Questions about Making Metal Flasks

Looks good Melter, I take back what I said about uplift. You casting doesn't have anywhere near the surface area I guessed at. Especially since you have already done one.
Just picking nits here, but my "Foundations of Mechanical Accuracy" suggests that style had sensitivity to temperature variation and were not self proving. What level of accuracy are you hoping to achieve?
I agree with Rocketman's description of the slotted locator ears on one dowel. Some combination slotting of your four will still locate, but make separation easier.

 

Thank you for your interesting comments. That style of SE is sensitive to heat compared to what style? Are we comparing to his very complex (but rarely used) four-sided SE? And the concept of a “self-proving” straight edge is new to me. Please fill me in. That termed used regarding squares is old hat, but new to me concerning SE’s. Perhaps you are speaking about the need for 3 square or round plates to develop master flats to prove them against each other due to the need for rotation as part of the self-proving process?

With respect to the level of accuracy hoped for: my personal straight edge of this design is scraped to 40 points per inch against a granite plate that is AA accuracy. 5 others are in circulation. 3 are in use by professional scrapers and two by enthusiasts. So far, no complaints about inability to scrape them in to whatever degree of precision/accuracy desired.

With respect to heat sensitivity. Yes, lighter camelback SE will be a little more easily affected by temperature. But the user had better be very careful concerning heat with any SE or trouble is certain. The usability of a light weight SE comes into play quite obviously the fiftieth time that day you lifted it onto your work from the surface plate. This is especially true if the way you are scraping is vertical. So, horse for courses. A very lightweight SE has a definite place. I have people waiting anxiously for me to get some made so they can purchase them.

I like the suggestions made concerning guide pins/vees and intend to make the appropriate modifications.

 

I took the general term “aging” as I have usually heard and used it in the past to mean simple ambient temperature aging in distinction to high temperature accelerated aging most commonly referred to as heat treatment or furnace stress relief. No doubt the amount of sand, type of sand and binder, flask dimensions and flask material would all affect the degree of residual stress relief achieved by cooling a casting in the sand. But my point was that taking care to allow initial cooling slowly in the sand will always be much better than early shakeout.

So far, re-stress relieving has not been required or opted for by purchasers of the SE following milling. I have taken considerable care to minimize stress induction and and have developed a proprietary method to reduce twist in machining.

 

As soon as you showed your SE I looked it up and those were the first labels for that shape. Then Moore proceeded to explain the four sided SE with the comment that it was self proving. I take it that he meant it could be rotated 180 and checked on a surface plate since the opposite edges were parallel. He bemoans the fact that the straightness can be no better than the 25 millionths of an inch flatness of the master surface plate. I'm just a beginner here, but appreciate what you are doing. I like to work to .0001's , but have the best luck with that if I change the sketch measurement to reflect that level of accuracy after I make the part .

 

Moore's four-sided SE's are indeed a maximally rigid and stable design engineered to the nth degree. But, I do not think they offer any special self-proving qualities. You can see that if you had two and checked them against each other, they could show a perfect print of indicating ink, but they could still have minute corrugation with the "ridges"parallel to the long axis and spaced,say, 1/2 wide each, you could not detect that. That is why he explained you need three surface plates that could be rotated 90 degrees to each other and checked in all relative rotations to each other. And, though his SE design is elegant, it is not practical for day-to-day use in machine scraping and reconditioning. I know of no one who uses one with nearly everyone using camelbacks and a few people using (heavier) parallel designs for actual way checking and reconditioning.

Anyhow, I think mine is flat to within a tenth or two (I don't recall the specification right now for AA) or so along its 36" length. That is far better than I need for any practical purpose. It has been quite stable over the two or so years since I initially scraped it in. It has not been unduly temperature sensitive.

 

I would appreciate pics if you can supply them. Sorry I did not respond earlier today.

I have only two pins on my setup with one longer than the other so that the short one just engages as the mold closes and the long one only engaging for a half inch of travel or so. But, you are right that if the cope were guided in x and y for a little more than the depth of the pattern cope depth, there would be less chance of spoiling the cope sand. I think I have a pretty good idea what you have done, but seeing would be good. Also seeing others' solutions might be helpful.

 

Did you consider box tubing, stress relieved then split for you flask members?

Beautiful design!

 

No, I didn’t. 4 x 2 by .187 would do it. But how to split it economically? Maybe bandsaw it? And how to stress relieve it? I presently have to take my straight edges a hundred miles for heat treat. If unheat treated, would the spring crazily when sawn lengthwise. I have sawn a lot of square tube to length but never split it lengthwise more than a foot or so for some special purpose. That might have been a cheaper way to go.

 

Dull red and a slow cool would pull the forming stresses out of sheet metal. Seems like you could build a loose brick furnace and heat treat at home. Controllers are pretty cheap theses days. I use an electric kiln with a cheap controller ($35 or so) to cure small refractory parts on a 30 hour cycle. You could probably pay for heat treat furnace materials for the cost and time to go 100 miles and back to pick castings up.

Stitch cut with a plasma like you do beams then cut the ties and you wouldn't get much warpage, or stitch cut then anneal then separate.

 

I think he referred to it as furnace stress relief but I didn't grasp the reference that seems to indicate he does not need to do that.

 

I am talking about furnace stress relief of grey iron for my straight edges. I presently am using the services of a commercial foundry for that process though I do have in mind in the not distant future buying a used pickup toolbox and modifying it a bit and lining it with insulation and reisitance wire to make a heat treat oven. Just have other things more immediate to do now.

I think a trial of split box tubing for future flasks would be a good idea. I think I will split a foot or so today just to see how it behaves. From prior times when casually cutting short sections I don’t recall remarkable warpage. And having the pieces of plate bent like I did was not the perfect solution either as that caused modest (maybe 3/8” in 4 feet or so) warping on the long axis that had to be dealt with.

 

Regarding splitting square tubing: For my forklift project I cut a 4" x 4" x 1/4" wall square tube in half lengthwise. It had significant bowing from internal stresses. I would estimate about 1-2" over 8'. I was able to use it by placing cross braces every 2 feet.
Robert

 

I followed your advice and lengthened my pins and “ovalized” the short pin hole. I milled the 3/8” round hole by extending it left and right by50 thou for a total lengthening of the hole by .100. I also used a file to generously champher both holes to reduce their tendency to bind even if the pin is not quite vertical as the draw is made. Both the locating round hole and the oval holes are just a few thou over 3/8 in smalles dimension so they still do a pretty good job of minimizing registration errors.

And to make an already complicated hoist setup a little more complicated, I am going to place a little strut from the top of the spreader bar to the middle of one of the long sides of the flask. That will prevent rocking of the flask around its long axis as the draw is made. I need this flask to come straight up and off.

Here are couple pics of the lengthened short pin with its ovalized hole.

 

That is quite a bit of bowing. Good to know. Thanks.

 

How was it cut? One torch cut, segmented torch cuts, sawn?

 

Heres some shots of a commercial flask I own plus one of my home made jobs.

 

Thank you Rocketman for the pics!

I mentioned earlier in this thread that I intended to make a stabilizing strut that would connect the long side of my cope flask to the lifting spreader bar. My hope is that this will keep the cope lift nice and straight by preventing rocking of the cope as it is drawn off the pattern. The complexity of this pattern makes a clean lift essential to avoid spoiling the print and the size of the mold makes fine control of the lift challenging for me. So here is the addition that I am anxious to test out. (It would be possible to also attach the strut to the follow board as I lift it off the drag before ramming the cope. So far, that has not been needed as that lift is pretty easy to control.)



The sleeve shown below allows for length adjustment of the strut.



This picture illustrates the full scope of my blacksmithing skills! ;-)


Here is the upper end of the strut. I put a tab on both sides of the spreader bar so that there the strut could be attached no matter which way the spreader was oriented. Probably not needed. But with the TIG already setup and realizing my ability to get things backwards....

 

Just took the flasks and spreader bar out to the barn. It LOOKS like this is going to really help control cope draws. Nothing to do now but pound sand tomorrow the the weather may not cooperate.