Looking for help on early (1908) Curtiss Engine

[eweidenh]

Hello All,

 

I'm new here. I hope that this is a reasonable topic for this forum.

I'm working on a 3d-printable model of a classic aeroengine--basically an (intermittently) working prototype from the early history of aircraft. This is the Curtiss No. 3, which famously powered Silver Dart, Cygnet II, and various other experimental craft created of the Aerial Experiment Association (AEA) that included Alexander Graham Bell, Glenn Curtiss, and others. The engine first ran on 23 October, 1908 and was used by the AEA through at least February 1909. It was later used in a small fishing boat in Nova Scotia, which sank. The engine was later recovered and now sits in a display case at the Canada Aviation and Space Museum (CASM) in Ottawa.


The good people at CASM made a scan of it for me a month or two ago. I've been using that scan as the basis for a printable model.

Here's a rough 1/8 version FDM printed version in primer:



Here's a 1/25 version in UV-cured resin that I've painted up: 

 

 

Edited December 12, 2019 by eweidenh

[eweidenh]

The engine as it survives today is missing various parts, including spark plugs, electrical cables, cooling water tubing, and (I think) a gear-driven water pump between copper cooling jackets and radiator. I'd like to try to model a more complete version based on some archival research. This is challenging because:

1) The engine was modified pretty radically over its operational life. For instance, the engine that left the Curtiss Factory had one carburettor for each cylinder, one mechanical exhaust valve per cylinder (the intake valve was a suction valve), and the distributor located directly in front of the valve timing gear. The later engine had two carburettors (one per cylinder bank), two mechanical valves per cylinder, and the distributor moved to the opposite side of the engine. I'm more interested in this final configuration.
 

* These are details from Smithsonian National Air and Space Museum archival photos NASM-CW5G-0908 and NASM-CW5G-0917

https://airandspace.si.edu/collection-objects/aerial-experiment-association-aerodrome-no-4-silver-dart-mccurdy-john-d-glass

https://airandspace.si.edu/collection-objects/aerial-experiment-association-cygnet-ii-propulsion-engines-curtiss-v-8-silver

 

2) I haven't found a lot of photo documentation so far. 

 

3) My knowledge of piston engines (not to mention historical aircraft engines) is very limited.

I was hoping that some of you pros could help me with some technical questions, or provide suggestions about where to look for archival info. In any case, this topic should serve as a repository of info on this engine, which isn't particularly well studied. Here is how the engine was described in Scientific American on December 12, 1908 (p. 434). This refers to an earlier configuration as mentioned above.

 

The engine of the “Silver Dart” is a 3 ¾ [bore] by 4-inch [stroke] 8-cylinder, water-cooled, Curtiss motor, capable of developing 50 horse-power at 1,600 R. P. M. Including radiator, water, oil, etc. the weight of the complete power plant is 250 pounds, but the engine alone can be stripped to 165. The cylinders are fitted with copper water jackets and auxiliary exhaust ports. Circular concentric valves are located in the cylinder heads, the inlet valves being automatic. The crank-shaft is of specially treated vanadium steel. It is bored out hollow and is 1 ⅜ inches in diameter. The connecting rods are steel forgings, the cylinders and pistons being cast iron. The crank case is made of special aluminum alloy. The main bearings of the crankshaft are continuously flooded with a bath of oil by means of an oil pump of the gear type. Individual aluminum carbureters are employed with all eight cylinders.

 

Edited December 14, 2019 by eweidenh

[eweidenh]

Here is my first question: There appears to be an additional assembly that is missing from the motor as it survives today.

 

It looks to be gear drive--possibly an oil or water pump, though I'd assume that the oil pump would be inside the engine case.  It's evident in various photos as a smaller gear above the valve timing gear, roughly in line with the  lower portion of the carburettors. Any ideas or information on this? You can see it in the following images:

* Detail from Smithsonian digitized journal "Aeronautics" Sept 1909.

*Detail from Parkin 1964, p. 126.

 

* Detail from Smithsonian National Air and Space Museum archival photo NASM-CW5G-0911

https://airandspace.si.edu/collection-objects/aerial-experiment-association-aea-aerodrome-no-4-silver-dart-mccurdy-john-d

 

 

[eweidenh]

Second question:

 

There is a row of alternating elements on the top surface of the engine case that are mostly missing from the surviving engine. 

 

These appear to be threaded plugs alternating with some sort of adjustable opening. Any idea what these these are for?

* Detail from Smithsonian National Air and Space Museum archival photo NASM-CW5G-0917

https://airandspace.si.edu/collection-objects/aerial-experiment-association-cygnet-ii-propulsion-engines-curtiss-v-8-silver

* Detail from Smithsonian National Air and Space Museum archival photo NASM-CW5G-0903

https://airandspace.si.edu/collection-objects/aerial-experiment-association-aerodrome-no-4-silver-dart-propulsion-engines

[John Aero]

In post 3, the large gear wheel is, I think the Cam shaft drive for the valve timing. At the reverse end of the camshaft is what looks to me to be the Magneto with the insulated output in white ceramic. In the second photo down in Post 3 there are what I think are ignition leads so this might make it a combined Magneto/ Distributor. My guess on the items running along the flat top of the crank case might be adjustable oil pressure feeds for the cam followers or for such as the missing water pump or the upper cylinder mechanisms.

 

I've looked in my early 20th century books and I can find no reference to this engine which will help I'm afraid.

 

John

[Sabrejet]

I too have been looking at this and cannot find anything of use. However it looks to me that the extant drive gear (driven off the crank extension) is driving a second gear-driven water pump (just visible in the B&W photos). That would make sense because the prominent semi-circular pipe which extends around the top of the engine in that area would seem to be in the correct location to connect to the water pump, and thence to each cylinder's water jacket - which again it seems to do.

 

Again, guessing but I wonder if the threaded objects are lubricators for the camshaft bearings?

[Dave Swindell]

I'd concur with the above.

The cylindrical object in front of the large gear wheel is the magneto, connected to the spark plugs on the outboard lower side of the cylinder banks.

The large gear drives the cam shaft above the crannkshaft between the cylinder banks at half crank speed so it's a four stroke engine

The smaller gear above the large gear is driving a water pump connected to the jacket headers on the inboard top of the cylinder banks, this woud normally be the outlet and would be pumped to the radiator, return from the radiator goes to inlet headers on the outboard lower sides of the cylinder banks.

As there's no sign of an external oil pump I'd suggest this is internal in the sump and internally gear driven off one end of the crankshaft.

The knurled valves and pipework on top of the camshaft between the cylinder banks is for lubrication of the valve gear. There's one valve per bank, and the oil is fed from the camshaft bearing oil galleries through the restrictor valve to a pair (left & right) of cylinders. This is almost certainly a total loss system, hence the restrictor valves which could be set to provide the minimum oil required for lubrication.

Interestingly and unusually, this engine appears to have concentric air inlet and exhaust outlet valves.

 

[eweidenh]

Thank you all very much for your helpful replies! I'll keep digging around for better photos of the water pump. It's a mystery to me how it was mounted, given the surviving motor.

 

I have a further question about the firing order of the cylinders: Do you know whether that was standard at the time? You can deduce part of it as there are a couple of photos that show wires leading from the magneto to the plugs. Also, assuming that the camshaft is intact, I could measure the angle of the rocker arms in the scan.

[Dave Swindell]

4 hours ago, eweidenh said:

I have a further question about the firing order of the cylinders:

It's a 90 degree V8, firing order will be determinded by crankshaft geometry (which you can't see without dismantling the engine to some extent).

Positioning of the HT leads on the magneto and connection to cylinders will give a clue for each bank.

These pages may help further

https://www.roadandtrack.com/car-culture/a19744474/how-v8s-work-visual-guide/

https://www.enginelabs.com/engine-tech/engine/firing-order-swaps-whats-best-for-your-engine/

[jeaton01]

I would think working back from the OX-5 would be a good starting point for ignition.  Here is what is in the The Curtiss Standard Model OX Aeronautical Motor Hand Book, which can be found for free on google books.  I have added the firing orders which I determined by examining the diagrams.  I think it is correct.  Note that the engine could run in either direction.  Apparently the starting point for all of the Curtiss engines was the V-twin motorcycle engine, and 2, 4, 6, and 8 cylinder engines were developed at first by bolting them together.

 

 

 

 

 

[eweidenh]

Thanks very much! I was looking at the OX-5 and the V-4 while trying to figure out the gear-driven water pump. I didn't run into the handbook, which will be useful. 

 

I should get some detailed archival images in January from the CASM/ Ingenium archive in Ottawa. Hopefully those photos will show more detail. 

 

 

[eweidenh]

Thanks again to all who have helped so far. I have received a couple more archival photographs since I last posted. None offer a clear view of the entire engine.

 

I did encounter this photo of the drive shaft, which supplements jeaton01’s firing order diagram posted above. This is from the Bulletin's of AEA digitized by the Smithsonian (university library subscription only). 

....

The gear-driven water pump on the non-drive side is especially unclear. Here are the details from the various images of the engines in (or near) its relevant configuration that I've found so far:

....

I don't know much about centrifugal water pumps. This images below show my best guess for those missing elements so far. I’d appreciate any advice that you might have.

Here’s a design and test print of the centrifugal pump recreation, which is based on the images below it. Keep in mind that the design is pretty coarse because it is designed to be FDM printed at 1/8 scale.

Below is a rough FDM test print with the recreated elements painted green:

 

Here's what I think is possible to know based on the information from the archival images above. This is a centrifugal water pump with two delivery pipes, likely emerging from a single point in the volute. Curtiss would later a patented a similar-looking water pump, but that had two separate intakes and chambers so that the pathways between the two banks were further separated.

 

• The water pump is driven by an open gear (~60 to 65 teeth) running off the timing gear (90 teeth). It is centred on the crankshaft. We can guess at its size and tooth count because it comes to approximately 2/3 the height of the carburettors.
• The water pump brings cooled water from the radiator through a hose attached to an upward-facing opening and forces it down two curving channels leading to the lower water manifolds of the cylinder banks. The height of these curving channels delimits the circumference to the impeller housing the pump impeller, at lease the top side of it.
• Cooling water from these channels enters two manifolds, one on each lower side of the cylinder banks. It is forced upwards through the copper cylinder jackets and exits through the upper drive side manifolds which are joined by a hose at the drive side. This brings hot water to the radiator.
• An earlier configuration placed the pump to one side of the engine, in front of one of the cylinder banks. I don’t know why this was abandoned. Possibly the different lengths of the two water channels created differences in flow to the cylinder banks. It’s also possible that the need to install clamps to prevent the solder joints around the cooling jackets from failing took over the place used by the pump.  In any case, this gives us a maximum depth for the pump (assuming that the pump wasn’t changed) since too long a pump body would have intersected the front cylinder jacket on that side:
• 

 

Here’s what I think is a plausible configuration:

• Given the location of the centre of the gear driving the pump impeller relative to the curving channels leading from the pump to the lower manifolds, it seems likely that they join the pump at the top.  The bends leading to these curving channels are probably sharp given the contiguous appearance of the arc of the two channels in some photos. 

 

The pump probably resembles various centrifugal water pumps designed by Charles B. Kirkham of the Curtiss Airplane and Motor Corporation. Here are some of those:

 

The Smithsonian has a water pump from an OX-5, which you can see here:

https://airandspace.si.edu/collection-objects/water-pump-curtiss-ox-5-v-8-engine/nasm_A19660385002

 

This is Kirkham's patented two-chamber pump. The images shows sharper elbow bends than the OX-5 pump. 

 

A couple of other things:

- It's likely that this pump was made by another manufacturer, possibly an automotive company. For instance the concentric valve cylinder heads used in this engine were made by the Franklin motor company. I haven't yet looked into this possibility.

 

- Probably the best image of the pumps comes from an early configuration of the engine when it was being modified at the Curtiss workshop. I don't really understand what's going on in this image. In particular, I don't understand this element highlighted in green. I don't think that it's related to the water cooling system, but I could be wrong. Maybe it's a piece of instrumentation for testing purposes? 

Electrical system is next...

  

 

Edited April 1, 2020 by eweidenh