Interesting scraping videos I found 2023

[rmcphearson]

I stumbled across these recently. Some good material for discussion.

Why were fuel pump sealing surfaces (and other sealing surfaces) scraped long after gaskets were available? What material is the fuel pump in that video? Did I read somewhere that some CI exhaust manifolds (GM?) did not have gaskets?

 

[Paolo_MD]

Just a couple of quick comments.
Scraping (with the goal of having ~80% contact surface) is done to guarantee perfect sealing where you cannot have gaskets for several reasons (e.g. high pressure steam flanges, certain gearboxes, etc.). Probably, that fuel pump has been designed in a way that tolerances were too tight in order to allow for the thickness variability of a gasket, or similar engineering issues (yes, I guess that an o-ring would be perfect there, however, you would need to machine a rather intricate groove to accommodate it, no CNC available back then, plus cheap labor).

Second video: "Good and fast enough for the job"
Alex is a brilliant engineer very capable of thinking outside the box. Yes, he goes against general wisdom and uses materials that are far from optimal. However, let's focus on the problem: he needed a good enough straightedge in the shortest possible time. Given that casting was out of question and that welding cast iron to steel would have caused more problems than solved, he came up with the best solution reasonable for his problem: a symmetrical design (I-beam), which, I believe, is the structure that maintain best it straightness with changes in temperature and with (a lot) of residual stresses.
He developed his technique for scraping steel, given that it is rarely practical, or even possible, to get something cast for prototyping. Therefore, willing or not, he has to deal with mild steel for his kind of work and he developed his methods accordingly.

As a side note regarding the use of "unconventional materials" for straight edges and other masters, Busch Precision has a series of tooling made with cast aluminum bodies to which they bond cast iron "wear plates".

Paolo

 

[sfriedberg]

Busch Precision has a series of tooling made with cast aluminum bodies to which they bond cast iron "wear plates"

I purchased one when the original Busch was liquidating all their inventory. 18"x36" square with 3" wide working surfaces. While it is no featherweight, I can at least lift it with one arm, which means I can actually use it where I would not be using a reference that requires a shop crane to move around. The aluminum casting is excellent and the bonding of the body to the cast iron wear surfaces looks impeccable. I keep it in a temperature controlled shop, so I can't say anything about temperature stability, which various people poo-pooed about the last time I saw these mentioned on PM. Since there is about 5/64" of cast iron bonded to 1.5" thick aluminum, and my working temperature does not vary, I am not too concerned about differential expansion.

 

[just Dave]

I ran into gasket less sealing surfaces on boiler feed pumps at power plants. Changeing mechanical seals involved disassembleing the plain bearings and holders. Oil bath through components and no gasket.

 

[TGTool]

Just a little trivia relating to the 1940's video. CNC toolpaths and O-rings really took care of the cover seal issue on that pump that scraping addressed, but you still have sealing between the gears (gear pump) and the housing.

From a former employer who built test stands for hydraulic pump and motor manufacturers I learned that they still fit the housings to gears the old way. That is, the housings are machined with a very small interference fit. The gears are pressed in and the pump assembled. On the test stand the first operation fills the pump and rotates it to shave that last little bit from the housing. Swarf is collected in a strainer and subsequent filters before oil goes back to the reservoir. That does ensure the closest fit from gear to housing.

The boss (chief engineer) also asked what I would expect about the pressures between gear teeth from the low to high pressure side. I expected incremental increases tooth-to-tooth space. Not so. Testing shows that as soon as a rotating tooth closes the space from the inlet side, it's already at outlet pressure.

 

[magisterFactotum]

A few other non-machine tool scraping jobs I've stumbled on:

Nothing ultraprecision here, but interesting to see where the old methods hang on.

 

[Richard King]

Just a little trivia relating to the 1940's video. CNC toolpaths and O-rings really took care of the cover seal issue on that pump that scraping addressed, but you still have sealing between the gears (gear pump) and the housing.

From a former employer who built test stands for hydraulic pump and motor manufacturers I learned that they still fit the housings to gears the old way. That is, the housings are machined with a very small interference fit. The gears are pressed in and the pump assembled. On the test stand the first operation fills the pump and rotates it to shave that last little bit from the housing. Swarf is collected in a strainer and subsequent filters before oil goes back to the reservoir. That does ensure the closest fit from gear to housing.

The boss (chief engineer) also asked what I would expect about the pressures between gear teeth from the low to high pressure side. I expected incremental increases tooth-to-tooth space. Not so. Testing shows that as soon as a rotating tooth closes the space from the inlet side, it's already at outlet pressure.

Kearney and Trecker mills had a gear pump like that and would wear out after time. We would rebuild them much like what you did. We ground the gears that had worn thin and ground the squash plates were scoured up where the gears spun against. At first we applied a gasket between the plates but ended up scraping the 4 sides - center section 2 side and plates that were on the outsides. We figered .0002" clearance. It was a trial and error fit. I now see how that's the way it worked TG. Thanks