ML Midge Diesel

[steve-de24]

The ML Midge is a 0.8cc compression ignition model aircraft engine that was designed about 20 years ago by Mark Lubbock. The design is fairly simple so fits my requirements as (a) it will be my first i.c. engine and (b) it should be small enough to make on my Cowells table top machine shop.

Can I say up front a big thank you to the experienced guys who have so willingly shared their expertise on this and other websites - I don't think I would have even started on a project like this without others showing how it can be done. It is not going to be a quick build because I don't get that much 'shop time' but I'm hoping that keeping up the build log will prod me into continuous, if slow, progress.

A build log needs pictures so I've just bought a camera but even the 'Quick Start Guide' runs to 14 pages; and there is nothing like the print size on a set of Japanese camera instructions to make you realise your eyesight is 65 years old!

To start I will post a couple of pictures showing my set-up to bore the crankcase for the cylinder. I did this last week but set it up again because I wanted something to take photos of to see if I can successfully post pictures on here. The idea for this set-up is taken from the Midge build log on modelenginenews.org

Of note is the small faceplate bracket I had to make and the boring bar held in a cross-slide mounted block to make it as rigid as possible. For those not familiar with the Cowells lathe the faceplate is only 90mm (3.5 inch) diameter.

[scroc]

Looks like a great start Steve. Have not heard of the Cowells lathe over here but it appears to be a quality machine and the idea of tabletop size sounds good.
Keep up the good work and posts,
scroc

[Hans]

I look forward to this build, Steve. I enjoy following these projects which are confined to machines the size of the Cowells and Sherline.

~Hans

[Admiral_dk]

Yipee - another diesel (CI) engine build 

You are off to a good start - I will follow along 

[b.lindsey]

I am looking forward to it as well Steve. I have a cowells on the way so am very interested to see your set- ups and how you go about things.

Bill

[steve-de24]

Thanks for the comments guys.

Bill, you've a lot more experience than I have so don't follow my set-ups without question on the assumption that I know what I'm doing!

In my last post I forgot to mention that a 'button' is screwed to the bracket to locate the crankcase. The button is the same diameter as the cavity in the back of the crankcase and has a flat machined on it which is set parallel to the faceplate. By measuring the full diameter of the button and its 'diameter' across the flat it is easy to calculate how far the flat is from the button centreline which is also the crankcase centreline. The flat can then be used as a datum to face off the crankcase to the correct 'height' when machining the cylinder bore. Not my idea, I got it from modelenginenews.org.

Steve

[steve-de24]

The only things left to do on the crankcase are to tap the threads that hold on the cylinder and backplate plus a bit of cosmetic skimming of the outer surfaces. I'm leaving this until the cylinder and backplate are finished so I can spot through to get their positioning correct. (At least that's my excuse for putting off tapping 8BA (that's about 2-56) into blind holes!

Onto the cylinder. Looking at the drawing one thing I don't like is the thinness of the cylinder wall where the transfer ports are machined into it. The nominal cylinder wall thickness is only 1mm (39 thou inch) and with a 15 thou deep transfer port that leaves the cylinder wall only 24 thou thick (see drawing). I'm thinking of reducing the cylinder bore to 0.354 inch (9mm) which will increase the minimum wall thickness to 44 thou inch.  Thoughts welcome. 

Steve

[sco]

Nice start Steve - will be following this closely.  Your Cowells looks very clean - is this it's first proper job ;-)

Simon.

[steve-de24]

Hi Simon, good to have you on board. I think I was very lucky getting my Cowells lathe and mill second hand about 6 years ago. From the paperwork that came with them they were new in ~2000 and spent the first 10 years of their lives making parts for a gauge 0 railway, its loco's and rolling stock. They've always been kept in the warm and dry and are in good condition. I haven't used them very much, there always seems to be something else taking priority, but I've decided I need to get off my @rse and make something, hence this build log.
Steve

[steve-de24]

Made a start on the cylinder. I decided to go with a bore smaller than that shown in the drawing because I am worried about the out-of-round stiffness that could result, especially if the tolerances stack up the wrong way.
A piece of 5/8 inch bar was faced off, centre drilled, and then drilled through 4, 6, 8 and 9mm. For the two larger sizes I had to use stub drills because of the lack of space along this tiny lathe bed.
A small boring tool was then used to clean up the bore - I think it ended up at about 9.2mm but I don't have a good way of measuring bores. I don't really care because the piston will be made to match. The resulting minimum cylinder wall thickness increases from 0.6mm (0.024 inch) to 1.0mm (0.039 inch) a very useful increase in bending stiffness.
Next the outer surface was measured for the locating flange and a start made on reducing the outer diameter that will fit in the crankcase (I've left the swarf on the lathe in this shot just for Simon).

Steve

[steve-de24]

A quick 20 minutes at the lathe enabled me to finish the diameter on the cylinder that locates in the crankcase. I wish I was more confident doing this sort of job - perhaps then I'd be faster - I seem to be constantly worried that I'm going to go from 'oversize and won't fit' to 'undersize, wobbly and unacceptable'. If anyone out there has a fool proof method then please let me know. However this time I got lucky and managed a push, wobble free, fit.
Steve

[sco]

Nice looking swarf! 

[Admiral_dk]

Looking good so far.

As to the reduced boring diameter - I read many places that going below 1cc swept volume increases the difficulties with starting and running a small CI (diesel) engine by a order of magnitude, and if this is correct, then any decrease in boring will increase the demand on the quality off your work ....

If you keep all other dimensions the same, you might be able to increase the boring later if needed ....

The original dimensions are a proven design - not that I'm saying do not change anything as this is a hobby and we all experiment 

Best wishes

Per

[steve-de24]

Per, thanks for your comments. Last night I was looking at the late Ron Chernich's excellent modelenginenews.org website. Regarding the Midge he shows the first page of Mark Lubbock's original 1996 article from Aviation Modeller International (but unfortunately just the one page!). In Mark's introduction to the project he explains that the engine was originally designed with an 8mm bore and 0.5cc capacity and he reports that it was an easy starting, nice running engine. Mark uses his engines for flying model aircraft and it was in the search for more power that he increased the bore to 10mm. The AVI article actually presents both the 0.5 and 0.8cc versions of the engine. From this I think I will be ok with my 9.2mm bore - if nothing else I might have the only 0.66cc Midge in the world. Time will tell.
Steve

[steve-de24]

I managed a bit more shop time this afternoon. The chuck and cylinder now need to be transferred to my simple indexing fixture on the mill so that the ports can be drilled and transfer passages milled. Last time I drilled using this kind of set-up I wasn't happy with the result and came to the conclusion it was because the work piece was not rigidly enough supported. So this afternoon's job has been to make a simple machine jack from bits in the odds and ends box. The result is shown in the picture.

I was concerned that there would be insufficient space when using the small milling cutter so a trial set-up was made. If I avoid milling in line with the chuck jaws I can do it with about 3mm clearance between mill spindle and chuck. Fortunately the small milling cutter I've got has a 35mm long shank (most of the others are 25mm) so I'm able to have 25mm in the collet and 10 sticking out to give me clearance. Drilling the holes shouldn't have a problem with clearance.

However it all has to wait because duty calls and I've got to go shopping - oh whoopee.

Steve

[Ramon Wilson]

Steve - that's some nice clean work you are doing there, the Midge is a nice little engine 

Re getting your clearance on the liner for milling - have you considered mounting it on a mandrel?

Either a straight one with an end cap (washer) and screw or an expanding one (easily made) Just make sure you relieve it where the milling breaks through so that as you remove the liner it doesn't burr and possibly gall on the mandrel.

Good luck with the rest of the build - nothing quite like a diesel 

Regards - Ramon

[steve-de24]

Ramon, thanks for your kind words. Can I also thank you for so readily sharing your expertise on this and other sites with the build logs for your ED Racer, Super Tigre and ETA engines.

Regarding my Midge, I did think about a mandrel but it was fear of jamming the cylinder on the mandrel with burrs on the ID from drilling the ports that put me off. Would an expanding mandrel that was only 12mm long ie. half the length of the cylinder, hold the cylinder well enough?
Then I'd be drilling into free space and burrs wouldn't prevent removal of the cylinder.

This evening I've thinned down (from 2 to 1.4mm) a parting tool that came on the end of a 6mm square HSS tool bit; I will need it to cut the fins in the cylinder head. I can't make up my mind whether it is better to cut the fins before or after drilling the holes for the hold-down bolts. One way gives me lots of burrs in lots of holes to tidy up, the other gives me an interrupted cut with a parting tool.... 

Steve

[sco]

I'd  take the interrupted cut route as the holes are more likely to stay in line. I also think I'd go  for a full length expanding mandrel with axial relief to avoid the problem with the burr.

Simon.

[Ramon Wilson]

Hello Steve - Thank you too.

I have done both but deburring the holes is a real pain. Holes first are definitely the better way to go  - as Simon says  (oops) there is far less chance of drill wander. The interrupted cuts are nothing to be that wary of, as long as the tool is sharp and the cut is gently applied you will soon be past the holes. Deburring the holes then is a doddle - just run a drill through on 'finger power'.

You should get away with using an expanding mandrel as you suggest. As long as its length in the bore is longer than the diameter it should easily resist the cutting forces. If you can arrange tailstock support (if doing it in the lathe) so much the better or possibly a temporary jack using a bolt and nut underneath if on the mill. It's quite amazing just how much holding power there is with such little amount of torque on the expanding screw on these mandrels. BTW it's definitely worth turning a cap-head head to 60 degrees to match a centre drilled hole than using a countersunk head to force it.

I have also relieved the mandrel to give clearance but this does mean cutting away some of the clamping force and can spoil the mandrel for further use.

You could also make a plain, stepped, mandrel and a stepped cap and just use a cap-head to hold it in place.

Hope that helps some - anything else I can assist with do say.

Merry Xmas to you and to you too Simon - any news on the L&B ?

Regards - Ramon

[sco]

Hi Ramon - still dreaming about the Lane and Bodley but also wondering about a Hoglet.

Merry Christmas to one and all,

Simon

[steve-de24]

Thanks for the advice guys - drilling the holes first seems to be the way to go, but that will have to wait until the cylinder is finished. I'm hoping to get some workshop time over Xmas but not sure I will achieve it.

Simon, is that a Corliss valve Lane and Bodley you are referring to? that would be nice.

A picture of my favourite piece of L&B 19th century engineering is attached, I just love their concern for health and safety. It even seems to come with its own vulture to clean up the blood and bits.

Best wishes to all for Xmas and the New Year.

Steve

[sco]

Simon, is that a Corliss valve Lane and Bodley you are referring to? that would be nice.
Steve

Yes it is a Corliss valve engine and is nice but the castings are chuffing expensive and only available from the States so postage is astronomical :-(

Simon.

[steve-de24]

I decided to go with an expanding mandrel to complete my cylinder machining. I made it just short of the finished cylinder length and will use partial engagement when milling the transfer ports. The mandrel can then be used to complete the remaining turning by reversing the  cylinder.
I took a photo to show my set-up for tapping the 4mm thread, a 4mm long grub screw will be used to do the expanding. I used my 3 inch tap wrench and this feels like a comfortable set-up down to about 3mm taps but below that feels clumsy. I got some good ideas for tapping smaller sizes from gbritnell's hay baler thread.
Next photo is my very clumsy set-up for using a slitting saw in the mill, its only saving grace is that it gets the job done. I will have to have a go at machining my own arbors for the milling machine  - I can see quite a few applications if I can manage it.
After finishing the mandrel, and out of interest, I attached the cylinder as it would be in the mill and checked the run-out on the turned section. Only 0.5 thou inch total - so I'm going to bed a happy man.
Steve

[steve-de24]

This is just a quick post to replace the photos that went missing when the site went down.

The first 4 show the machining of the cylinder, drilling the ports and milling the transfer passages.

The next 4 are of machining the cylinder head. The last pic shows my temporary minimal cross-slide handle that's needed on the Cowells lathe to allow the top-slide to swing around to 95 degree to the lathe axis.

Steve

[steve-de24]

The last catch-up post to show the machining of the fins. I followed SCO and Ramon's advice and drilled the hold down bolt holes first which was good advice - thanks guys. An unexpected benefit was the indication this gave to stop cutting when the groove depth had passed the holes - the wall thickness is only ~1mm.

Steve

[steve-de24]

For the first time in ages I got some 'shop time' yesterday. The result of my 3 hours efforts was a banana shaped con-rod    I think it must be due to residual stresses in the bit of HE15  (UK equivalent of 2014) aluminium I used. I machined it from a piece of 3/8"  dia rod and the mistake I think I made was to machine one side fully (including holes) before flipping it over to do the second side. The result is the crank pin hole is not parallel with the wrist pin hole!  Time for a rethink about the sequence of ops.
Steve

[Admiral_dk]

Dam - but at least you are learning something and can put it down as experience in the end 

Looking forward to the next chapter in the series 

[90LX_Notch]

Sorry to hear that Steve.

-Bob

[zeeprogrammer]

There's always a plus side. 

[steve-de24]

Thanks for the kind words guys; they say you learn more from your mistakes than when everything goes ok - I just wish I wasn't learning so much at times....

But upward and onward. Today I decided to rough out the crankshaft from a piece of 3/4 inch bar and I needed to reduce 45mm of this from 19mm to 7mm dia.

I'm only using leaded EN1a (UK equivalent to 12L14) but I'm hoping it will be ok for an occasionally used engine - and I'm making this engine mainly as an exercise to get familiar with the different procedures involved.

On another current thread someone is asking for information on the Cowells ability to machine steel so this looks like a good opportunity to give a practical example. I will put more details than usual in this post - apologies if I bore anyone.

I use HSS tools and keep them as sharp as possible to keep the cutting forces down.
The roughing cuts were carried out using the back gear at 180 RPM.
I used the auto traverse which gives a feed rate of 0.055mm per rev.
This would be extremely boring to watch but the Cowells has a cut-out to automatically disengage the lead screw so you can set it going and get on with something else.

My rule of thumb for machining like this is :   Depth of cut(mm) = 16 / Dia(mm)    but I round down to make it easy to keep track of how much I've removed. I also don't take cuts greater than 1mm (ie 2mm off diameter). One of the advantages of machining like this is the tool chip breaker works and the swarf is in easily managed small curls.

So I reduced the 19mm diameter to 7mm with the following passes.

(a)  One pass with 0.5mm depth of cut - diameter reduced from 19 to 18mm
(b) Two passes with 0.75mm depth of cut - reduces diameter to 15mm
(c) Two passes with 1mm depth of cut - diameter reduced to 11mm

At this point work stopped for a cup of tea and Jaffa cakes (hey I am English).

When I got back I put on another 1mm depth of cut and set it going. I got two surprises; firstly, by mistake I had put the depth of cut on twice (before and after tea break!), so it was now 2mm; secondly, the Cowells didn't miss a beat - so I left it alone.  This reduced the 11mm diameter bar to 7mm in one pass. I am impressed with what the little Cowells can do - but I think I'll stick to a 1mm maximum depth of cut in future knowing that it is well within the lathe's capabilities.

The photo shows the resultant swarf - that on the left is from my 'normal' roughing cuts, nice small curls which are easily vacuumed - on the right are the big curls from the 2mm depth of cut pass.

Steve

[Tennessee Whiskey]

Looks like the Cowells do a great job. I'll also say, that by looking at the swarf, you definitely know how to grind a HSS tool bit  . Nice job Steve.

Cletus

[steve-de24]

Cletus,
I can't take credit for the tool grinding because it was already ground when I bought it; it's from a set by SOBA, a cheap Indian make. I do however make sure it is super sharp, nothing fancy, just a piece of 1200 grit wet and dry sitting on a nice flat piece of 1/2 inch aluminium plate used to hone the face, and a rolled up piece of the same wet'n'dry is used to hone the chip breaker 'gutter' which forms the other side of the cutting edge. My only criticism is I'd wish they'd used a better grade of HSS because it's a bit soft, and these tools would be no good for production work because the cutting angle is far too acute - so it's not going to last long before it needs re-honing.

I've attached a couple of photos which show the 1mm and 2mm depth of cut passes.

Steve

[paul gough]

Steve, Thank you for taking the time to give details of some cuts and illustrated with the photos. It shows that this little lathe can take a decent cut without any trouble if one uses appropriate feeds, speeds and tooling. It also demonstrates the utility of the power feed and of course that now rare item, back gear. As Jo mentioned cutting fluid, I suspect with a spouted can dripping some soluble oil or like lubricant/coolant onto the cutting point the tool would last longer and perhaps even allow for larger than 1mm cuts without too much problem if you did not mind 'working' the machine a bit harder. Regards Paul. 

[steve-de24]

Paul,
Although I've demonstrated that you take a decent depth of cut with the Cowells you have to realise that in order to do so I'm having to use a very slow feed rate.  The carriage is only travelling at ~10mm per minute so each pass in the example I showed was taking 4.5 minutes. If it wasn't for the auto cut-out on the traverse it would be like watching paint dry. You can't beat one of the basic limitations of this small lathe which is that it has only got a 90 Watt motor. I think it is fair to say that the Cowells uses it's limited power as well as possible but if a high metal removal rate is important to you then perhaps a Cowells is not for you.

The other Cowells limitation is the size of work that can be accommodated. There's still quite a bit to choose from - the smaller steam engines, or scale down from larger plans - i.c. engines up to 1.5 possibly 2 cc - gauge 1 locomotive should be possible?. The range of models increases dramatically if you have access to a friend's larger lathe for flywheels etc. you can still do the bulk of the work on the Cowells.

For me it was either a small table top lathe or no lathe at all - so I live with the Cowells limitations.

Regards,
Steve

[b.lindsey]

Steve, I am enjoying and learning from your comments also. Haven't tried back gear yet but your 1 & 2 mm cuts are impressive.

Bill

[paul gough]

Steve, Thankyou for your additional comments. I am aware of what is appropriate for them and was not intending it could be an alternative to a Myford or some such.  My intention behind the enquiry was to gain a credible/practical confirmation of this lathes ability to take a heavier cut or machine a hard material occasionally than I was able to determine from other sources. My first choice was for a Wabeco 2000, about the same price as a Cowells though larger, but have been turned off by the problematic performance of the 1.4kw motor and or electronics reported to me from a number of different sources. Like you I am going to have a one lathe and mill set up and as the Cowells definitely fits a particular capacity range I wanted to be sure it could do what I require. Again, thanks for your help. Regards, Paul

[steve-de24]

A quick half hour enabled me to take the crankshaft blank over to the mill and finish off the roughing out of the crankpin. Last year, as a preliminary to starting this engine, I made a fixture for machining the crankshaft that Ramon recommended on a different thread; the photos show my version of it reduced to Cowells size. Its first use is to hold the crankshaft on the mill so that the material surrounding the crankpin can be machined away. First the centre of the crankpin is centre drilled and a small endmill touched on the surface to mark a 'keep out' zone.
The oversize crankpin was machined by eye to be roughly octagonal to ease the final turning. The last pic shows the result.
The crankshaft blank will now be put away until I've made a con-rod.
Steve

[steve-de24]

And the 2nd con-rod is consigned to the scrap bin 
This time I messed up reaming one of the holes - it ended up 'lobed' 
I think I'm going to have to take some time away from this project and get my reaming technique nailed. I will have to practise with 6082 instead of HE15 (equivalent to 2014) because I've got limited supplies of the latter.

The problem I've got is that I have to drill and ream a 3/32 inch hole and a 1/8 inch hole and I need advice on :-

The appropriate drill diameters (I have a metric set going up in increments of 0.1mm).

Reamer speeds and feed rates (remembering that I'll be manually cranking the handle).

What lubricant - I've got WD40 and a Drill-Tap cutting oil.

Any advice gratefully received because at the moment I'm not sure what will come first - making the con-rod or running out of material!
But in the meantime it's  for me.

Regards,
Steve

[90LX_Notch]

I like that fixture.   That can be a very useful piece of kit.

-Bob

[Jo]

The problem I've got is that I have to drill and ream a 3/32 inch hole and a 1/8 inch hole and I need advice on :-

The appropriate drill diameters (I have a metric set going up in increments of 0.1mm).

Reamer speeds and feed rates (remembering that I'll be manually cranking the handle).

What lubricant - I've got WD40 and a Drill-Tap cutting oil.

Hi Steve,

HE15 drills, reams and machines beautifully  Its that other stuff 6082 (HE30) or worse  that is more troublesome but that is not suitable for conrods.

I normally drill and ream my connecting rods as the very first operation, the holes providing the reference for the rest of the rod. They are co-ordinated drilled in the mill leaving 0.2mm for reaming. Then I put the reamer through with the mill at the same speed, sometimes as it goes through I think I should have slowed that down  but it works ok. I don't normally use lubricant for HE15.

Jo

[Jasonb]

Like Jo I would go 0.2mm smaller. There is not much flute space on small reamers so you may need to withdraw part way through to clear the swarf

I do tend to use a tapping/reaming fluid so your tap magic would be similar.

If you have machine reamers then the general reccomendation is half drilling speed. If like me you mostly have hand reamers then I run those as slow as I can in the mill or lathe faster and the taper can cause chatter.

[Bluechip]

I don't know about reamers, but this stuff stops step-drill & cone cut chatter dead in it's tracks. 

https://www.rapidonline.com/g-j-hall-09a-60-cutting-compound-60cc-86-2292

Few months ago step-drilling 2mm SS sheet to 24mm with Trefolex and was having no end of grief.
Hideous shrieking noise and the drill bit was doing the same ... 
One spot of this and all was sweetness and light.

Dave

[steve-de24]

Jo, Jason & Dave,
Thanks for your replies. I think Jason may have solved my problem in that I was using a hand reamer in the mill but going too fast (half drilling speed) and I got chatter. So next time I will try to go as slow as possible and use my drill-tap fluid instead of wd40.
I assume I should also feed the reamer through the hole (and back it out with the mill still turning) slowly, say 5 seconds per mm, slower??
Thanks again,
Steve

[Jasonb]

Yes keep it turning all the time and feed slowly in and out to what feels right, the amount of metal to come off will depend how true your drills cut to size.

[steve-de24]

Jason, thanks for the quick reply.  Steve

[steve-de24]

If anyone is wondering about the lack of progress it is because I've discovered one or more errors in the drawings I'm using. I wish I'd calculated the resultant port timing from the individual component dimensions before I'd started - I'll know next time. So the project is on hold until I can sort out how to proceed. Also not helping is that we're heading into summer and fine weather hobbies take precedence.
Steve.

[Admiral_dk]

If anyone is wondering about the lack of progress it is because I've discovered one or more errors in the drawings I'm using. I wish I'd calculated the resultant port timing from the individual component dimensions before I'd started - I'll know next time.

Very annoying to say the least 

I've been more or less the same place, when I tried to restore my old Viking 2.5cc diesel. My attempt at a new crank interfered with the bottom of the piston at BDC, so I calculated the port timing and "discovered" that the exhaust and transfer times where a lot more aggressive than the inlet time, so I just decided to remove a bit of the bottom of the piston - Wrong Move 

But I will say that wrong drawings are worse than just taking the wrong decision - I know what I did wrong, but you will have to figure out what is wrong with drawings made by someone else 

Best wishes

Per