Diesel engine : Kromhout Gardner 4LW 1940

[minh-thanh]

Hi All !
An update :
A little for today
I tested the oil pump, everything went according to my plan
The oil adjustment bolt for the cylinder head does its job well, the overpressure protection valve is not completely sealed because it's not important , it only works to maintain a pressure in the oil pipe.

 

[minh-thanh]

Hi All !
An update :
Some tests...for tonight
I have to sharpen tweezers so I can get the wedges
and use a magnifying glass to check ...

 

[minh-thanh]

Hi All !
An update :
I have already assembled the pump, but I am not satisfied with the result, the pump function, and the speed control is fine, but the hand pump part doesn't work as expected.I am improving it and also I am looking for a new solution

 

[lee webster]

The oil supply pipe to the main bearings looks as though it is soft soldered. Will the engine oil get hot enough to melt the solder? The reason I ask is that I made a similar oil pipe for my 1930 Austin Seven car. I haven't run the engine yet. I discovered that soft solder and hot oil are a bad idea. So I will have to strip out the pipe and make a new one with silver solder. You are doing an amazing job!

 

[minh-thanh]

Hi Lee !
Thanks !
I soldered it with a silver solder ( Harris 0% silver brazing rod )

 

[minh-thanh]

Hi All !
An update .
There is always a difference between design and reality.
As said in the post , manual fuel pump function is not good , I have 2 options : 1 / Do not do that function , 2 / Redesign , try to make ...
And I redesigned the fuel pump, and with a few samples to test..., the results seem fine
Quite a few parts (in the attached image) have to be removed if I choose option 2
I will do option 2 and hope it will be fine.

 

[minh-thanh]

Hi All !
An update .
After change..., I made a temporary hand pump to test, everything went according to my plan: hand pump is fine

 

[minh-thanh]

Hi All !
An update .
I made the oil pipe for the head cylinder and assembled it to test, It's ok

 

[ajoeiam]

Hi All !
An update .
I made the oil pipe for the head cylinder and assembled it to test, It's ok

snip

Interesting banjo fittings - - - - will have to remember that.

Did you put copper washers between all the surfaces?
(I can't see evidence of such therefore asking.)

 

[minh-thanh]

Hi @ajoeiam !

Interesting banjo fittings - - - - will have to remember that.

Did you put copper washers between all the surfaces?
(I can't see evidence of such therefore asking.)

I thought of copper washers and made them
But today, when making bolts and assembling, I noticed it made the position too long and I didn't use copper washers.
As long as the surface is good enough it will prevent the oil from coming out, and the oil has no pressure - The oil pipe only carries the lubricating oil to the cylinder head
I have assembled the oil pipe and adjust the amount of oil for the cylinder head and tested, everything went according to plan.

 

[minh-thanh]

Hi All !
An update .

After many checks and adjustments...everything was fine so I assembled the fuel pump's cylinder
After assembling, I turned the engine by hand - many times - to check the smoothness of the pump components, then turned the engine with a hand drill : everything went according to plan.

And here is the entire fuel pump - unassembled pump head

 

[GreenTwin]

Its looking good !

.

 

[minh-thanh]

Hi All !
An update .
I'm having a problem with my ball valve block, only 2 ball valves are sealed - perfectly sealed .
My challenge is not the pump cylinder but the ball valve, since making ball valves I always have difficulty sealing the ball valve

 

[xander janssen]

Give the ball a gentle wack with a hammer into the seat to form the seat. Then throw away that ball as it will be damaged and repace by a new ball.

If possible (temperature and chemical properties of the fluid to be transported) use solid Viton balls.

 

[minh-thanh]

Give the ball a gentle wack with a hammer into the seat to form the seat. Then throw away that ball as it will be damaged and repace by a new ball.

I tried .
With the outlet ball valve, it is "relatively" easy to seal
Inlet ball valve: the ball is held in contact with the valve seat by a fairly light spring, which makes sealing difficult.
I'm changing a bit with the ball valve., hope it will be better
For me, making it completely sealed is harder than making the fuel pump cylinder

 

[Roger B]

Ball valves for small volumes are a problem. That’s why I decided to deal with the alternative problems of using an inlet port closed by the pump plunger as on a full size pump. As you say the delivery valve is much easier as it has a much higher pressure differential across it.

Unless you are using a fuel feed pump the inlet valve has to open at a very low pressure, the fuel head, which limits the strength of spring if it is installed horizontal. If the valve is vertical and gravity can keep the ball in place it is easier.

The movement of the ball must also be very limited otherwise all your injection volume is lost as the ball moves back onto its seat. If you compare your anticipated injection volume to the volume of the ball (what size are you using 2mm, 3mm?) you can see how limited the movement must be.

Do you have the design information I posted on ball vales as well as the information on diamond charged ball laps (expensive but work)?

 

[rutzen]

I make clack valves for model steam locomotives and the 'give the ball a whack with a hammer ' isn't nowadays regarded as the way to get a seal. Instead, set a ball in a little holder and put it in the lathe with the valve face in the chuck, apply some oil and apply gentle pressure as it rotates. This forms a much better seat. I have holders made for all the ball sizes I use. You can pretty much guarantee a gas tight seal from this technique.

 

[minh-thanh]

Ball valves for small volumes are a problem. That’s why I decided to deal with the alternative problems of using an inlet port closed by the pump plunger as on a full size pump. As you say the delivery valve is much easier as it has a much higher pressure differential across it.

This engine has a displacement of about 9cc per cylinder, the plunger stroke is really small - about 0.3 to 1 mm so its inlet port is not effective.
Actually, with my 1 cylinder 10 cc diesel engine, I also designed the inlet port, but in my tests it was completely useless. For the inlet port to be really effective I would have to make the cylinder diameter1.5mm

Do you have the design information I posted on ball vales as well as the information on diamond charged ball laps (expensive but work)?

Yes, I read it
I'm trying to figure out why some valves are closed and some are not
Really , my problem is not the fuel pump cylinder but the ball valve

 

[Roger B]

The density of air is around 1.2kg/m3 so 10 cc of air is around 0.012g. Taking a fuel air ratio of 14-1 this would require 0.0008g of diesel fuel. The density of diesel fuel is around 0.8kg /dm3 so at full output around 1mm3 of fuel would be required. This could be achieved with a plunger of 2mm diameter and a stroke of 0.8mm (plus a bit to allow for leakage).

The theoretical stroke agrees with your values if you are using a 2mm pump plunger (looks like it from the pictures).

The volume of fuel injected is much less even that the volume of a 2mm ball.

Do you have anything to limit the movement of the inlet valve balls? If they take too long to return to their seatings it will appear that they are leaking and this will probably be fairly random.

I am using a 1mm inlet port and a 2mm plunger. The plunger is stopped when the inlet port is fully open and I have an additional 1mm plunger travel to allow the port to close. When the port closes the plunger is travelling with some speed which gives a sharp pressure rise and hopefully improves the injection quality.

 

[minh-thanh]

Hi Roger !
What you say is quite true
But the problem is that there are 2 pumps that do a pretty good job of pumping fuel, meaning the problem lies in the other 2 sets of valves.
It is possible that the inlet valve is not completely sealed, or the output valve is not sealed, or both of them are not sealed
Definitely not the pump cylinder , because they are the same , and I did the test : cover the finger at the top of the cylinder and with 1 mm plunger stroke it sucks very strongly -There is pain in the area of the skin being sucked by the vacuum. , and hold that negative suction for more than 3 minutes - That's when I pull my finger out while the suction is still there
I'm lapping valves again