Marine Twin Cylinder Build (Steamech)

[ths]

Hi Kim,

Just found this thread, and I'm enjoying it greatly. Making the cylinders seemed like fun, and the bedplate was an exercise in remarkable concentration. Therefore I'm remarking on it. Keep going, I think you have quite a talent!

Cheers, Hugh.

[Bogstandard]

Kim,

I do have DRO's on my machines, and use coordinate milling and drilling all the time, but like yourself, I still usually mark out the material, just as a safety factor.

A belt and braces technique that anyone reading this should start to do if you don't already do it.
You will find that you won't have so many failures.

Very nicely executed.


John

[Kim]

Thanks Jo, Hugh, and John for the encouragement!

I'm needing it right about now!  I'd started over on the Cylinder mounting plate yesterday (which I haven't really posted about yet) and had  caught up to where I was when I mucked up the other one.  Then I proceed to mess up the new one even worse!!     I didn't have the part clamped down tight enough.  It jiggled around on the rotary table as I was milling and suddenly I realized I wasn't making a circle anymore, I was making an ellipse...    Its just 1/8" Al, I thought one clamp would do it.  But I guess it didn't.  Or it wasn't tightened enough or something (but I've ruined those feeble Sherline T-nuts by over-tightening before, so I'm trying to be careful... guess I was too careful...)

Anyway, it was time for me to come in and take a break.  It's one more do-over for me on this one.  Not an expensive hunk of metal, but it feels like I just wasted a lot of work    I know - what doesn't kill you just makes you stronger...

So I'm writing up my next build log post.  Give me something different to do for a bit...

Kim

[Kim]

The Cylinder Plate is what the two cylinders will mount to.  It's another cool looking curvy piece.  I thought it would be pretty easy compared to the Bed Plate because it's half the thickness (1/8") and has fewer holes you have to mill out.

I started with a 1 3/4" x 2" x 1/8" piece of 6061 bar stock, squared it up on the mill, then blued it and laid out the important parts on it.


Next I got my dimension chart all ready, zeroed it all up in the mill and spotted & drilled all the holes.


Here I'm drilling out the 1/2" hole for the cylinder.  Turns out my 1/2" drill cuts a fairly accurate 0.500" hole.  So since this isn't really part of the cylinder, I'm going ahead and just drilling it.  Seems like that should be OK.


All the holes drilled.  The easy part is done. (The orange arcs and numbers are explained below.)


I spend the next several hours of my hobby time calculating where the two diameters intercept each other so I know the length of arc's I need to sweep for the big circle, which go around the cylinder holes, and the little circles, which go around the support posts.  This took a lot more trig that I had originally anticipated.  I even resorted to using a spreadsheet to help me figure it out.  But eventually, I got there and I had great numbers to cut all the arcs.

I had to do the big circles in 3 pieces each.  I could do parts 1 and 2 in one setup, then rearrange the clamps and realign for part 3.  The three arc sections are shown in the picture above with the orange numbers.

To clamp the part in place on the rotary table I made a little brass bushing that fit in the 1/2" hole. This slipped over a 3/8" bolt screwed into the center or the rotary table.  The threads in the Sherline rotary table are 3/8"-16, but standard bolts seem to fit somewhat loose in it.  Regardless, when I tightened things up, it seemed to pull to center well and hold things nicely.  I added an extra clamp just for insurance.  This setup made it very easy to center the part on the rotary table.  All that was left of setup was to get a reference angle on the rotary table so I had a point of reference for all the arcs.

After all this setup, I followed my careful calculations, and cut arc number 1.  Cool!


Then, without a change in setup, I raised the spindle, rotated over to the starting place for arc number 2, brought it down there and swept that arc.  Super Cool!


While I had that all in my mind, I flipped the plate around and did arc 1 & 2 on the opposite corner (don't you love symetry? )  And this is what I had at that point.  You can also see the nifty brass bushing I made for this operation still in the hole on the right.


The main point of this picture is to show you all my carefully calculations.  Yeah, a little too proud of those calculations isn't he?


And pride cometh before the fall...

I rearranged the clamps, re-zeroed the table and cut the final arc, arc number 3, on one side.  Then I noticed it... All my careful calculations and I was off by two degrees in my arc sweeps.  It wasn't my trig that was faulty, it was just simple addition.  I had to add the arc sweeps to the actual starting angles when I got the rotary table all zeroed in.  And I was just off by two degrees.  Go figure.

Here's a close up of what I noticed.  I made red and greed arrows to highlight the issue; the green arrows show the correct width, the red arrows show the bad part.


I thought about ignoring it - trying to blend it in with a little fudging here, a little filing there... But the difference was big enough that I could see it with my naked eye.  And these are going to be concentric around the support posts, which would highlight this discrepancy even more.  I decided it was just too far off to pass my QC department.

So, back for one more try.

I spent all day yesterday (well all day I had in the shop anyway) getting back up to the same point.  I was feeling pretty good about it.  And I was ready to cut the arcs around the 4 outcroppings.  It was a similar exercise - center the hole on the rotary table, clamp it down and find the horizontal reference angle.  Then start milling the radius.
 
Oops....  About 2/3 the way around my radius, my turn table stops rotating.... What?... Oh... My clamping setup apparently hits the dovetail for the mill headstock.  Bummer...


It just barely clips it, but there's no squeaking by.  So I re-do my hold-downs, re-align the part & table, and start the arc again.   (As an aside, I realized later that I could have just pulled my end mill out of the collet another 1/8" and it would have let me raise the headstock 1/8" and solved my problem with no re-aligning required.  But I wasn't that smart at the time.  )

This time, just about the same place I start to notice that my arc isn't being round anymore - there had been a bit of vibration just before this - I think because the mill bit was cutting into my sacrificial base plate that I'd been milling on.  That had been happening all along and didn't seem like a big deal.  But apparently, my current clamping arrangement wasn't sufficient for this and the part moved a bit. And messed up my radius cut.

Here's what my new part looks like now.  Note the top right protrusion - it starts with a nice radius, then WHONK... not so nice anymore.


So that part goes in the scrap bin and I start looking at my OLD bad part. Somehow, it doesn't look so bad now.  Maybe I can fudge it...  I'll just give it a try...

So, I setup again for the same radius. But I'm smarter this time and I use TWO clamps, and I make them nice and low so I don't hit the back of the mill head dovetail.

And this time, it turns out beautifully!


I repeat this process a couple more times.  Now I've got three of the four done:


And that's where it sits.  Because now I look at it and it doesn't pass muster.  I've fudged as much as I feel I can get away with, but there are issues I just don't think I can look past.  The lower left protrusion is the worst.  I've already cut a smaller radius to try and get it 'looking round', but I just can't get it smaller enough.  And to top it all off, the larger circular arc that is going to follow the cylinder shape has a nic taken out of it (lower left again, inside the protrusion) where I was trying to hurry and cranked the turntable too far and cut into the other side.

So, yeah, I'm disappointed  . But writing about it for the last hour has proven therapeutic and I'm feeling much more philosophical about it all now.

What's my rush?   If I finished this engine, I'd just pick something else to do, right?  So might as well enjoy myself with this one and do it so I'm happy with the end result.

So, off to bed, and next time I'll start at the beginning of this post and work my way through. And I'll be a little smarter, and a little more careful!

Thanks for listening! 
Kim

[ths]

The only reason it doesn't pass muster is because you've got the plans to compare it to. Don't bin it yet, even when you get a passable one, you might get all sentimental about it, and decide that the original one actually has some pretty good curves on it. Keep going, all's fine, just annoying sometimes.

Hugh.

[steamer]

Kim

Striving for perfection is a wonderful virtue...but take heart in what Hugh said...could the radius be smaller?....I'd take 15 minutes and think about that before cutting the stock for another one...just sayin

I commend you on the blow by blow post....that took heart and integrity...you can't buy that trait...and you can't train it.

 

Dave

[zeeprogrammer]

Wonderful post Kim.
I think it describes well the trials and feelings that we all have gone through and for many of us, still go through.
You got much closer than I would have.
Calculations a bit off, rotating the RT only to discover later something is in the way, vibration. I could have written that post several times.

Whether you cut another or make do with what you have...the next project will benefit from your learnings.

[Ramon Wilson]

Hi Kim, I've been watching with interest for a while now and enjoying your post.

If I may comment on your current malaise - as I think I know where you're at with not being satisfied 

Just a couple of thoughts....

Matching up radii on differing setups on a R/T is not easy at the best of times so with what you're are attempting is challenging to say the least. Firstly, given the size of your table, I think I'd be inclined to make some smaller clamps. These are actually from a very old Unimat but are used virtually all the time in preference to others - not that I have many others as these seem to do 95% of whats required ....

The R/T is 6" diameter and the clamps are about 30mm long by 18 wide, 5 or so thick. Bolts are 6mm capheads - I use nothing larger.

Secondly why not make up two filing buttons to fit the center bore and a couple more for the lugs. Mill to within say .2mm and then file to finish. If you don't have any silver steel to make the larger buttons mild steel would do but ink them up first to give an indication when you get close. You should be able to fair the inner rad to the outers with ease.

I can't see any evidence of it in your pic but do you mark your limit of travel on your R/T with felt pen? A tip I have tried is to make two marks - the first a few degrees before the final position which can be a big help in preventing overshoot.

Hope I'm not teaching granny here - good luck with the rest of the build

Regards - Ramon

[Kim]

Thanks everyone for commiserating with me on this!  Sometimes, that's all you need...

Dave, Hugh, Zee,
Its not in the scrap bin yet, and who knows, maybe it can redeem itself yet.  But it'll be a challenge!  Its like an ex-girlfriend.  After you've broken up you start to magnify all the things that bothered you about her.   And that's where I am with this part.  There are several little things about it that I'm not satisfied with.  And it just feels too early in this build to compromise so much.  Maybe these are just my own mental games, but I've gotta play them .

Ramon,
I like that "Teaching granny" - is that more like "preaching to the choir" or "casting pearls before swine"?   I always love a good new phrase! (I just want to be able to use it right!) .

Good idea on the smaller clamps.  I may do that.  Although I worked out a way to use those bigger ones, they sure get in the way of seeing anything, which is important to a non-DRO, manual mill guy like me.  And the marks on the RT are a good idea.  I'll have to pick that up and add it to my bag-o-tricks!

As for filing buttons - Hmm... Yeah, that may be what I do, but somehow I really wanna get this to work.  It ought to!  Dog-on-it!  I've got a new toy and I wanna use it!   We'll see.  Maybe I'll do the filing if I can't get it in another go.

Thanks for sharing your advise everyone!
Kim

[Ramon Wilson]

Kim - the full phrase is 'Teaching Granny to suck eggs' and is usually used if someone is preaching to those already well conversant with whatever's being preached  .

I don't think you could ever find use for 'Casting pearls' on here though

I admire your persistence to do it on the R/T and get it right - you might find that inking the radius previously cut will help see things more clearly as you approach it....



Ramon

[Bearcar1]

Greetings Kim, to go along with some of the suggestions that have already been made, may I add one more to the mix. It seems that most of the problem stems from your clamps either slipping or getting in the way of things. Maybe you could consider attaching the part to a sacrificial plate that the clamps could bear upon and thereby not get in the way. Also. if some forethought is given, the plate could be made with a locator stud(s) that could be used to get the symetry when the piece is changed end-for-end. Filing buttons are our friend as well. I'm enjoying watching your progress and know hat you will conquer the beast.


regards


BC1
Jim

[mklotz]

Given that the area to be rounded has a central hole, a rounding jig may be simpler than the RT and quicker than filing buttons...

http://www.modelenginemaker.com/index.php/topic,352.new.html#new

[Kim]

Thank you all for the excellent advise.  For my next time around I tried to use as much of this wisdom as I could cram in! And I think it worked

Before my third attempt at the Cylinder Plate,  I took a little break and made a couple of small clamps similar to what Ramon showed.  It was a fun little side project!  It only took me an evening to do. Unfortunately, I didn't have very many evenings out in the shop this week.

Just for fun, I took some pictures of the clamp construction.  I started by cutting some 1/4"x2" blanks of random hardware store steel bar, drilled a row of holes and connected them with a 1/4" end mill.


Then I trimmed off one end at an angle.


After that I just softened the corners on the belt sander.  And they look good to go!


These little clamps really did make it easier to fit everything in on my little rotary table.
Jobs are much easier when you have the right tools, and this one helped!


Here's the start of my third try on that Cylinder Plate, all squared up and marked out.


Now I've drilled all the holes and am up to the point where I messed up on my 2nd attempt.


Carefully centering the hole over the center of the RT.


And with some trepidation, I set out to round off the first protrusion.  I used tape to mark the beginning and ending points of my radius.  Yet another excellent suggestion from the list.  And note the increased visibility around the much smaller clamps!


And after some work, I got all four of those nubbins rounded, and I could hardly be happier!


One more close-up shot of the completed piece.


I'll eventually have to drill the through holes to match the holes for the cylinders.  But that will come...

Next up will be the support columns (I think...)
Kim


Edit (3-18-13): Moved the pictures to the right folder & updated post to reference the new location.

[ths]

No compromises with you! Good on ya.

Hugh.

[Jo]

Looks good Kim 

One of the tricks I learnt from following George Britnell's work is that he puts a brass button into the holes as he rotates the table. Thus when you get to within a smigin of where you need to get to the button will rotate and you will be reminded to stop. 
 
Jo