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

[Lloyd-ss]

Hello everyone.
...............................................
This thread actually started a few days ago on the general engine discussion forum, but it seemed like it really belonged on this forum, so here it is.
https://www.homemodelenginemachinis...smatched-rings-and-piston-and-cyl-bore.31104/
................................................

I have wanted to build a diesel for a long time, actually a 2 stroke complete with Roots blower. Why not bite off more than I can chew? I call it being greedy; and it usually gets me in trouble. EDIT:I have since revised my expectations and have dropped the blower and 2 stroke concept.
About a year ago, I found Find Hansen's beautiful engines on Youtube, and that almost pushed me into the project. Then I ran across his new video of the build of an A-Frame diesel. Darn! Then I found this great forum with lots of smart and helpful and friendly members. That's a rare combination to find on the internet! So I dove in head first.

The selection of 33mm bore x 66mm bore had an odd beginning. In one of my junk piles I had some short pieces of 8" welded steel pipe (actually 8-5/8" O.D.). I thought, that could make a great flywheel, what size cylinder would it go with?

So the first item is the one that for me, took the least amount of decision making. Like using 4 spokes or 5 spokes? The 5 curved spokes spoke to me. It is bolted construction, with inner and outer rim, 5 spokes and a hub, and hidden bolts.

I started by slicing off 3 pieces of the 8-5/8" O.D. x .30 wall welded steel pipe: rim, inner rim, spokes. I used a Harbor Freight $120 portable band/hack saw and had to saw all the way the way around each ring. I sawed them all in one session, using plenty of oil on the cuts. The Lennox blade held up fine. The saw didn't whimper.

My lathe is a Grizzly 10x22 and that pipe definitely pushed the limit, but I got the outer rim roughed out, clamping from the inside out with the 6-1/4" 4 jaw. Indicating the rings in to get them to run optimally for least amount of material removal was a chore. The outer ring was the first part rough machined. To make the inner spoke ring, I sawed out a 2" section of one of the pipe slices, and clamped it to force it into an 8" O.D. Then I welded it, multiple passes, grinding out the flux and inclusions between passes, One thing that really helped the welding was preheating the joint area with a Mapp gas torch. Welding that 5/16" wall just wouldn't have worked, not a my skill level anyway. Then I rough machined the new not trying to get 100% cleanup.
I machined the last ring to the finished cross section of the spokes.
The hub was rough turned from 12L14. It has an O.D. groove the width of the spokes, and has 5 flats machined into that groove.
I rough cut the 5 spokes to length plus about 1/8". It is bolted construction, with each spoke attached to the rim with a single 8-32 bolt, with the head counterbored into the rim. After all 5 spokes were attached to the rim, I fit them to the hub and determined how to trim the ends of the spokes and install the bolts.

After it was all lightly bolted together I chucked it in the lathe and pushed it around to get minimal runout and wobble, which ended up being about 30 thou. I tightened the bolts up, using a generous amount of 680 loctite. I then trued it up ont he lathe, changing the chucking at least 4 times.

Then the outer rim was attached using (5) 10-32 set screws that fit into pockets on the I.D. of the outer rim. After it had a final alignment, one last set of clean-up cuts were needed to get the non-painted surfaces all pretty looking. Lastly was several sessions of epoxy and JB Water Weld. I was shocked at how well that combination worked. i cleaned with acetone, and lightly preheated before applying epoxy, and used a heat gun to get the epoxy to flow properly. If this wheel were dropped hard enough to break it, I think a cast iron one would have broken too.

Here are the final as-built specs for the flywheel.
Outer rim O.D. 8.58"
Outer Rim width 1.99"
Inner rim 1.13 wide x . 25 thick
Spokes (5) at .88" x .20" cross section
Hub 2.30 long x 1.82 O.D.
Hub Bore .75"
Weight 7.96 pounds

........................................
Starting with a chunk of 8" nominal welded steel pipe.

Three slices of pipe.

L to R, Outer rim, spokes, inner rim.

The bolted assembly with the epoxy and bolts visible.
The JB water weld has NOT been applied yet.
EDIT: Each spoke is machined for a snug fit into its loacation, including a snug fit into the matching groove in the hub. There is one 8-32 bolt radially thru the inner rim into the spoke. There is also a 10-32 bolt that goes thru the spoke radially into the hub.When these 10 screws were all tightened and secured with Loctite 680, the flywheel became a rigid unit. The outer rim was a slip fit onto the inner rim, and epoxy was was used to secure it in place (and possibly kill some vibration) and was also secured with setscrews radially outward from the inner rim into pockets in the outer rim. This method allowed the outer rim to be aligned with the hub to minimized radial and axial runout. The additional epozy and JB water weld fillets are mainly for cosmetic purposes, but I am sure they also will contribute to strength and vibration damping.

After filling, sanding, priming and painting, here is the Glamour photo
of the finished flywheel, using special lighting,
and the photo editor, for those extra special high-lites, LOL.

More to come, but not really fast.
Lloyd

 

[minh-thanh]

Nice flywheel ! I really like a diesel, hope you will succeed .

 

[Lloyd-ss]

Thank you minh-thanh. I have been cautioned about the difficulty of injected diesel, but if it doesn't run, it wont be because i didn't give it my best shot.
Lloyd

 

[Johno1958]

Good luck with it Lloyd.

 

[Lloyd-ss]

Good luck with it Lloyd.

Thanks Johno, I am sure I'll need a fair helping of luck along the way.
I am working on the injector prototype right now. Started the design with a pintle that opens outward, but am now seriously considering making one that opens inward. They seem inherently safer (I think). With the open outward, if the tapered head broke off into the combustion chamber there would be a serious disaster . With the open inward style of pintle, at least there is no head to break off.

 

[Lloyd-ss]

I made a few of the parts for the injector. The plan is to get it built and tested before making any of the other parts for the engine.

 

[George Punter]

Lloyd I was interested to read this thread and see your photographs of the progress on your Diesel engine. About six years ago I followed the same path and for the same reasons. I had drawings for the 6/1 Lister Diesel engine and used these to make a scale model. The engine is 20cc and uses direct infection into a pre combustion chamber. I used a modified gass jet as the injection nozzle and the injection pump had a bore of 4mm and a stroke of 1mm. A manual bypass was used to pass the excess fuel. The injector fuel line had check valves to prevent fuel coming back down the line. I am very much a " a try it and see if it works" type of model engineer and don't get bogged down with too much theory, but there was some discussion relating to it not being a "real" diesel. In my book if there was no spark plug to ignite the mixture so it was a Diesel engine. I have included some photographs of this engine and should you want to see it working it is on U tube - just type in model lister Diesel engine. Good luck with your project and I will continue to follow your progress.
George.

 

[Lloyd-ss]

George, nice engine and video, too. Funny about the discussion on whether or not its a "real" diesel. Looks like one to me, too. Gotta keep a sense of humor and have fun, and in the video it looked like you were having fun.
Whenever I watch one of these model engine videos, especially the diesels, I realize how many questions there are that I didn't even know I was supposed to have. Wow! I usually rough out an entire design or concept, and then detail the drawings as I go, but most importantly, I try and be really good about cleaning up my cadd design with as-built drawings as I make each individual part. This engine is based very loosely on what one of Find Hansens horizontals look like, but i am doing my own drawings and design, and when I am done, it will be quite different.
On the unanswered questions, I am wondering about the 33x66mm (56cc) size. Maybe that's too big, IDK.
Cooling. It will have a water jacket, but I watched a video the other day (dieselpilot directed me to it) where the builder had a little bitty radiator with his engine. Hmmmm.
Lube systems. Seems like bronze bushings and bearings with drip lubrication are popular. What about needle bearings or something that doesn't need to be oiled? BTW, I am not trying to keep this authentic to any period or design, I just want it to work and run, and look kinda cool, too.
Glow plugs. Anybody ever use a glow plug for starting? Real diesels do. The airplane engine glow plugs continue to stay hot after the electricity is cut off due to the reaction with the nitromethane fuel, but I don't know if they would do that running diesel. I would only want it for starting, definitely not for running.

A real can of worms, but already a lot of fun, and I definitely like meeting all of the Forum members. That doubles the fun and probably cuts the mistakes in half, LOL.
Lloyd

 

[Lloyd-ss]

I finished making all of the parts for the injector, and next will be the final assembly of it, and then testing it for functionality.
I hope it works, LOL.

Here's how its "supposed" to operate.
There will be constant low pressure from a pump on the engine to keep approx 50psi (TBD) of pressure on the fuel in the injector. There will be a check ball in the fuel line near the injector to protect the pump during actual injection. The injector has a .062 dia piston at the top, that will be operated by a rocker arm that adjustable via a governor to control the stroke length. Stroke length could be a long as .115". The .062 dia keeps the force to operate the piston low ( less than 5 pounds force) but still be able to produce over 1,000 psi. The piston is hydraulicly coupled to the pintle; the pintle does not open by mechanical force, just by the pulse in hydraulic pressure. (I am hoping there isn't too much "give" in the hydraulic system and that the pintle does indeed open crisply. The return spring on the pintle holds it closed until the opening pressure (TBD) is exceeded by the hydraulic pressure from the rocker driven 1/16" dia piston. The travel of the pintle is limited by the two locknuts to about .005" (TBD) so that the pintle doesn't just blow open and dump all the fuel at once. The plan is for the fuel to atomize better thru the barely-open pintle.
There are 2 high pressure ferrule compression fittings on the valve body. One is the fuel supply, and the other is to attach a pressure gauge during testing. I will probably have to do something to protect the test gauge if I run the injector at speed. After testing, the extra port will be sealed off with a 10-32 seal screw.
One thing I like about this design is that there is only one dynamic O-ring seal in the injector. All the other o-rings are static seals. Of course, there is always the pintle which probably won't seal the first time. Sigh.

That's the plan, anyway. You can learn a lot from failed tests. right?
Lloyd.

 

[J.Bueno]

Hi. Great job. This is my first participation in HMEN. I'm building a diesel engine too. I'm limited in tools. I only have lathe.
I have already obtained good results in the high pressure pump and the injector through a method similar to this one.
Initially, I started using the oring for sealing the connections but gave undesirable elasticity in the hydraulic system, then replace them with paper joints.
Glad to see your project.

I'll be following up.

 

[Lloyd-ss]

J. Bueno, thanks for your comment. Good luck on your project too.
Lloyd

 

[werowance]

Curious about the part that looks like a spring. is it a spring? if so what material did you make it from? it appears to have been wound with some sort of flat stock or shim stock.

 

[Lloyd-ss]

Curious about the part that looks like a spring. is it a spring? if so what material did you make it from? it appears to have been wound with some sort of flat stock or shim stock.

It is indeed a spring, home-made. I needed a very stiff, but compact spring, to hold the pintle closed, similar to what you might find in a regulator bonnet. There was nothing suitable in my collection, but I did have some straight lengths of .041 dia stainless steel spring wire. I clamped one end of a 5/64 allen wrench (to serve as a mandrel), together with one end of the 12" piece of spring wire. Then I just manually wrapped the wire around the mandrel, keeping the spacing as nice as I could (ha, ha, that's a joke!) and wound it about 4 times as long as what I needed. I snipped out the best section of the spring and then, with a bench grinder and a tuna can of water , squared up the ends. I think that using an allen wrench as a mandrel helps to keep the spring wire from slipping and therefore reduces spring-back after you remove it from the mandrel. But the spring I.D. did open up to about .100-.110, just what I needed, but the O.D. was too big to fit in the hole. So I slipped the spring onto a barely-loose-fit piece of stiff wire and went back to the bench grinder. I held the spring against the side face of the grinding wheel and let it spin moderately on the wire mandrel, using my finger to control the speed. Plenty of dips into the water and it ended up with a wire cross section somewhere between round and "D" shaped. It is now about .010 in dia smaller than the hole, to allow for dia growth during compression. Using an online spring calculator the spring has a rate of about 300 pounds per inch and a coil-bind force of about 60 pounds. You can't readily buy anything like that . I'll be using it at a force of 10 to 15 pounds or so.

Springs are fairly easy to make, but it is difficult to make one that looks like it was made in a factory, ha, ha. There are all kinds of gizmos that you can make or buy to do a better job, but not usually worth it.
Lloyd

 

[Willyb]

I finished making all of the parts for the injector, and next will be the final assembly of it, and then testing it for functionality.
I hope it works, LOL.

Here's how its "supposed" to operate.
There will be constant low pressure from a pump on the engine to keep approx 50psi (TBD) of pressure on the fuel in the injector. There will be a check ball in the fuel line near the injector to protect the pump during actual injection. The injector has a .062 dia piston at the top, that will be operated by a rocker arm that adjustable via a governor to control the stroke length. Stroke length could be a long as .115". The .062 dia keeps the force to operate the piston low ( less than 5 pounds force) but still be able to produce over 1,000 psi. The piston is hydraulicly coupled to the pintle; the pintle does not open by mechanical force, just by the pulse in hydraulic pressure. (I am hoping there isn't too much "give" in the hydraulic system and that the pintle does indeed open crisply. The return spring on the pintle holds it closed until the opening pressure (TBD) is exceeded by the hydraulic pressure from the rocker driven 1/16" dia piston. The travel of the pintle is limited by the two locknuts to about .005" (TBD) so that the pintle doesn't just blow open and dump all the fuel at once. The plan is for the fuel to atomize better thru the barely-open pintle.
There are 2 high pressure ferrule compression fittings on the valve body. One is the fuel supply, and the other is to attach a pressure gauge during testing. I will probably have to do something to protect the test gauge if I run the injector at speed. After testing, the extra port will be sealed off with a 10-32 seal screw.
One thing I like about this design is that there is only one dynamic O-ring seal in the injector. All the other o-rings are static seals. Of course, there is always the pintle which probably won't seal the first time. Sigh.

That's the plan, anyway. You can learn a lot from failed tests. right?
Lloyd.
View attachment 108191

Good day Lloyd-ss
I would't install a pressure gauge in your injection system unless you are absolutely sure it is full of fuel and has zero air in it. Most pressure gauges are made with a curved tube with one end connected to the pressure source and the other plugged. This makes it very hard to get the air out of the dead ended tube. If you have any air in your injection system, it's not going to work. The miniscule amounts of fuel being injected in these small diesels can easily be offset by a bubble of air in the injection system. When injection takes place the air compresses instead of the fuel being injected.
Keep up the good work.

Cheers Willy.

 

[Lloyd-ss]

Thanks Willy. Yes, I was thinking about the purging process. The gauges would only be used during some prelim testing, and it might or might not work, but I'll give it a try. I was planning on doing the testing with a water soluble coolant/lube, mixed fairly rich. Should be safer that way but I don't know how close the performance would be to diesel fuel. We'll just see how it goes.
Lloyd

 

[weir-smith]

Lloyd

Just noted your diesel build and its good to see someone else giving it a go as there are very few around. Approximately two years ago, I commenced to build a true diesel of no particular type. It was made from material that I had on hand and to my own design based on a horizontal type engine with a 40 mm bore and a stroke of 80 mm using an auxiliary shaft to drive the fuel pump and valve cams. The valves were mounted on the side of the cylinder head e.g. top and bottom. The purpose of the build was to see if I could make it work using diesel as the fuel. I have learn much during the build - mostly what not to do however, I have learnt a lot about diesel engines.

Initially, all the components were made and the first test to start the engine was a complete failure. The problem was where to start to work out what was wrong. The first thing was the compression, as the design required 500 psi or better if it was to work. I removed the head and replaced it with a temporary head with an inlet valve and the exhaust valve which had a pressure gauge fitted. The valves were inline types that were good for 2000 psi. After cranking the engine, the best I could do was just over 200 psi. The piston was remade to allow for four more rings and retested and I now had better than 500 psi. Next was to test the valves and both failed at approx. 300 psi, so more grinding and heavier springs. I now had the basis of an engine that should work however, no such luck.

I put a pressure gauge on the fuel pump and the best I could do was about 600 psi. Eight more attempts were made to build a pump and the best I could do was just over 1200 psi. Putting that to one side as I was fed up with it, I tested the injector using an injector tester. The results were not very good or consistent so I tried several designs and they all failed in some way. A friend gave me a pintal injector nozzle and out of desperation I used this as the basis of a new injector. The results were very good in that at 2000 psi it produced a atomised cloud of fuel. At lower pressures, it produced a very fine cone spray however it had a tendency to drip. Probably why it was replaced in the first instance. A new one was ordered but it is not quite the same and I will need to modify the injector body.

I still had the problem of the fuel pump. I was given a CAV pump which I dismantled to see how it was made and while doing so, someone else advised that there were cheap small injector pumps available on Ebay. So for $25 one was ordered and while is was well out of scale, I could see how I could machine parts down to bring it closer to what I needed. However, before doing so I wanted to test it as is. On the bench, it easily produced 3500 psi which was the limit of my pressure gauge. The pump was fitted however at the required pressure 2000 psi, the auxiliary shaft was deflecting so a new mounting bracket for the injector pump cam was made. I was now in the position to give it a try and on the first test I had large clouds of white smoke which indicated unburnt fuel. The problem was in the timing so I made a disk with degrees marked and fitted it to the aux. shaft which rotates at half speed. The cams were reset and this was followed by another test and at last the engine fired but would not sustain continuous running. My temporary fuel control was very hit and miss so to speak, so that is the next thing to do to allow better control.

This account has rambled on some what however, it does show that with persistence you can achieve an outcome. I have included a photo of the pinal nozzle and I am happy to answer any question you have. The best advice I can give is to suggest that you test the critical parts at the time to ensure that they can produce what you are expecting. I also agree that you can't put a pressure gauge in the line between the pump and the injector because of the air problem. The amount of fuel is very small and it rises quickly to the "pop" pressure and then immediately drops to almost zero until the next cycle. Air in the system no mater how small will prevent the system from working. Also, I have used copper tube for the delivery to the injector and to date it has not failed however, I have purchased some 3mm stainless capillary tube to use.

My current build is a vertical air blast injection engine which has its own unique set of problems and a further engine at the startup stage is a Fairbanks Morse 3 cylinder R80 engine.

Bruce

Note, the coin is approximately 16 mm

 

[Lloyd-ss]

Hi Bruce, I want to tell you how much I really appreciate your taking the time to detail the trials and tribulations that you went thru on your engine. And how about an award for persistence? But honestly, this is exactly the type of help I was hoping to find on HMEM. The areas that vexed you will probably vex me too, so the caution flags are up.

Your advice about testing each critical component as it is built is well taken. That's why I built and tested the flywheel first, LOL. When mounted on a shaft with bearings, it maintains its momentum. That is at least one item in the win column.

The pintel and seat in your photo is a commercially available part, correct? If I get stuck, the source info on that would be helpful. I have already had to re-design the internals of the injector. The items in the earlier photo functioned fine, but did not seal very well. A couple of cc's of 1000 psi air would bleed thru in about one minute, all the way down to zero. The new design is with a pintel that opens inward, like the one you have, responding to a pressure imbalance. It is interesting that I thought it was a long shot that the first design would seal properly, but I built it anyway. Lesson learned.

The second injector should be finished today and ready for prelim testing. I do indeed expect better results with this one.
Lloyd

 

[weir-smith]

Lloyd

Just a quick note. I think you would be better off using diesel fuel to test your injector. You will need to make an injector test pump with a pressure gauge e.g. 2000 psi or better and include a non return valve to allow the pressure to be built up. I made a test pump from an old 3T hydraulic jack. Just needed to do some cutting with the band saw followed by some welding. A bit rough however it worked and went back to the bin when I was finished. You may be able to put your hand to the real deal as there are plenty of injector test units about as they have passed their use by date with electronic injection control.
I have included the info re the pineal nozzle and note, we now pay GST on our purchases hence the higher price.

Just one more thought re fuel pumps. I collect old injectors for bits and pieces and I was recently given a reasonably new one. The valve at the injector end is about 4 mm (approx) and fitted into a a bore that was of very close tolerance which is what you would expect. The material is very hard e.g. better than 4140 however, I think it could be ground down which I will try on my tool and cutter grinder. I need a high pressure low volume fuel pump for my current project. Using such materials makes life much easier as to machine and grind to these exact tolerances is very difficult at best.

Bruce

e.

 

[Lloyd-ss]

Bruce,
Thanks for the additional information. I might be able to obtain that Kubota injector locally.

I ended up changing the internals of my prototype injector so that the injector pintle opens inward in a more conventional manner as the pressure goes high enough to pop the injector open. But it still leaks at the pintle tip seal so that requires some attention.

Here is the assembled injector. The injector is also the high pressure pump.
The plan is still for the injector to be fed by a constant low pressure pump at about 50 to 100 psi, driven by an eccentric on the shaft that drives the cam shaft. There will be a check ball inside the inlet adapter. Opposite to the inlet fitting is a screw for bleeding and test hook-ups. The small pin that sticks out of the top of the injector is actually the top of the pump piston, and will be driven by a rocker. The end of the piston that is inside of the injector is .062" in dia, so very small. Also, inside the injector there is very little fuel volume, maybe 0.3cc, so once the system is bled of all air, and the check ball is working properly, building high pressure and popping the injector open with each piston stroke should be possible.
For scale purposes, the input line to the injector is 1/8" O.D. stainless steel, and the thread at the tip of the injector is 1/4-28.

Just for show-and-tell, this is a hydraulic burst test fixture that I made to test high pressure air tubes. Capability is to approx 30,000 psi. The black tube on the right is 1.25 O.D. x .095 wall 2024T3. Out of the picture on the left is a Harbor Freight hydraulic hand pump that goes to 8,000 psi, and serves as the first stage of the hyd system. The 2 long steel tubes to the left of the black tube are a hyd multiplier with a large input piston on the far left, that drives a smaller piston that is in the tube on the right. Multiplication is about 4 to 1.
When testing the black aluminum tube on the right, the first failure occurred at 9,000 psi when the tube swelled and an o-ring popped. I fixed that so that the o-ring held pressure again, and then the retaining screws for a valve plug in the end of the tube plowed their way thru the aluminum tube. The failure was quick and definitive, but the tube never did really "burst". I did test some 7/8 x .065 A513 Ty5 ERW DOM steel tube and it did actually burst at 16,000 psi, which was several percentage points above the predicted failure. The test was reassuring.
Lloyd

 

[weir-smith]

Lloyd

Firstly, an interesting setup for pressure testing. I must admit that I haven't seen something like that before. As for your injector, again I haven't come across a combined injector and pressure pump. However, it doesn't matter what the configuration of pump and injector you have, it all comes down to the volume of fuel (about the equivalent of a single drop of water or less) and it needs to be atomised or very close to it. I would suggest that you need to build a jig or similar to mount the injector/pump with an appropriate lever to operate the piston plunger. You can then see what happens or not. The problem with making small high pressure pumps is fitting the piston to the cylinder with (in reason), zero clearance. While you can't compress a liquid, because the volume in this case is so small, it fills the smallest of clearances and as a result you can't achieve the required pressure. I don't know how they achieve this commercially but they do and they work very well and hence I collect old injectors for the parts. As I said before, use diesel for the test and put some paper towel under the injector nozzle and have a critical look at the spray patten. In your setup, you will not be able to use a pressure gauge to check the pop pressure because of the air in the gauge so it will be down to looking at the spray patten.

I will be interested to see how it goes and also, your machining looks good.

Bruce