Fuel Injection Systems Part 2

[Roger B]

I continued testing the injectors. The second injector (1) has an 8° angle the same as the first (0):

<a href="https://www.youtube.com/watch?v=eiltz3fsyAE" target="_blank">http://www.youtube.com/watch?v=eiltz3fsyAE</a>

The third one (2) has a 12° cone and did not give a good spray pattern.

<a href="https://www.youtube.com/watch?v=fGT4Zcx7Nkw" target="_blank">http://www.youtube.com/watch?v=fGT4Zcx7Nkw</a>

The forth one (3) has a 10° cone and also did not work well.

<a href="https://www.youtube.com/watch?v=eQi4ddWMdSg" target="_blank">http://www.youtube.com/watch?v=eQi4ddWMdSg</a>

The needle injector gave a fine spray although it was somewhat offset.

Finally I did a flame test with the needle injector.

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

And with the best mushroom injector (0)

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

The 8° angle mushroom injectors seem to give the best spray pattern and 'buzz'. The needle injector seems to give much finer droplets. All the more to think about 

[Vixen]

Hello Roger,

I always enjoy your gas lighter experiments, the results are always spectacular and seem to be getting better with each design iteration.

Are your both of your eyebrows still intact? 

Cheers   

Mike

[Roger B]

Thank you Mike  I hope I am learning useful things with these trials however I am nowhere near the conditions inside the engine.

My eyebrows and hair are unscathed, I do have a damp towel on hand as the first line of defence and a dry powder extinguisher by the door as a backup.

The next trials are with the two injectors fitted to the engine and cranking at around 500 rpm with the battery drill. The fuel rack is set to where the engine would previously run. There is a significant difference in the combustion characteristics with a much lower fuel volume than from the test pump. The needle injector gives an almost explosive burn whereas the mushroom injector is a sluggish and smoky flame. This may be where Find Hansen's small injector shows it's benefits     

<a href="https://www.youtube.com/watch?v=T70SwvtFZwQ" target="_blank">http://www.youtube.com/watch?v=T70SwvtFZwQ</a>

<a href="https://www.youtube.com/watch?v=cuNx4eX4GrI" target="_blank">http://www.youtube.com/watch?v=cuNx4eX4GrI</a>

[Roger B]

I received a new M1 die from Volkel which was a different design and cut the 8mm long thread in one go without problems. The holes in the new die are much smaller than in the original which I think will give more rake angle    The nuts were made from the same 2mm brass hex, drilling 0.8mm and tapping with some old Swiss M1 taps I got in a work clear out, which cut with just finger grip on the shaft. The threads fitted together so I made a 1mm guide bush for the tailstock drill chuck to support the needle whilst I turned the shank down to 1mm diameter and then 0.8mm at the cone end.

[Roger B]

The top slide was then turned 5° to cut the 10° cone and finally the needle was parted off to give a diameter at the end of the cone of 1.2mm.

Next make a couple of nozzle bodies for the 1mm and 1.2mm designs and harden and polish the needle cones.

[Vixen]

Hello Roger,

Turning that needle taper must demand the cutter is spot on centre height or the taper will become waisted. Do you cut from the small diameter outward or the other way round. How do you deal with needle deflection with these small diameters?
I am full of admiration as you are working to sizes I can bearlly see,these days

Mike

[Admiral_dk]

I'm curious about the same as Mike !

+ I agree - I got a good number of Volkel Taps and they are Light-Years better than anything els I have tried on Metal 

That said - they are Bollocks on Plastic - Not surprising as most Plastics should be cut with Completely different Geometries of the Cutters 

Will say that this latest design looks really good in the pictures Roger 

Per             

[Roger B]

Thank you both   

The machining sequence is as follows:

A length of 1.5mm silver steel is faced and centred with a 0.5mm centre drill close to the collet chuck. I use a DCTG tip set to within 0.1mm of centre height.

The stock is then bought out about 10mm and supported with the tailstock centre. It is then turned down to 1mm diameter in 0.05mm stages and threaded M1 for about 8mm.

The tailstock centre is then changed for the drill chuck which holds a short length of 3mm brass rod with a 1mm hole drilled through the centre. The stock is bought out another 12mm and the thread is supported by the brass bush whilst the next section is turned down to 1mm diameter (actually around 0.95mm so it will pass through the 1mm nozzle) again in 0.05mm steps.

The stock is then advanced another 5mm and is turned down to 0.8mm with the previous 1mm section being supported by the brass bush.

The stock is advanced again 5mm and the top slide is set over 5° to cut the taper. The free end cannot be supported but the full diameter stock is held in the collet chuck and the cut is from small to large so the forces are taken by the chuck. The cone is turned until the tool just contacts the 0.8mm section.

The needle is then parted off using a 0.5mm Aplitec parting/grooving tool.

[Vixen]

Hello Roger,

Thanks for the very detailed response. Wow!! so many sequential steps, so not exactly mass production. It makes a lot of sense to turn the taper as the last opp, close up to the collet, which is holding the full stock diameter.   0.5mm centre drills and 0.5mm parting/ grooving tools are in a completely different world to most model engineering.

Sounds like a case of 'When in Switzerland, do as the Swiss (watchmakers) do'.

More impressed with your injector experiments than ever.

Cheers   

Mike

[Roger B]

Thank you Mike     there have been a lot of ‘trial’ pieces made whilst learning/developing the various techniques. Even reaming a nominal 1.5mm hole on size and without bell mouthing took around 10 experiments with drill sizes and feeds.

On with the making.

The two blanks were parted off and then held in a 5mm collet to turn the feed end. The other nozzle was inserted into the back of the collet to balance it. The 1mm nozzle is slightly shorter to allow for the length (or lack of) of the small drills.

I then moved to the FD150 for drilling and reaming, higher speeds and more feel. The 2.2mm for the spring pocket and the 1.4mm clearance for the 1mm needle went ok. I then chose a Ruko 0.8 mm drill as it was longer than my Proxxon ones. Unfortunately it broke on contact but luckily the broken pieces fell out and I was able to continue with a Proxxon 0.8mm drill.

The nozzle was then reamed 1mm (actually 0.99mm as it was on clearance offer) and the needle was checked for size. The cone is currently a bit too long, but I will see how it looks after hardening, polishing and lapping.

[steamer]

Those are coming along really nice Roger.   Creating the pressure is a HUGE challenge but you seem to have conquered that!

[Kim]

Fascinating work, as always, Roger! 

I love seeing all your experiments and how you do your work.  You're quite methodical about it all.  Very impressive!

kIM