  - asking for injector information and am being blessed with a lot of knowledge from that source.

Okay, having done with the blather, here are some photos of the WD-40 parts.
The WD-40 parts.
Just to give an idea of the size of the parts, the pump plunger diameter is roughly 0.329" (8.37mm) and the cast iron pump barrel outer diameter is about 0.669"(17mm).

The nozzle outer diameter is 0.545" (13.85mm) at the tip end.  The pintle major diameter is 0.236" (6mm), the step diameter is 0.098" (2.5mm) and the pin diameter is 0.039" (1mm).

I will have to get the bore and stroke of one cylinder or the displacement of the WD-40 to see how it compares to my tiny volume to see if there is any way I could scale down the tip to work with my design.  It's gonna be a headache to machine in any case.

I'm still thinkin' on it.  It's getting to a point where I think I may be able to build a hydraulic injected Diesel.  If any more people say it can't be done, I guess I'll have to do it.  After all, Find Hansen has proved it can be done so this redneck should be able to make it work, too.  No sense in leting Fred get all the glory!  :-)

Now, I think I'll go out and listen to stack music from The Upside Down Engine whilst my gears turn.  Maybe the engine will drown out the grinding noise coming from my head.
27 January 2018:
More ideas are coming to me after looking at even more injector tip types.

Injector types.
29 January 2018:
Here are a couple of drawings of tips I've come up with.  Whether or not either one will work is a question.

I've learned how to make really small orifices from a SMOKSTAK contributor.  His method is to press a small plug into a larger hole in the part you want to make the really small hole in.  Before pressing the plug into the part, a scratch is made in the length of the plug.  When it is pressed into the hole, there is a very small orifice created.  Of course, it is not round and totally predictable but it is a small hole.  It will also not be quite concentric but what is wanted is a fine spray.

Some Bright Ideas.
I've also thought of fitting a plug into a hole in an injector tip that has a tiny clearance around it which will, in effect, be a small orifice.  With the small amounts of fuel to be delivered, the tip could possibly be made by punching the plug through a hole in the tip that is slightly smaller than the plug.  When the plug is in the hole, there is theoretically a very small clearance.  If not, the tip could be polished a bit for clearance.  I could use it in conjunction with a pintle but getting the parts concentric could be a problem.

If a really simple injector will work, one with a simple nozzle and check valve, here is one that could work with an insulin needle, soft soldered to the injector tip then carefully cut to length and the end polished.

In the search for a small diameter orifice, I went out to the shop and attempted to remove a needle from a disposable syringe.  They are somehow glued or epoxied into the plastic housing and I did a number on the needle getting it separated.  The O.D. of the needle is measured at 0.0145" but I can only guess as to the I.D. at about half of that or about 0.007" or so.  What I need to know is if I can get the needle to tin using soft solder.

Here's the insulin needle.  
It's getting close to the time to shoot the engineer and start production!
30 January 2018:
Today, in the mail, I got the donated John Deere injector promised by Frank Heins.  Thanks, Frank!
  The disassembled John Deere pencil injector.                               On the test stand with adaptor soldered on.
The inlet fitting of the injector must be metric because I had nothing that would fit it.  The diameter and thread are close to 1/4" NPT but no cigar.  I cut off the flare and made a brass adaptor that went from the 5/16" diameter inlet tube of the injector to the 1/8" copper tubing of the fixture and soldered the joints.  I cleaned up all the parts and put it back together, setting the cracking pressure to around 1,200 psi.  The larger threaded part adjusts the spring pressure exerted on the pintle to set the cracking pressure and, as near as I can tell, the small screw and pin contact the end of the pintle when the pintle opening limit is reached, limiting the maximum fuel injected.  I guess this adjustment is made on the test stand.  I have it backed-off.  

Either this injector is really worn or, as I suspect, it doesn't like the small volume I am going to have to flow through it.  The pattern of the three jets is more a squirt than a spray.  Being able to take it apart and have a look at how it's constructed is very helpful.

I may make a quick and dirty fixture to see if the insulin needle will produce a spray.  I think if I can get the velocity of the stream high enough, it should break-up into a mist when it hits the dense and hot air in the combustion chamber. 
31 January 2018:
Today, I fiddled with the hypodermic needle idea.  

The teeny needle on the left and the itsy bitsy needle on the right.
This morning, I was helping  my neighbor who is diabetic and mentioned my project.  He volunteered to give me four of his diabetic needles which are even smaller than the ones I had.  The needle on the right measures 0.010" O.D. and I can only guess that the I.D. is around half of that.

I decided that I would make a brass "jet" so I can solder the needle in the smallest hole I can drill.
              Drilling a #65 (0.035") hole partway through.                                                              Redrilling a #70 (.028") hole to where the head will be.
I found a tube of #75 drills (0.020") but have no chuck that will fit a drill that small.  Drilling these small holes is tedious even in brass with the lathe running full tilt.
          Here's my test fixture for the needle idea.                                                                The "jet".                                                              The needle prepared to be soldered into the "jet".
The test fixture is made from a 1/8 tubing fitting, a length of 1/4" bar stock and a length of 1/4" hex brass bar stock.  It is threaded 6-32 and screws into the end of the round bar stock part.  The round bar stock part is soldered into the tubing fitting.  After tinning both the needle and the "jet" and blowing excess solder out of the "jet", the needle was laid into the hole and soldered into place.  I found that the needle could be removed from the plastic part by using the heat gun to soften the plastic.  The needle could then be withdrawn from the plastic part without bending or otherwise damaging it.

It's hard to see but the thing only squirts.  No spray.
I am disappointed in the way it works.  Rather than a mist, it just makes a stream.  I suppose the velocity isn't high enough for the stream to break up into droplets.  There are a couple of things that are probably causing this.  One is that there is no pop-off to allow pressure to build which may not be allowing the pressure to build-up quickly enough.  Another thing is that the needle I am using has a lance on both ends.  If I am able to somehow grind the lance off the "business" end to get a flat end, maybe it will do better.  I'm still thinking of a way to make a proper pintle to work in conjunction with the needle orifice.

Tomorrow's another day.  (It -better- be!).    

6 March 2018:
It's been a long "tomorrow".  I took a month to generally catch-up on some projects and get The Algore Edition Green Hybrid Hoyt-Clagwell (which means it's nearly worthless) ready for Zolfo Springs.  Just got back.  Had a good time there and am now ready to do something on this engine, even if it's not groping around for a decent injector idea.

What I'm going to do is finish the engine CAD with some changes to both simplify it and make it a bit cheaper to build.  The water jacket is going to be smaller and I am going to use simple spur gears for the timing instead of the spiral and miter gears.

I figure that, after I get the engine built, I will have some idea of a working injector.  I have been thinking of moving the injector to the side of the head and making a sort of pre-combustion chamber for the injector to squirt against to break up the stream into a sort of spray.  If I can go direct injection, I will do it, because the pre-combustion chamber idea smacks a lot of Bronz or early Cummins with timed fueling instead of untimed.

I'll post a drawing once I've gotten to where it looks like something.

Boy!  This is fun!

In case you see me about to do something really stupid and want to warn me, here's my email address
[email protected]
You are visitor number

Since 23 August 2017