Duesenberg Straight 4, quarter scale

[peterl95124]

I started this Duesenberg Straight 4 too long ago for this to be a complete build blog, but am hoping this inspires some folks to ask for my plans, of what I consider to be both a beautiful and historically significant engine. The original is of course the "Straight 8" but I decided that would be too large for my bookshelf and didn't really want to make 32 valves, 32 valve guides, 32 cam follower buckets and 32 follower guides, so cut it down to a "Straight 4", with the hope that it will still be recognizable as the classic Duesenberg (especially if painted bright green with polished aluminum covers as it seems all full size engines are). I think it would be easy for someone to use my plans and build either a straight-4, straight-6, or straight-8. The design and plans are still a work-in-progress so hold off asking for them until they're ready.

I contacted Lou Chenot about plans for his 1:6 scale Duesenberg engine, he said he didn't keep any, but gave me his "factory drawings" on a CD, and said he thought they belonged in the public domain so I'll make them available for anyone that asks (I've already uploaded some of the drawings to BAEMClub website). In any event 1:6 scale was way too small for me to to 4-valves-per-cylinder, so I've gone with 1:4 scale derived from the factory drawings. The factory drawings are incomplete as in missing a couple pages, and show what machining operations are to be done on castings but are only approximate about the castings themselves so some interpretation from photographs of full size engines was required.

 

[peterl95124]

Here are the custom 35-degree Vee-blocks in action for machining the bores for the combined valve guide and seats into the head

 

[ddmckee54]

You don't happen to have any more pictures of machining the head do you? From what I can see in the photo above, that was no easy task.

Don

 

[peterl95124]

You don't happen to have any more pictures of machining the head do you? From what I can see in the photo above, that was no easy task.

Don

I was able to dig up two more old photos.

the Vee-blocks themselves were a challenge, its not that they need to be exactly 35-degs, they don't, but they need to be perfectly parallel, so I messed around for a long time with a dial-indicator in the drill chuck moving along the block to make sure there were no height or depth variations across the lengths of both exposed surfaces of the block before milling into it.

the real problem with 35-deg angled valves is getting things set up accurately, I scribed a line down the center of the head and then when it was in the Vee-blocks I got the mill accurately over that line using a pointed wiggler and a 10x magnifying loupe, then with some trigonometry I could calculate how far to move the mill to get to where the valve should be. this setup didn't turn out to be very repeatable, when I had to do subsequence tweaking and/or re-design machining then I'd use a dial-indicator in an existing valve bore to zero the DRO, but because I took so much time to make the Vee-blocks parallel I could run to the other end of the head and I'd be zero'ed right where the DRO said I should be.

here's using the Vee-blocks to machine the inside of the combustion chambers and bores for valve cages (you can see the scribe line I had created before hand). also if you look closely you'll notice the V-block in the back of the mill vise has been bolted to the fixed part in place of the normal vise jaw, in the previous photo I was lazy and C-clamped it in place)

here's using the Vee-blocks to machine the top side where cams and cam followers will go (it too has a scribe line down its middle)

the scribed center line is just one way to do the zero-ing, I might have been able to locate over an edge except this was extruded aluminum and didn't have a sharp edge, also if I knew the exact depth of the cut in the v-block I could have used some trig and measured from the side of the block, but the height and depth was only approximate when I made it (though I could get pretty accurate with a dowel-pin and a depth mic).

anyway, you should not let this scare you if you've built any engines before, any "hemi", the Offy engines, Ferrari V12, and now most modern engines, all have angled valves.

ok, enough rambling for now...

 

[peterl95124]

I have so far planned on using "XL" type cog belts and sprockets for timing,
I even already bought a set of sprockets for it (from stock-drive-products),
the full size has what appears from a distance to be roller chains and sprockets,
but I recently stumbled across a close up (below), and after much googling
think I can say these are "inverted tooth" (sometimes called "silent") chains,

it appears that there's only one manufacturer that has these inverted tooth chains
down to 3/16" pitch (a bit over scale, still I could live with it), but only one retailer
and for 10-ft of 3/16-pitch its $220 for 2-wide and $275 for 3-wide.

regular roller chains down to 3.75mm (.1475") pitch are much more widely available,
but only in 1-wide, I'd have to disassemble and reassemble with new pins to make
my own 2-wide and 3-wide, which would take about 110 pins each for two chains,
that would be quite a chore to say the least.

so for now I'm sticking with cog belts and sprockets, with the option of upgrading
in the future.


if you look closely in the photo below you can see the chain has teeth that fit in between
the teeth of the sprocket, in this photo it is clearest for the topmost sprocket.

 

[peterl95124]

valves are now all slotted for lapping
you can see the jig to hold a valve in the vise, and the slotting end mill,
which was a "#202.5" (1/16 thick x 5/16 diameter) Woodruff Key cutter
until I ground it down to .030 thick in the lathe using a dremel grinder,
and at the bottom you can see the custom round-nose screw driver for
doing the lapping (thanks to Dwight Giles for that), the second hole in
the jig is for my Cirrus valves with different stem diameter.

 

[peterl95124]

now on to lapping the valves in, here's how I do it

 

[petertha]

@peterl95124 I'm assuming there must be a check valve or maybe its a tire inflation stem so that it maintains pressure when you remove the fill hose? Can you elaborate.

 

[peterl95124]

@peterl95124 I'm assuming there must be a check valve or maybe its a tire inflation stem so that it maintains pressure when you remove the fill hose? Can you elaborate.

Petertha, yes, that's a standard Schrader valve from a bicycle inner tube, hence the tire fill fitting on the end of my compressor line. that bicycle valve fit into a 5/16" copper tube so its loctite'd in, but not until after silver-brazing on the taper-pipe-threaded cap for the pressure gauge.

I thought long and hard about how to capture the valve cage in the tester, I wanted something that was twist on because when you've got say a V-12's worth of 4-valves-per-cylinder you want it to be fast and convenient and finally came up with what you see in the video.

 

[petertha]

Thanks. I tested my valve/cage seal assembly with vacuum. But I want to make an elevated pressure (leak down) tester like yours for the completed cylinder assembly.
Actually I've never quite understood if the 2 methods were excusive or somehow tested different aspects of seal? Vacuum pulls from the underside to say max 14 psi. Positive pressure on the valve chamber side could easily be +3X that like your example off a compressor tank.

 

[peterl95124]

pressure and vacuum are just two sides of the same coin, the only difference is as you noted vacuum is limited to 14 psi, while 1000s of psi are possible with pressure.

FYI, after everything is assembled in the head its hard to tell which valve is leaking, in that case I put pressure into the spark-plug hole and use a stethoscope in the ports to identify the leaky one (finally settled on this after trying nearly every other imaginable way).

 

[petertha]

Thanks. Thanks that corroborates my thinking. I have vacuum tested my valves & cages prior to head installation & then again through the backside ports once cages were installed. It was recommended to me by someone before initial run to spot check the head assembly with positive pressure via a pressure fitting replacing the glow plug. I guess everything comes into play now, valve seats, ring fit, ring gap. I guess if one is rapidly falling off & others are fine, you know something. From what I've seen of the auto style, yes they go on an audible 'hisss' finding mission, or maybe soap bubbles is some exposed surfaces. I do have a model engine kicking around I could spot check, I've never actually done this kind of measurement on them though.

 

[peterl95124]

the trickiest part of making split bearings is doing the splitting,
here's what I've found that works for me,

note the sacrificial aluminum blocks taking the place of the hardened vise jaws,
note the squirt bottle of cutting fluid, without which a slitting saw always overheats.

I cut from both sides, and have never had the cuts match up in the middle, but
the halves still fit together, and the middle gets bored out anyway, so after this
its off to the 4-jaw in the lathe...

 

[peterl95124]

when you're machining a split bearing without a flange then the stock can be pretty small compared to the hole you're about to bore in it, in that case to avoid the metal getting bent by the force of the drill you can put a clamp on it, a lathe dog works just fine

 

[peterl95124]

and here are the bearings for the Duesy camshafts

 

[peterl95124]

The aux shaft housing that contains helical gears to drive the fuel and oil pumps is very distinctive on the Duesy, here's a shot of some rotary-table work to get that shape as an integral part of the crankcase (the rod is part of the tailstock, not part of the engine, its to keep the work from vibrating while the rotary table rotates during cutting)

 

[raveney]

The aux shaft housing that contains helical gears to drive the fuel and oil pumps is very distinctive on the Duesy, here's a shot of some rotary-table work to get that shape as an integral part of the crankcase (the rod is part of the tailstock, not part of the engine, its to keep the work from vibrating while the rotary table rotates during cutting)

Peter,
How is the crankcase secured to the table of the rotary table? Is there a dead center or clamps hidden from view?
...I really appreciated your video on valve lapping and testing by the way...

 

[peterl95124]

Peter,
How is the crankcase secured to the table of the rotary table? Is there a dead center or clamps hidden from view?
...I really appreciated your video on valve lapping and testing by the way...

there's an aux shaft hole on each side of the case, this one for the oil pump drive and on the other side for the water pump drive, through the water pump drive shaft hole there's a clamping stud secured to a Tee nut in the table, and in the oil pump shaft hole there's a morse taper #2 centering plug that matches the hole size.

one thing you'll want to get first thing after a rotary table is a couple morse taper collets that you can insert into the table and into which you can insert drill rods for centering parts on the table, in my case I made some #2 centering plugs from scratch, and then in the end finally got smart and bought a collet as you can see here, the only disadvantage of the collet is it sticks up above the table which isn't always acceptable, so I still use the home made centers a lot.

good to know the video didn't put you to sleep, I'm aware that I'm not the best talker/speaker...

 

[peterl95124]

its been a busy month making a plethora of shafts, bearings, lock nuts, and eccentrics for the Duesy's timing gears. I'm using type "XL" timing belts and pulleys because I'm not able to source metal chain small enough for this scale, oh well,

the double set of holes for one of the aux shaft bearings is because when I finally got around to sourcing 45-deg helical gears to drive the oil pump I couldn't find any that were scale and had to move the shaft outboard another 1/8" inch to fit what was available, oops !

no, I don't think the engine will run with rubber bands !, but the belts are on back order from "Stock" Drive Products (hoping they don't need to rename and rebrand :-( ! the problem is that 63-teeth is in the catalog but in reality multiples of 5 are what's actually available, I might round up to 65 and make a larger diam tensioner pulley to get on with the project).

the black magic marker is what the eventual outline of the front of the engine will be, after I no longer need flat surfaces to clamp in the mill vise.

 

[peterl95124]

when making an oil pan I always hog out the inside first, and be sure to leave a straight vertical interior rim around the inside of the flange in order to be able to put an adjustable parallel in there for clamping when switching to doing the outside of the pan, this way even with only 3/32" of flange in the vise it can still be clamped down with a lot of force as the adjustable parallel inside the pan keeps it from crushing

hopefully the photos make clear what probably isn't obvious verbally,

the first shows the adjustable parallel inside the pan after it has been hogged out, the second shows how the pan can then be clamped to machine the outside since the adj-parallel allows a lot of clamping force to be used, finally the nearly finished pan which still needs some corner rounding on the outside.

the big round cutout on the inside is scale, that's where the oil pump goes, even in the full size engine the vertical shaft that drives it is so off center that this is necessary to enclose it, and there's a corresponding bulge on the outside.

the black magic marker is where the front timing gears cover goes, but photos of Joel Nystrom's full size rebuild indicate that the pan is not blended to that shape, it stays rectangular up front. very thankful such photos exist, the internet can be a wonderful thing.