P&W Wasp Jr. in 1/6th scale

[Mike R]

This is the start of me tracking my build of a P&W Wasp Jr. in 1/6th scale, from drawings by Bob Roach and castings from Bruce Satra.  I started this about 11 years ago when I acquired the drawings and the castings, and about 9 years ago I got a tiny start on building but realized that my skills in the shop were not up to par yet so it sat for a long time.  In the mean time I've played with a few other small projects and dabbled here and there on how to cut the fins on the cylinders.  That was a real challenge for me but I recently figured it out, which has prompted me to post my progress.
So here is the status to date.

• Cylinders are nearly done, need to have bolt pattern drilled for fixing to crankcase, and a pair of flats added to the large flange. 
• 8 connecting rods done (earliest parts made - not perfect but probably good enough for bits that will be hidden.)
• Crankcase done (but may yet get remade).  Has a few discrepancies I may want to address, second (or is it the third one) will go faster.
• intermediate crank support - by far the nicest part I've made to date - from a fit AND finish perspective.

I'm posting this here but as my progress has been so slow that glaciers outpace me don't get too hung up waiting on the next chapter.  Work interferes still far too much.

Attached are a few pics of the current status, one of the cylinders and the connecting rods after I made them.


Mike

[Jo]

You are moving along nicely Mike

I have a few sets of castings that are maturing while I wait for my skills to improve sufficiently that I feel confident to do them 

Jo

[Jasonb]

Lot of work there, this may spur you on. The engine built by Ron Harris that was at the recent London MEX

[Vixen]

Nice work Mike 

It's always good to see someone else embarking on another 'round engine'.

Keep the occasional report coming our way.

Mike 

[Mike R]

Thanks all.

Jo,
I also have other castings waiting their turn (Wright J6 1/4 scale) as well as a locomotive (LMS Duchess 3 1/2" gauge).  Going to keep working on the supporting bits carved from solid until I'm comfortable tackling the castings...

Mike,

Your work on the Bristol Mercury is inspiring, keep it coming! 

Jason,
No need to torture me with that sort of photo to spur me on.  I'm motivated but lacking time due to the need to eat (i.e. work).    Simple math says I need to work 10 more years but I'm aiming for less.  The shop does provide a nice break occasionally but rarely in doses large enough to make real progress, hence the slow pace. Maybe once the kids are adults I'll have slightly more free time.

Mike

[Mike R]

In the spirit of making use of the tools I've collected and modified over the years I decided to put my Emco Compact 5 into service making the 18 valves required.  The Emco is a CNC machine that I've converted to newer stepper motors and modern electronics for fun(Mesa 7i76e Ethernet card and LinuxCNC). 


I choose to make the valves out of 303 stainless after much searching for what material is suitable for small engine valves.  As it appears that just about any steel is serviceable I decided on 303 SS so that there is at least some corrosion resistance for the long periods where this engine will not be running.


The 1st technique I tried was one I'd seen recently on a couple of Youtube videos, where the desired small diameter is turned in one pass from a much larger piece of stock (think 10:1 or more diameter ratio of stock to final part).  This technique failed on first attempt even in 3/8" aluminum due to the lack of rigidity of the Emco - it would grab and then stall.  Next I decided to go at it in steps.


I programmed up the lathe to cut the valve in distinct steps (using Fusion 360 CAM and some hand editing afterwards):
1.  Turn the stem top down to diameter (3/32") for about 5 mm
2.  Machine the retaining groove (home ground tool, 0.44 mm wide (~ 17 thou)
3.  Turn the next 5mm or so down to diameter
4.  Turn the remaining stem length to diameter and shape the head
5.  Part off
6.  Return to start position to help set bar for the next valve.


I initially trialed the program on some 3/8" diameter aluminum and all went swimmingly (sure sign of issues to come ).  Once the 303SS was in I had to make several program tweaks to accommodate the harder to cut material.  The Emco is about as rigid as a wet noodle when compared to my Standard Modern 1340 so I needed to learn more about its capabilities and limitations.


In the end I succeeded in getting a repeatable program but the parting off - even at 3/8" (~10mm) was hard for the lathe. I could actually see the lathe deflect as it tried to cut. Total time per valve was on the order of 20 minutes when taking into account a few programmed pauses I had to let me check on the progress.


My concerns are that the valve stems are still in need of what I would consider final finishing as they are a couple of thou too large, and the surface finish is not what I want (some are much better than others).  I'll probably end up chucking the head up in a 3/8" collet and lapping the stem to size and finish.  Anyways here are the parts as they sit now in a partially finished state.  There are 22 of them as I will inevitably either loose one, or stuff one or more up in setting up for lapping, etc.   I got distracted and moved on to another part / project after this so I have the lapping to come back to...  (look forward to?).


If you are into boring machining video I took one of the Emco working on one of the valves, its quite slow going as it nibbles away at the metal!  The video is 5 min 46 secs and its only the middle and head profiling portion, with some terrible audio- me talking very little and even then much too quietly to be heard and the lathe squealing at times... You've been warned. 
  <a href="https://www.youtube.com/watch?v=2LJdOdJzXs0" target="_blank">http://www.youtube.com/watch?v=2LJdOdJzXs0</a>
[/size]
[/size]Mike

[petertha]

Very nice. Was this engine ever featured in a magazine like Strictly IC? Or maybe I'm just recognizing the names (Bob Roach & Bruce Satra) from that era. I think there is a 1/4 scale version made by Paul Knapp in the museum. Anyways just curious about the basics - displacement, fuel/ignition type, lubrication.... Also what are your cylinders made from?

[nj111]

That's very nice work so far.  Interested to know your method / tooling used to cut the very fine cyl head fin grooves. With regard to improving the finish of the valve stems I made some in exactly the same way as you have done - but with a little Denford Orac (which I upgraded to Mach3). I finished the stems prior to parting off using some replica external hones that I made some years ago - basically a home made version of a Delapena / Sunnen external hone, this works superbly for valve stems. Nick

[Niels Abildgaard]

  This technique failed on first attempt even in 3/8" aluminum due to the lack of rigidity of the Emco

 The Emco is about as rigid as a wet noodle

 You've been warned. 

From the boring video it can be deducted that overhang- that is distance from spindle bearing to work zone- is at least 100mm.
It is not possible to see the  load path from cutting point to lathe bed and then to Gaia but let me suppose it is some kind of QCTP on the miserably engineered  EMCO tool slide.
My advice is to make or buy a short ER  collet system and a decent compound slide system.
See another boring (but much shorter) video





and less overhang (modified WM180 lathe)

<a href="https://www.youtube.com/watch?v=49ISMaCAB1k" target="_blank">http://www.youtube.com/watch?v=49ISMaCAB1k</a>

[Elam Works]

Mike,

A few thoughts. I see in the video that you are using molded indexable carbide insert tooling. These do not have a very sharp edge, indeed they are manufactured with a slight radius to make the cutting edge stronger. Various coatings increase that radius slightly. They work more like a bulldozer shoving the metal aside than shaving it off. So when you are turning very small diameters like your valve stem it just pushes the work piece aside. Also, as the work diameter gets smaller, and you take shallower cuts, the edge break becomes larger in proportion to the work. You may not even be getting off the edge break on very light cuts, resulting in effect a very high negative cutting angle. Combine that with stainless steel, notorious for work hardening, and it is no wonder the surface finish is poor.

You would be better off with a ground carbide cutting edge. You could take a used insert and grind a new cutting edge on that. It also looks like the insert you are using has too large a nose radius for the light finishing cut you are trying to take. That is working against you by increasing the length of the cutting edge engaged with the work, increasing the amount of torque and rigidity required.

Why not consider using a ground HSS tool bit? It will take a keener edge than carbide can hold, and you can grind a very high top positive rake that will reduce the cutting load (and so reduce torque and work deflection). It will also hold a much sharper tool nose radius. For really small diameters like you are attempting, the nose radius ought to be practically dead sharp, with just a stroke of a fine oilstone to give it a thou or two radius. You are going to be feeding slow anyway, you don't need a big nose radius.

I don't see the cut off tool, but I wonder if it is not the standard 3/32 wide carbide insert parting bit? If so, that is too wide in proportion when you are parting off work that is just 3/8 inch diameter. You want something that is just a 1/16 or little less wide. A ground grooving bit (not molded) would be ideal.

Summary: Sharp, ground edges (the keener the better); positive rake; and minimize the length of the cutting edge engaged.

Nick, I like those external hones; thanks for posting the pictures. I have a project that would benefit from something like that.

-Doug


[fix typos. 18Mar20. -Doug]

[Mike R]

Hi Nick.


I appreciate the external hone pictures, I was planning on making something and that has given me some good direction on what I'm doing.  I'll fill everybody in on the cylinder fin cutting in a proper post, basically involved using a slitting saw on the mill and rotating the work.  The heads are as purchased so no credit to me on that work except perhaps having the foresight to buy them while they were available. 


Hi Neils,


I do realize that the overhang from the spindle bearing is quite long.  I briefly looked at an ER collet chuck for the lathe (cheap Chinese version) that would bolt in place to the chuck adapter I have fabricated and surprisingly it provided almost no benefit in terms of shorter overhang.  That leaves customizing my own ER collet onto the strange nose of the Emco.  Doable but I just wanted to use the lathe for once, not take on yet another non-engine project.  I may revisit but as I don't have any smallish ER collets I'd need to buy them as well and its just not something I'm interested in at this time.
It does have the Emco tool changer - not sure if its miserably engineered but it certainly has its shortcomings.  I did reduce the stick out of each tool to a bare minimum, I could have maybe gained another 5mm but with more effort.  Again I wanted to use the tool changer as that is the appeal of the CNC - it can in theory make the whole part in one go, changing tools as needed.  So changing out the tool changer for a compound is not in the cards, I'd like to learn what the lathe is capable of and how I can maximize it through better choices of tooling, tool paths and setup.


Hi Doug,


I appreciate the feedback.  I was being a bit lazy in terms of using what was available to me on a few fronts. 
I recognize that the insert was maybe not ideal.  I've actually ordered a set of inserts designed for aluminum which at least provide the sharp edge but not the smallest nose radius so that gets 1/2 way there.  I had actually gently lapped the insert on a diamond sharpener to try and get a sharper edge but may have made matters worse rather than better.
I was using the tool changer on the lathe and it has positions for 3 square shank turning tools and 3 round positions (i.e. drills etc).  I needed 1 for the groove at 0.44mm,  1 for the majority of the turning and 1 for parting off.  In hindsight I could have left the valve attached and parted off in the manual lathe, freeing up a tool position for a "roughing" tool and 1 for fine finish. 
The parting tool is ground from a HSS bit, and its about 0.055" wide (under 1/16").  I could have made it narrower but again my laziness prevailed as I thought it was small / thin enough.  Maybe one day I'll learn...   
This was one of the first times I've tried setting up the lathe for "production" parts and if I'm so inclined I can have another go at the valves as the material was cheap in this size and I consumed less than 2 ft (about $6).  Might be worth it just to learn how to setup the lathe better to get parts that are "ready to go" off the lathe. 


- Mike


 

[Mike R]

Pertha,

I don't believe it ever featured in a magazine.  Ron's Model Engine News had a few mentions of it but I've never seen a full magazine / blog / thread on a build.

[Art K]

Mike,
Just wanted to say hi! and that I am following along, or at least I am now. I look forward to your work on Your Wasp Jr. I do like round engines and have 2 on my bucket list.
Art

[Niels Abildgaard]

It does have the Emco tool changer - not sure if its miserably engineered but it certainly has its shortcomings.

- Mike

Hello Mike
Found it on Lathes

http://www.lathes.co.uk/emco/page8.html

Was not aware that Emco had made such a thing

Kind regards

Niels

[Jasonb]

can't you use some tailstock support, I always use a very small BS 0 ctr drill on the overlength valve stems and then turn the full length rather than in steps. Excess length turned off later by holding stem in collet or a split bush.



DCGT insert will help a lot and also angle the toolholder or get one that presents the insert so there is more clearance when cutting the shape of the valve head as the larger engagement width is when you are getting chatter.



I'd saw up your stock into 2 x valve + 25mm then you can keep the ctr section to hold then if you need to lap the valve stems ( I have not found the need to) and also when lapping the valve into it's seat

If it won't part off well then saw off and hold by the stem to face head to thickness with light cuts, again hold stem in collet or split bush.

[Vixen]

Hello Mike R

I agree with Jason about the use of DCGT inserts and making double ended pairs of valves. I go as far as also turning the outside diameter of the bar stock after turning the first valve. This to ensure the stock is round and concentric with the first valve, before it's reversed for the second valve.

I also have a EMCO Compac 5 CNC which I use for this sort of work. It is a light, flexile (as in bendy) machine but with light cuts and the right tooling, it can do great things. I discarded my Emco auto tool changer in favour of a quick change tool holder and half a dozen tool holders. This allows you to get the saddle closer to the work and reduces tool push off. Unfortunately, this makes Jason's suggestion of a rotary centre, more difficult.

For any work that may require accurate repositioning, I use the EMCO collet adapter. I know it's an ugly device with about the same overhang as the three jaw, but it is so much more accurate at recentering than the three jaw. A set of metric collets will cover the whole range of diameters from 16mm down to 1mm. You dont need to buy 'special' imperial collets, you can hold anything round in one of the metric collest.

Mike

[Mike R]

Well, I haven't revisited the valves with any kind of commitment yet, though I did happen to cut a piece of 1144 steel (3/8" diameter) on my (big) lathe to make a pin, turned it down in 1 step and boy oh boy did that cut nice.  I may give it a go instead of the 303 SS.


What have I accomplished?  A whole lot of non-model engine stuff (rebuilding a cottage).
And a lot of shop related stuff (rewiring the mill was a big one, lathe metric gear change set, etc.)


On the engine:
Looking over things, its has been a long time since I posted. 


I had been playing with gears for the lathe and gave making the gears for the engine a go.  I decided to redo the layout of the geartrain to use metric gears (mod 0.5) versus the as drawn DP48, as the cutters were significantly less expensive. 


So after drawing it up in CAD to be sure it was all going to work I set about making some gears.
They turned out to be relatively easy to make with the 4th axis setup and turned out quite nice.


I just recently got the master connecting rod fabricated, after much relearning about my mill and its capabilities and more importantly, limitations.


Not perfect but as it will live hidden inside I may just leave it - some small cosmetic issues to do with the head on the mill rocking around under high cutting forces or even just high accelerations.


I may have lost count of the number of small tools I broke getting the final product, but at least being small they cost less to learn with.


Also half fabbed up the 8 pins to hold the link rods.  The drawing calls for some special geometry on the ends that I'll need to setup to do (small step at half width on either end).


I also started on the crankshaft but stopped while I got the mill rewired.  So lots of tasks to pick up again (valves, crankshaft), but am hopeful that as I spend more time in the shop I'll spend less time making scrap and more time on useable bits.

[Mike R]

So I had another go at making the valves -  10 months later!


Taking advice given I made the following changes:


I added an ER32 collet chuck to the lathe
I used a DCGT type insert (i.e. sharp insert intended for aluminum)
I set the insert holder at an angle to the axis to allow clearance while profiling
I skipped the parting off and used pieces cut to 2 x valve length + about 20 to 25mm


Another fairly major change I made was to increase the speed of the spindle. I had been running the motor flat out, but was in the wrong belt position and had a limited speed at the chuck (probably ~1000rpm).  With the belt moved over I found a happy speed at ~2300rpm (higher and there is some awful lathe/motor noise produced even without cutting!).


I programmed in some backlash compensation on the cross slide (X axis) which helped with shaping the head area.


And lastly I changed over to 1144 steel - this last one was more from being out of stock of the 304 stainless I had been using.


To show the improvement I have taken a picture of a previously made valve with a new one.
You can clearly see the surface finish improvement.
Also, most have come out within a half thou of the desired size (0.0937")  Chasing that last one, to half-thou variation is a real pain, and I'm not confident I'll solve it with a reasonable amount of effort.
For now I'm happy as I can likely adjust the valve stem diameter down 0.0005 to 0.001" by lapping if needed -I'll wait until I've made the valve cages to see how close a fit I can achieve.

[cnr6400]

 

Outstanding improvement on the old vs new valves! Hope you collected arr that swarf - you could start a pot scrubber manufacturing business!     

[Vixen]

Pleased the advise proved useful. Nice finish on those valves

Mike

[Mike R]

So I thought I'd keep at the valve train and see if I could do some work on the rockers.  Need to make 9 each of a left and right (intake and exhaust).  As drawn there are 2 options, one carved from solid phos. bronze, and 1 from 7075 aluminum with an oilite bushing pressed in.  The later is the one I had chosen to draw up. Regardless of the rocker design, the adjustable tappet screw at the end is the same.


The debate I'm having right now is the tappet screw end seems overly complicated and I'm looking for alternate designs before I begin making the rocker arm in case it is affected.  I don't intend to run this engine hard or long so longevity isn't a primary concern, though I do not want it to chew itself to pieces after 20 minutes.


Attached is section view of the rocker arm and tappet screw. The screw is simply called an M4 on the drawing (presumably M4 x 0.7 but could be x 0.5).


Inserted in the end of the short screw is a 3/32" ball that is captive and has a flat ground on it, but is intended to be free to rotate in the screw thereby always pressing with the flat on the valve stem.
The screw also has a 30deg bevel at the ball end.


Seems like those are going to be a challenge to make.  Anybody have options on simplifying this setup, or am I worried about nothing? 
Unfortunately,  I can't move the adjustment feature to the pushrod end due to the head casting being closed over back there (see picture).


Regards,


Mike

[cnr6400]

Hi Mike. the ball with the flat seems like a difficult way to get a hardened surface running on the valve stems. The other thing is, unless the setscrew is long enough to accept a nut, or the setscrew is pinch clamped somehow, the setscrew will likely start rotating and loosen itself, ruining your hours of valve adjustment quality time!

Maybe a longer setscrew (no inset ball) made of 0-1 tool steel and through hardened, then tempered to straw colour, would be better. A "model size" M4 nut, with reduced hex size vs screw dia, could be used to allow locking the setscrew in place on the rocker after adjustment. The smaller hex would give a more "scale" appearance, although a normal M4 nut would likely  function just fine. I'd also suggest at least 4 full threads in the rocker, or more, even if you have to thicken it a bit on the outside face to get the additional threads. This will help stabilize the setscrew and reduce risk of it and the nuts coming loose. Just food for thought. 

[Mike R]

Thanks CNR,
The rocker arm does call for it to be split at the end and pinched to retain the tappet screw.
I'm also not opposed to a little chemical fixation to keep things in place once I'm happy (loctite)
Digging through the drawings much further towards the back of the set I came across a valve train layout drawing that helps me understand better the space constraints, while keeping things looking scale from the outside.
There's no room for a protruding screw and extra nut as it would interfere with the valve cover, and even as drawn this design is very close with no extra space.
The only other idea I've had is to make the rocker arm fixed with a curved tappet face similar to most commercial small 4 strokes (creates a line contact on the valve stem face) and then to adjust I'd need to

• do something with the pushrods (again not possible due to pushrod tubes covering the pushrods)
• make the rocker pivot have an eccentric relative to its mounting hole in the head. Figuring out the right amount of eccentricity to provide adjustment that is reasonable is a big experiment I don't think I want to take on.

I may just soldier on and make as drawn.  Hopefully it doesn't drive me crazy

[cnr6400]

Hi Mike, Looking at the valve rocker housing design section drawing you posted, it sure isn't optimal for space inside! I think the ball in screw design is a pretty bad / optimistic design feature by the designer, at the M4 screw size noted. If it was a 10 mm dia screw, and the socket could be machined with a ballnose cutter with lots of metal either side, different story maybe......

Apart from the adjustment need, a cylindrical roller mounted on a pin in the rocket might be better in terms of lube and reduced friction vs the ball setup. If you do change plan to adjustment by eccentric pivot bolts for the rockers, the roller might be worth considering. The rocker would have a slot cut in it to receive the roller. if a case hardened dowel pin is used with a cast iron or bronze roller, parts will run nicely together and there is a flow thru path for lube oil. The rocker holes for the dowel could be peened at both sides to keep the pin retained. Very space efficient.  Just a thought. Roller rockers are very common and well proven in high performance V8 car engines.

In any case, good luck with it. 

[dieselpilot]

I would think that was intended to be modified from a flat ball point set screw. They didn't make this clear in the prints/parts list?

https://www.elesa-ganter.com/en/www/Machine-elements--Ball-point-screws--GN605

[petertha]

{deleted post, I misunderstood sketch}

[Mike R]

Dieselpilot - thanks for the link - buying this premade is a real good idea.  At worst I just have to modify it slightly. 

B.T.W. the instructions are:
"Tappet Screw - Cut from M4 M.S. screw.  Lightly crimp to retain ball. Ball must swivel freely"
And the socket for the ball is indicated to be made with 3/32 bull nose endmill (my belief is that is meant to be ball nose but who knows...)
And the ball is indicated to have 0.4mm taken off to create the flat.

CNR - A roller tip is a good idea if needed to go a different route than as designed.  But still would need to figure out an adjustment mechanism that is reasonable to build in this scale.  I also realized that making the rocker shaft eccentric wouldn't work unless I opened up the shaft holes through the head and then put bushings,  starts to get ugly quick.

I think I'll try building it as designed, with purchasing the premade ball tip setscrews, and hope it works well enough for the little running it will get

[Mike R]

Pistons (and rings)

As I wait for various materials and items to show up I thought I could tackle the machining of the pistons. 
The pistons are supplied as a casting but they leave very little material for clean up.  I also only have 9 so no practice parts allowed!
The desired finished piston diameter is by the drawing is 0.002” under the bore, so 0.873”
The as cast diameters vary from max ~0.899 to min 0.884” meaning at best case I have 0.013” to remove (radius) and worst case only 0.0055”.
The drawing set does call up a method of machining that I have decided to use as it seems likely to succeed. 
The piston is chucked up in a collet and  the skirt lightly is lightly faced, and a small counter bore machined 1/16” deep and to 0.740” diameter (to sit on a fixture).

I machined up the fixture to hold the piston, centering the skirt on a 0.740” diameter shoulder protruding 1/16”.  The piston is then held by a pin through the wrist pin hole, into a ¼-20 screw with a loose fitting cross hole.
The pistons have cast into them the wrist pin hole and it was quite clean, allowing me to insert the reamer (7/32”) in via the shank end and to then clamp up the reamer in the drill and ream while pulling through, the shank acting as a guide to get the reamer started at the right angle.
And that is where I got to so far.

Reading ahead, I need to machine the diameter to size, and put in 2 piston ring grooves and this is the point where I again question the design.
1.   Are 2 piston rings really necessary on a cylinder with 7/8” (22mm) bore and 0.984” (24mm) stroke?  My thoughts are 1 will suffice but I have more research to do here first
2.   The piston rings are only 0.41mm thick (0.016”) and this may be the biggest reason for me to deviate from the drawings,  making rings (and good grooves) at this thickness will likely be my demise – I can envision broken rings littering my workshop, along with hundreds of pounds of cast iron swarf trying to generate 18 rings of acceptable fit and finish.
I have read up on making rings (https://www.modelenginemaker.com/index.php/topic,3794.msg68914.html#msg68914) but the decision as to how many is still up for debate. Following the guidelines in the referenced post, I see that my suspicions on ring thickness (w) are likely well founded from the following quote:
 Many published designs use rings which are too wide, & I suggest w = 0.03 x B as a guide, with a minimum of 0.04”.
Using the formula 0.03 x B (assuming this is Bore as there is no B elsewhere in the article) we get 0.03 x 0.875 = 0.0262” so I would be looking at the recommended minimum of 0.040” or if I’m daring the calculated 0.026” which is still 0.010" thicker than the drawings.  The radial width (t) would nee to be modified appropriately as well, again the drawings have it at a quite wide 0.065" versus the articles recommended D/25 to D/30 which works out to roughly 0.029" to 0.035"
At the end of the day, I’m looking to build a runner, not a power house.  So whatever in theory works best for an easy to start, lightly loaded and relatively low RPM engine in this size is what I’m aiming for. Being able to machine said parts is also a necessity as it will never be a runner if I can’t build it! I'm afraid I have more reading ahead before I continue with the piston machining.

[Mike R]

Time for some long overdue updates.  I'm going to start in with where I am as I am seeking guidance (see below).
I was performing some of the last major operations on the cylinder heads when disaster struck.
I've uncovered a casting flaw - a pocket of plaster / dust / sand whatever that was unseen previously but connects from the combustion chamber dome to the wall of the hole for the glow/spark plug.  This was found on the 8th of 9 castings (the 9th was ok).
The first picture shows it in the holding fixture when I realized that there wa a problem with the swarf coming out - it just wasn't right.
The next shows it after an inital blast with compressed air - I didn't realize just how bad this was yet as I've circled the barely visible material that I had not seen previously
The last picture shows that the defect connects the glow/spark plug hole to the combustion chamber after I blew it out with compressed air some more and picked at it with a scriber.



Now the question is - what to do about it.  I fear that the castings my not be available anymore though I am trying and track down if there's a successor making the castings that Bruce Satra made.
Failing to replace it I will need to repair it or face the reality of a non-operating engine (or cylinder). 


Repair thoughts?  Being in the hottest area of the engine of course, makes the idea of using an epoxy dubious at best.  Even the best epoxies are only rated to ~190C (400F) and I expect that local temperatures when running will exceed that.


TIG welding?  maybe if I pack the area with carefully collected shavings of the same casting material I can gently weld it being careful to not overheat the head.  being a casting I fear that it won;t take kindly to welding though and I've not got any spare material to practice with.


Other ideas?  the defect is at a strange angle and location so its not conducive to just drilling it out and plugging in a new piece and shaping it...


Mike




 

[vtsteam]

What is the cylinder head material?

[Mike R]

Cast aluminum is the best I can do for material specification.  I checked all my notes and correspondence and can't find any reference to material specifics.

[vtsteam]

Then I'm guessing that's a blowhole caused by faulty homebrew de-gassing, based on the white powder residue. Or could be investment.

A new casting would be the best bet. For repair, you can talk to a professional TIG welder and get an opinion.

Is it possible to bore out and press a plug in? Doesn't look easy from here, but you'll know better with part in hand.

There are some aluminum brazes that might work. How hot can a model head run near the exhaust? Do people run CHTs and EGTs on these? If too close to aluminum melt temp, there's a thin line between braze application and damaging the head. Some so called "welding" brazes and solders for aluminum are much lower temp products, and I wouldn't bother with them

I know what I'd do, if absolutely no chance of another repair type, and that is, I'd fill the interior with greensand, blocking one opening, and then pour molten aluminum from my furnace into the other. But I can hardly recommend that to anyone else, and that's just last ditch. I have done that occasionally in the past and it has worked for me, but I wouldn't take a chance on a casting as valuable as yours.

[Mike R]

To give an idea of the scale of the issue and a different view of the "worm hole" as I'm starting to call it in my head

[Jasonb]

I think I would set it back up in the fixture, face that surface back a bit more then open it up and cut a fine thread for a top hat bush that goes all the way through to the combustion chamber so it closes off the large end of the flaw.

Turn up a suitable bush an fix with JBWeld (550deg F) then when set machine as you were when you found the problem

[Admiral_dk]

Jasons sugestion is probably the best ....

Welding such a small part is moore or less out of the question - but hard-solder could be an option :

https://www.tav3.de/    This stuff works really well - but would probably still be too much on this scale.

Best wishes

Per

[uuu]

I'm voting with Jason also - a sleeve in the spark/glow plug hole will seal things up.  Perhaps then a solder mixed with metal powder?

Wilf

[Vixen]

Hello Mike

Bad luck in discovering a flawed casting after all this time and after all your work. It would be nice if you could find a replacement casting and start again, but I doubt that's going to happen. So, we need to think about the best way to recover the flawed casting.

My take:        Normal TIG welding is out because of the small size and delicate nature if the casting. Forget any of the 'wonder' aluminium solders, oxide removal would not be possible inside the tiny void cavity.

I agree with Jason, an aluminium top hat insert would repair the plug thread and hi Temp JB weld to fill the void cavity. However I think you will need to do some extra preparation and exploration before applying the JB Weld.

Looking at you IMG 5907. I suspect the flaw extends further to the left between the two valve seats. There is a dark stain with what appears to be a crack/ flaw. you may need to open the edges of the obvious flaw and see if it extends towards the left. You may need to open the sides of the hole and explore the extents of the flaw You must to be sure all the white powder (investment??) and rubbish is removed.

A dental amalgam packing tool/ carrier/ gun (search ebay or amazon  one-use packing tools are not too expensive.) would appear to be a good way of packing the Hi temp JB Weld tightly into the cavity. Gaining a good access is another reason for slightly opening the hole inside the combustion chamber. 

I cannot think of any other alternative method of packing the JB Weld into the cavity.

Good luck

Mike

 PS don't be too worried about combustion chamber temperature. Methanol burns significantly cooler than petrol (gasoline)

[Jasonb]

Out of interest what holds the valve cages in place?

[Floating around]

Hi,
that’s annoying! Give me a couple of days and I will pm you the details of the current owner of the dies. He may have a casting or 2 left. There are plans to make more but no timeframe.

[Mike R]

Thanks to all who've supplied suggestions, I'll try and answer here as best possible.


I was probably still in a bad frame of mind when I posted and still recovering from the shock and disappointment of the discovery to have a full think through of the issue.

Floating Around - Please do PM those contact details - I've been attempting to track down known associates and have emailed authenticscale.com who listed Bruce as a contributor to their engineering and consultants in the hopes they can put me in touch with someone.

To those who suggest JB Weld (or equivalent) I may need to reconsider - I had in my haste assumed that the head would be too hot.  But on further review and research it does appear that the head temperature could be within the acceptable range for those high temperature epoxies, especially as I will not be running this engine hard. 
Jason - the valve cages are held in with Loctite 638 - rated to 180 C,  as recommended by the drawings - so if that temp rating is good enough for the valve cages then the epoxy is probably good too from a bulk cylinder head temperature perspective.



I got out my dentistry tools and poked and scraped to find the extents of the defect.  Then I gave it a long cleaning in the ultrasonic bath, so I'm pretty sure I got all the junk out that is contiguous with this void. I have mapped out and drawn on the part the "subterranean" cavern.  It is effectively hemispherical within the bounds I've drawn.  Making a plug that fits well between the valve cages is going to be a real challenge.


I'm actually not that worried about the side of the glow plug hole.  There is enough metal to support the thread of the plug that to me its just a matter of plugging up any gas paths.  I do agree that the proper fix for that portion is the top hat type plug and that is what I'll plan to do. 
I was more worried about the effect of combustion on exposed epoxy for the section that is nearest the valve cage.  But on thinking about it more, there is no reason I can't use both epoxy and a bit of metal facing - as long as the metal is held properly by the epoxy and is maybe keyed.  That also could be overthinking the issue.
Perhaps better would be a mix of JB Weld and aluminum shavings - I'd think that might be the solution for the combustion chamber area, or a bit of larger aluminum pieces back filled by the JB Weld and small shavings...


I'll ponder it some more - there's no hurry - been plucking away at this for more than 10 years - no need to hurry now!

[vtsteam]

Sounds like you have a good plan with Jason's suggestion.

Just for interest's sake, re aluminum braze products, I thought the following might be of interest generally:

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

[Jasonb]

Looking again I was concentrating on your comment that the void connects the combustion chamber to the glowplug hole and that the engine won't function.

Actually all it is likely to do is result in slightly lower compression ratio as there is no connection from the combustion chamber to anything outside. So the engine will work with the void and any filling is really only to even up the compression.

Maybe fill a syringe with JBWeld (no needle) and see if you can pump it in until it flows out the side of the Glowplug hole)

[petertha]

Sorry to learn of your discovery. I just so happen to be looking at some Harris aluminum braze products. I had some grand plans & might still try some Al-Braze for personal R&D. But the melting temp is getting up there relative to base aluminum. So aside from visual feedback on the flux state, not too much extra torch dwell time between flowing braze into the cavity & a puddle formerly known as Head. I have seen some of the low temp sticks. I have not heard good things about strength, but maybe for this purpose it might be suitable assuming combustion temp stays under melt temperature by a safe enough margin. Is there any similar scrap part in your kit you can experiment on?

I assume the bronze? valve cages are already in place further aggravating the situation?

https://www.harrisproductsgroup.com/en/Products/Brazing-Soldering-Fluxes/Brazing-Alloys/Aluminum

https://www.harrisproductsgroup.com/en/Products/hpg-na-albraze

[john mills]

what we used when i was an apprentice on full size engines was tapered plugs made of similar metal .drilled a hole then tapped the hole
with the matching tappered thread screwed in the plug tight.then the need plug would be placed touching the last plug .then the plug ground down and peened in then ground to the original shape .plugs were used to repair cracks or any defect holes re tapped as required .
screwed plugs peened in might be good .
John

[vtsteam]

So John, what I'm understanding, and please correct if wrong, is, two tapered threaded plugs, one above the other, in a single tapered threaded hole. Both peened to tighten them further? 

[petertha]

I thought about a threaded / Loctite plug too, but seems like there isn't a great way to position the plug/hole diametrically to encompass the void without intersecting other features - valve cage and/or plug hole. Maybe not insurmountable but definitely more fussy work.

[vtsteam]

John was mainly talking about full sized engines. I think the JBW fill and JBW/aluminum fill are most easily do-able, assuming head temps and hot methanol don't erode it.

[Mike R]

I'm planning on using the high temp epoxy as a filler around the biggest pieces of aluminum that I can reasonably fit into the spaces, so the epoxy becomes just a bonding agent and gap filler.  Being in Canada I am encountering again the conundrum that what is readily available for shipping online and/or in store just 150km away on the other side of the border is not available to me here.  J-B Weld offer a "High Temp" 2 part  (i.e. liquid) epoxy but I can't seem to buy it in Canada.  J-B Weld do offer a High temp epoxy locally, but it is a "hand knead" type and it will be too "stiff" to fill in the tiny spaces I want to flow it to.

Permatex in Canada advertise a high temp (500 deg F) epoxy that I'll buy and try on a sample and subject to a few oven cycles (don't tell my wife!) before filling in the cavity in the head.  Another reason for me to venture in to the Bermuda triangle of tool stores up here - Busy Bee, Princess Auto and Ottawa Fastener Supply (withing 1km of each other).  If I don't re-appear its cause I'm stuck in a loop roaming the 3 stores...

[john mills]

I was talking about full size engines one plug in each hole.
we placed plugs next to each other cutting into the last one  to fill in cracks using as many as it  took to cover the crack,but we had different
sizes and also used to repair damaged holes for studs or bolts when the hole was damaged after removing broken studs with taps and easy outs may be a drill as well.the would be drilled till it was clean and a plug fitted and faced off.then a new hole could be drilled and tapped
as required .seats for injectors remachined.even had plain tapered reamer and plug that could be used on valve seats would have an interlocking screwed plug to hold it in place.the peanig was used mainly along crack repairs to try to make sure it sealed.the plugs were
cast iron or aluminium .
john