Fuel injection systems

[Roger B]

Thank you Andy   

A lot of it is to do with thinking about the scale you are working in. Most 'industrial' lathes have a better intrinsic accuracy than the cheap (GBP 700???) little Proxxon FD150  I am using. Could I make these parts on a Hardinge or Schaublin , probably not  In the first case you can't get close to see what is going on and in the second case the slides are so heavy you have no feel for the cut.  If I wanted to turn a piece of 20mm silver steel 300m long down to 14mm without a travelling steady I would probably use a speed between 250 and 500 rpm and a depth of cut between 0.5mm and 1mm. If I reduce this to 1/10th (what I am doing here, 2.0mm down to 1.4 mm) I will use a speed of 2000rpm or more and a depth of cut 0.05 to 0.1mm. The cutting tool also has to be sharp so I use aluminium finishing inserts.

The tip of the needle was hardened, ground and polished ready to be lapped into the body.

[Chipmaster]

Thanks for that Roger.

Andy

[Roger B]

The needle was lapped with some 1 micron diamond paste and then cleaned in the ultrasonic bath along with the parts of the injection pump for my two cylinder engine. Next it was assembled and tested with cooking alcohol. It has the characteristic creak and a spot about 30mm diameter at 100m, so a spray angle of around 16°     

<a href="https://www.youtube.com/watch?v=HGLnoBil0tQ" target="_blank">http://www.youtube.com/watch?v=HGLnoBil0tQ</a>

I will make a 20° version next to see if I can repeat the results.

[Roger B]

The 20° needle was started in the same way. I tried to take a video clip of turning the needle down to size:

<a href="https://www.youtube.com/watch?v=OTLVDq7Za4Y" target="_blank">http://www.youtube.com/watch?v=OTLVDq7Za4Y</a>

Unfortunately this nozzle wouldn't seal even after significant lapping  This may be due to the effect on the body of the previous floating needle  It may be better to start again from scratch, it doesn't take long now.

If Andy is still following, another critical point is getting the tool on centre. If you are turning a 50mm bar and the tool is 1mm low it will still cut reasonably well. If you are turning a 2mm piece and the tool is 1mm low it will miss completely 

[Chipmaster]

I'm still following Roger, 

Andy

[Zephyrin]

always interesting to follow, to turn and to mill at such a small size...
I wonder how far "conventional" machining with machine tools can go down...what is the smaller limit?

[Admiral_dk]

I'm in the middle of taking my MC apart. It just turned 17 and 100000Km. so in need of a bit of TLC ....
This evening I got the throttle bodies of the engine and could see the four tiny holes in each injector and this made me remember that they are LASER drilled at the factory .... and already oldschool - the new ones are up to twelve much smaller holes in them 

I'm pretty sure I will continue to follow this thread until completion .... and I really wish that you succeed Roger - simply because it would be nice and the rest of us will be enlightened along the way 

[Roger B]

Thank you for your interest and support 

Zyphyrin, I think that Dave (Steamer) did some very fine milling with a conventional machine at work using a barely visible cutter but I can't find the thread at the moment.

The trials with the diesel and the two cylinder engine have highlighted several problems that I will try and work through on here.

- The first one is the inlet port on the separate element pump. If the cone on the tapping drill goes to deep it will destroy the O ring seal. I have ordered some 4.5mm endmills to try rather than a drill which should reduce that risk.

- The delivery valve seating builds up some scale when hardening the pump element. I lap out the bore and will have to try the same with the seat. A piece of 4mm brass rod should work.

- The dynamic forces on the injection pump plunger are quite high. The ball race cam followers on the diesel broke, the bronze follower has spread, I think solid hardened steel is the next phase. The stroke adjuster on the two cylinder engine is also affected.

Lots to do and think about 

[Roger B]

The lap was crosscut with two offset small slots to hold the 5 micron lapping compound and faced off. Several boring evenings were spent lapping the delivery valve faces for the diesel and the two cylinder petrol engine. Both look ok but I need something different to get good photographs  I was surprised  how much the lap had worn  maybe I should have refaced it more often 

Next I will make a new body for the two cylinder engine pump and clean them all in the ultrasonic bath.

[Admiral_dk]

Re: Lap wear - as far as I remember it's the harder of the two materials that will be worn when you use a lap ... Why - the grinding particles imbed themselves into the softer of the two materials (if it can), and therefor cuts the hard material ....
So I guess that if both materials are hard it becomes unpredictable where the cutting happens ....

[steamer]

Thank you for your interest and support 

Zyphyrin, I think that Dave (Steamer) did some very fine milling with a conventional machine at work using a barely visible cutter but I can't find the thread at the moment.

The trials with the diesel and the two cylinder engine have highlighted several problems that I will try and work through on here.

- The first one is the inlet port on the separate element pump. If the cone on the tapping drill goes to deep it will destroy the O ring seal. I have ordered some 4.5mm endmills to try rather than a drill which should reduce that risk.

- The delivery valve seating builds up some scale when hardening the pump element. I lap out the bore and will have to try the same with the seat. A piece of 4mm brass rod should work.

- The dynamic forces on the injection pump plunger are quite high. The ball race cam followers on the diesel broke, the bronze follower has spread, I think solid hardened steel is the next phase. The stroke adjuster on the two cylinder engine is also affected.

Lots to do and think about 

It's around somewhere....I'll dig it up if anyone's interested.....Lapping and grinding can be non-deterministic..meaning that both the part and the tool change size as they interact.....the good news is it's pretty repeatable...

Been watching this thread....glad to see you sticking with it Roger....that's no easy feat!.....Detroit Injectors have a total tolerance of 33 millionths on diameter...for the plunger anyway....

Dave

[steamer]

.001 end mill.....

https://www.modelenginemaker.com/index.php/topic,8860.0.html

[Roger B]

Thank you both 

In theory the hardened steel body should have worn faster that the brass lap but it there was only motion in one plane it wasn't really true lapping. The real test is if they seat better when put back together.

As the weather here is currently hot and sticky (supposed to be a big storm tonight) I cut the blank for the body with the bandsaw rather than by hand. It is made in much the same way as before with a few more dimension checks. The fixing screws were moved 0.5mm so that they didn't break into the bore. The inlet valve port was finished with a 4.5mm end mill rather than a drill to avoid the cone breaking into the bore for the element as happened last time stopping it sealing properly.

[Roger B]

So the slot for the stroke adjusting wedge was cut with a 4mm (throwaway) endmill. The improvement in the positions of the various holes is quite obvious and after a good swim in the ultrasonic bath the petrol injection pump went together with very good alignment. Hopefully that means it will all seal under the 0.8-1.0 Bar from the feed pump (If the petrol feed pressure is too low it vaporises on the suction stroke).

[Roger B]

The next step is to try a 10° nozzle for the diesel. This was made much as before with a quick reflection over various trial needles and the evolution of injectors from my first experiments.

There is also a drawing of my current design and a picture of the Oldsmobile injector this design is based around.