1/3 Scale Ford 289 Hi-Po

[mayhugh1]

The tenacity of the super glue really surprised me. My original plan was to augment it with a pair of pressed-in pins at the front and rear of the workpiece for back-up until the threaded fasteners were added. Adding these pins though would have meant that I wouldn't have had room for the spare bearing, and so I tried the glue alone. I too was concerned about the between-centers operation. As I write this, the glue holding the parted off bearing halves together is still going strong after a one hour 250F oven bake, and I'll soon be going up to 300F. - Terry

 

[Gabe J DiMarino]

I have seen this done where the half's are silver soldered together . The way you made these could you of threaded the bottom half of the bearing and bolted them together .After boring and finishing , cut out the threads from bottom hemisphere of the bearing.

 

[ddmckee54]

I've seen the halves soft soldered together before, but never silver soldered.

I know Clickspring uses super glue a lot for work holding on a faceplate, and I think super glue is stronger in tension than it is in shear. The glue would have been in tension from the centers trying to wedge the pieces apart, right?

 

[Vietti]

I wonder if you could also use hose clamps to supplement the super glue in this application?

I have used a lot of super glue for machining and have had mixed results, sometimes it comes apart when it shouldn't and other times it is difficult to separate. I think cleanliness and surface texture play a role. Also large surfaces take a long time to cure even with accelerant.

 

[petertha]

I don't want to detract the post but I've had mixed results with CA in machining ops & I've used it for multiple decades in other hobbies. Hence my question. Sometimes it sticks (too well) even under machining/cutting loads. Other times it lets go. I noticed Clickspring CA face plates he always machines grooves. Initially I thought for excess glue to migrate to, or maybe maintains CA joint thickness to something 'thin' especially if you were targeting a dimension relative to the plate, machining to a thickness. But from my own experiments I think it also has to do with curing. Our metals are essentially zero porosity & permeability which is unlike a lot of conventional CA adhesion applications. I have smoothly 'machined' surfaces remain partially uncured for a lengthy time without the use of accelerant/kicker (another subject again) whereas roughly machined or sanded surfaces stick. Maybe the grooves have sufficient air allowance to cure in that area & which then extends into the much thinner joint for a full cure? I've also tried spraying one side with accelerator & the other with CA. That seems to help but often isn't practical.

In terms of release, that's another can of worms. Mechanical separation is easy enough to understand, the glue joint fails with a sufficient squeeze or a whack. But if you have very delicate parts, that's not possible without compromising them. So I have cooked parts to de-bond, but its not as pleasant as epoxy which somewhat predictably turns rubbery, softens & lets go. CA can sem to withstand higher temp to the point of burning when its at the point of releasing can make for ugly goo which is difficult to remove. Chemical de-bonding with acetone or CA-debonder is very slow because it only has a tiny exposed surface area of glue edge - a long journey to make its way into & through the thin joint. Even a 1-in2 glue area can be an overnight soak.

This would make a good (separate) subject post because there are a lot of learnings & examples to harness together.

 

[mayhugh1]

The individual bearings were parted from the workpiece and another sanity check made with each resting in the block and the test shaft running through them. Witness numbers was engraved on either side of each bearing's parting line to prevent mixing up the halves. The crescents were machined in the same setup. The bearings were returned to the lathe, and the oil grooves were cut. The grooves ended up deeper than intended (.070" rather than the .035") due to a radius/diameter brain fart. The temporary support fasteners were removed and the threaded holes reamed out for minimal clearances around the 6-32 cap bolts.

The super glue worked much better than expected, and the back-up threaded fasteners might not have been needed. Just two drops of glue were used to join the pair of six square inch surfaces of the workpiece. In the past I've had mixed successes with super glues and usually end up attaching myself to the parts as well as the parts to themselves. This may have been the first time I've used it as it was intended to be used.

I expected the parts to fall apart after a one hour 250F oven bake. But, after an additional hour at 300F the glue bonds had to be forced apart with a pair of wooden sticks cross wedged in the bores. After separation there were no obvious traces of CA on the mating surfaces, but the parts were allowed to soak in acetone for an hour anyway. Machining was wrapped up after drilling the oil passages connecting the crescents with the oil grooves
.
I had intended to nickel plate the lower halves of the bearings that will be in direct contact with the block. Bright freshly machined bronze and aluminum surfaces bolted together under oil saturated with combustion products sounded like a science fair project to demonstrate galvanic corrosion. However, when the bearings came out of the 300F oven with their chocolate oxide'd surfaces, I was less concerned and skipped the plating.

Fitting the crankshaft was less work than expected. With just the number two and three bearings installed, the crank turned freely with just the hint of a pair of snug spots that could have legitimately been overlooked. The number one bearing had a significantly tighter spot due to the greater runout of the number one journal. After an hour with bluing and 600g paper wrapped around various mandrels, the crank eventually spun freely with all bearings installed.

Plastigauge was used to measure the final results. I've used this calibrated wax thread to check clearances on full-size cranks where roundness and runout aren't issues, and crank angle isn't a consideration. With the model's less than perfect journals however, I can assume the clearance of each bearing approaches zero at a particular crank angle, and so what I'd like to know is each bearing's worst-case clearance. The Plastigauge helped me find these, and as the photos show they ranged from .002" to .003". - Terry

 

[user 48964]

The tenacity of the super glue really surprised me. My original plan was to augment it with a pair of pressed-in pins at the front and rear of the workpiece for back-up until the threaded fasteners were added. Adding these pins though would have meant that I wouldn't have had room for the spare bearing, and so I tried the glue alone. I too was concerned about the between-centers operation. As I write this, the glue holding the parted off bearing halves together is still going strong after a one hour 250F oven bake, and I'll soon be going up to 300F. - Terry

thought this might be of interest to you.

 

[stevehuckss396]

My spark plugs are in the engine he donated to the Henry Ford Museum. I made some replica BF-22 Motorcraft plugs. They are what would have been in the gt-40 at Lemans.

 

[peterl95124]

I find the "tenacity of super glue" to be unpredictable, I think for me most frequently on aluminum it tends to not adhere well, and that's even after filing the metal clean and leaving plenty of scratch marks. YMMV.

 

[L98fiero]

I find the "tenacity of super glue" to be unpredictable, I think for me most frequently on aluminum it tends to not adhere well, and that's even after filing the metal clean and leaving plenty of scratch marks. YMMV.

Cyanoacrylates, like Loctite don't work well on aluminum unless a primer is used, maybe that's the solution.

 

[mayhugh1]

After laying in the crankshaft, work began on the connecting rods. Gage pins in conjunction with a simple shop-made fixture were used to measure the actual rod journal locations with respect to the centerlines of the cylinders. These were checked against the assembly model which had been using placeholder connecting rods and pistons to establish clearances around the rods.

Compression ratio based upon the combustion chamber volume provided by SolidWorks was used to establish the rod lengths using placeholder pistons. The goal for compression ratio is between 6.5 and 7. The smaller value will give the electric starter a slight advantage when turning the engine over, and once running it'll be a little kinder to the head gaskets. The placeholder rods were finally converted into machinable designs.

In these particular heads, compression ratio seems to be very sensitive to the features inside the combustion chambers. The .020" head gasket itself makes a full one point difference. The exact volume will be affected by the geometry of the valves in their seated positions which hasn't yet been finalized. Once the valves are installed, the heads will be cc'd and an exact determination of the combustion chamber volume made before finish machining the tops of the pistons.

The rods are to be machined from 7075 in five cookie sheet batches of two rods each. Small batches are inefficient, but I wanted to use up the tiny bits and pieces left over from the huge chunk of aluminum that the rest of the engine came from. I can't believe it's nearly all gone. Surely, a shop elf is playing with me.

Each workpiece will be made up of a pair of bolt-together sections whose parting line passes through the center of the big end. The big ends will be machined to accept bronze shell bearings while the small ends will include integral bronze bearings. While safely hidden away from cutting tools inside the workpieces, the rod bolts will hold the workpiece halves together during machining. While the rods are being cut free, three roll pins will continue supporting the workpiece halves.

The actual first steps were to prepare the workpieces. In order to simplify later setups, all five sets were made identical to within a thousandth. The large workpiece halves were drilled and threaded for 5-40 rod bolts. The small halves were through-drilled to clear the 5-40 bolts and counterbored to clear their heads. After bolting the workpiece halves together, the assemblies were drilled and the roll pins pressed into position. Finally the workpieces are ready for machining. - Terry

 

[Charles Lamont]

On the matter of CNC machining and eating up aluminium stock, I often see on this forum what seems to me, in my ignorance, a surprisingly wasteful practice. It has puzzled me for some time. Why is it necessary to use a piece of stock so large that the part is milled down a pocket of waste material? Why not start with a piece just a bit bigger that the finished part?

Take, for example, those hose clamps. Now, I know they are tiny so the extra waste in that case is small, but it is a handy example to use for the purposes of my question. Why does each part need a wall of stock all round it, rather than pitching successive parts just a little more than a cutter diameter apart, so the milled pocket paths intersect?

 

[mayhugh1]

I agree it can be wasteful. It becomes a trade-off between material and time, and sometimes between material and machining precision. If you want to CNC machine a part with minimum hands-on and the fewest number of setups, you have to allow space around each part for the cutter as well as support framework that get's left. Big cutters need more space between parts, but bigger cutters can remove more material faster. Minimizing the number setups can reduce the errors involved in matching up surfaces cut in different setups which usually means file work.

The whole thing can be made more efficient if the dimensions of the starting workpieces can be optimally chosen for a particular nesting of parts. A commercial shop would do this and would likely order material that would give them the most efficient nesting for a particular part and the number of parts needed.

A lot of material is going to be wasted in the machining of these rods since I was stuck with the sub-optimal dimensions of the drops I had available, and even worse I could only fit two parts per workpiece. There'll be less material in the two parts than what's left behind in chips and support structure. - Terry

 

[gbritnell]

Charles, you mention wasted stock while making parts. It's just not the CNC process but also manual machining when it comes to making parts from a piece of metal as opposed to using a casting. And that's not the least of the wasted metal. After a large project like an engine I end up with a whole boxful of fixturing bits and pieces, ranging from large aluminum plates to special cutters made from drill rod. Most of the cutters are model specific and can't be used again. I have a whole drawer full of these.

 

[Charles Lamont]

George, I entirely agree with you, except that I don't regard even one-use tooling as 'waste'. I have a camshaft grinder that has, so far, only ground one camshaft.

However, that does not help with my specific question about the necessity for separated, fully surrounded pockets when milling a bunch 'nested' of parts.

 

[GreenTwin]

I can cast parts that are near net, but then there is the sprue, runner, gates, risers, spillage, slag, etc. that all are extra.
Luckily some of that extra material can be returned to the cruible.
For a full sized engine, I guess a full pocket may be a bit excessive.
For a model engine, I think a full pocket is not really too much, and would be very convenient to handle.

.

 

[kvom]

George, I entirely agree with you, except that I don't regard even one-use tooling as 'waste'. I have a camshaft grinder that has, so far, only ground one camshaft.

However, that does not help with my specific question about the necessity for separated, fully surrounded pockets when milling a bunch 'nested' of parts.

Parts that are profiled and do not require milling on both sides of the stock can be separated by the width of the profiling tool and be retained by stock at the bottom, which is then milled off in a separate op. Alternatively the stock is glued to a fixture block and separated later. Or if milling through the stock be retained with tabs on all sides that are to be filed or ground off. Clearance for vise jaws is also necessary.

 

[peterl95124]

have to chime in here, if my big end is X thickness and Y width I start with X by Y bar stock (I try to adjust my design such that X by Y is an available size, EG 3/8 x 3/4 or 1/2 x 1), drill the big and little end holes, and use them to secure to a piece that gets clamped in the vise while I machine the outside and the "I" beam pockets. very little waste, very nice looking rods.

of course that still does not excuse me for how I make my crankcases and oilpans, which usually start out as 5 or 10 pound blocks and end up as 5 or 10 ounce pieces ... :-( !!!

 

[mayhugh1]

have to chime in here, if my big end is X thickness and Y width I start with X by Y bar stock (I try to adjust my design such that X by Y is an available size, EG 3/8 x 3/4 or 1/2 x 1), drill the big and little end holes, and use them to secure to a piece that gets clamped in the vise while I machine the outside and the "I" beam pockets. very little waste, very nice looking rods.

Good idea!
Wish I'd thought of it ...

 

[peterl95124]

Good idea!
Wish I'd thought of it ...

I got it from Kozo Hiraoka's book "Pennsylvania A3 Switcher" !!!