Chevy V8 Scale Engine Plane

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
If your back jaw is perfectly trammed in to 90 degrees to your machine head using the method I suggested you can get within .0005 fairly easy. You take the jaw lift totally out of play that way.

I tried using a piece of round stock and was having no luck. I changed to a piece of copper wire and got slightly better. In both cases they were about the width of the chuck. I don't know if a shorter piece would make any difference.

I checked the vise ways and they are level within .0002 over their exposed length. The vise fixed jaw is square within .0001 over its height. The fixed jaw flexes .0003 when tightening stock. Mill is trammed well within a half thou over about 6 inches. The only thing I have found that helps is a piece of paper extending about .25 from the top of the fixed jaw. It overcorrects by between 1 to 1.5 thou but I don't have anything thinner at the moment. I think I'll pick up some cigarette paper and try that since it is thinner. I'm probably chasing my tail but I would like things square a parallel within a thou for the start of the block.
 
When using a piece of round stock use only 1 thin parallel up against the back jaw. Just tap the block down on the parallel don't hammer it to hard. After that for the next cut put the face you just milled against the solid jaw and do the same. If your jaw is within .0001 if you are real careful you should have no problem getting the block within .001.
PS don't over tighten you will push the fixed jaw out and loose your perpendicularity of the jaw on you finish pas just snug the vise no need to clamp the crap out of it.
 
Yep. That's the process I've been doing. I've got a single parallel at the fixed jaw, the shortest one I have. I give it three medium hits with a dead blow hammer at left, right then center. I have been giving a snug - tightening the jaws - after hammering the stock down. I'm using a fly cutter until I get things dialed in. I face side one then put 1 against the fixed jaw and face the next side. I then put side 2 down with side 1 on the fixed jaw and repeat.

I'll try no snug after tapping stock down and shorter piece of round stock (individually) to see if that helps. Otherwise, I'm at a loss as to why the difficulty in getting the stock square.
 
Try this put an indicator on the back side of your solid vise jaw before you tighten and check the movement when you tighten your vise.
 
I previously had put an indicator on the top face of the fixed jaw and tightened a 1-2-3 block. I measured .0003 when cranking down fairly heavily.

I think the main issue is moving jaw lift. I get significant lift unless I barely tighten the jaws. I may need to take it apart to see if something is wrong between the nut and the sliding jaw.

I've gotten three sides square a parallel within a thou and dimensionally within a thou. It takes a lot of time to keep going from the mill to the surface plate and back - what a pain. I am getting much better results if don't tighten the vise too much. That also resulted in the part shifting when I was milling .100 off to get a side to dimension. Fortunately I caught it in time before the block was ruined.
 
If you use a round stock on the movable side the lift of the movable jaw doesn't matter. The solid jaw has to be moving put the indicator on the back side of the solid jaw when tightening. If it moves then it is no longer perpendicular to the base which will make your part come out like you are getting.
 
Here is a little demo video I put together today to show the effects of solid jaw when clamping and how it will not stay perpendicular when to much clamping pressure is used.
 
Thanks for the video, Doc. I did some tests on my vise to see what was happening and took photos along the way. I swept the fixed and moving jaws unclamped. The rear jaw tilts away from the moving jaw .00015. The moving jaw has a .0001 bow with the middle protruding to the fixed jaw. Pretty good really. The vise is a Precision Mathews 5-inch. They claim it is quite accurate and they are correct. It is clamped directly to the mill table but only with two bolts at the mid-point. I used an old beam type torque wrench and did measurements of both rear jaw movement and front while clamping the 3-inch side of a 1-2-3 block. The torque values are written on the right side of the photos.

Basic setup. I have a .00005 resolution Tesastat on the clamping face of the rear jaw. A cheap Chinese knockoff on the top of the 1-2-3 block at the moving jaw. Both have been checked against gage blocks and are very accurate although the knockoff has more hysterisis.
IMG_20190504_102930.jpg


With only 5 ft-lb the block lifts .001. The rear jaw flexes back just under .0004. 5 ft-lb is barely tightening the jaws.
IMG_20190504_103042.jpg


Here is 10 ft-lb which is barely snugging the work piece.
IMG_20190504_103203.jpg


Here is 15 ft-lb which is a reasonable tightness; .001 deflection of the rear jaw and .0015 lift at the front.
IMG_20190504_103252.jpg


A jump to 25 ft-lb which is very tight.
IMG_20190504_103353.jpg


And 30 ft-lb which is about the most I can crank using the short vise handle.
IMG_20190504_103507.jpg


Last is a check to make sure the indicators reset to zero after releasing the vise.
IMG_20190504_103605.jpg


The combination of the rear jaw flex and front jaw lift were biting me. With the back tilting away from the work the rear lowers. Combine that with the front lifting and you get even more tilt of the workpiece. I'm now snugging the vise for critical work somewhere between 5 and 10 ft-lbs as close as I can tell. Much less than that and the work is too loose. I have to bear down for roughing or things get to loosy goosey.

I think I'll try four clamps like you suggested to see if it makes much difference.
 

Attachments

  • IMG_20190504_103203.jpg
    IMG_20190504_103203.jpg
    67.1 KB · Views: 293
No significant difference between 2, 4 or 6 clamping points. The differences were insignificant and likely caused by inaccuracies in the torque since I was using an old beam torque wrench with 5 ft-lb resolution. One thing that was interesting was 4 points had more lift (only half a thou) compared to the 2 or 6 points. Might be a bow of the vise frame allowing for more lift? In any case, I'll stick to 2 points and minimize torque.

I've got 5 sides of the block square, parallel and in dimension within .001. Now to get the last side done and start the real work.
 
I took three quarters (the equivalent of two semesters) of maching classes from a local technical college. Classes were 5 days a week for 5 hours a day. They didn't teach this stuff. As Stefan Gotteswinter puts it, "everything is rubber."

I took some mechanical engineering classes while I was in electrical. One was called "Deformable Bodies" by the students, which is the first class where ME students learn that everything bends, sags, stretches, squeezes and distorts.

The most valuable thing that got me through that class was a Pink Pearl eraser. (These if the name doesn't ring a bell) Everything they taught could be seen with rubber.
 
I took some mechanical engineering classes while I was in electrical. One was called "Deformable Bodies" by the students, which is the first class where ME students learn that everything bends, sags, stretches, squeezes and distorts.

The most valuable thing that got me through that class was a Pink Pearl eraser. (These if the name doesn't ring a bell) Everything they taught could be seen with rubber.

Oh yeah, I remember them but I used the white ones, they left less mess.

I'm a retired EE. I went to UC Santa Barbara. They loaded us up with so much electrical and math that I couldn't take an ME class if I wanted to. No strengths of materials, statics, dynamics or other interdiciplinary courses. It was so single disipline it was rediculous. Since the department head was into semiconductors we were required to take statistical thermodynamics - thremodynamics on a quantum level - but no classical thermodynamics. How stupid is that?
 
Oh yeah, I remember them but I used the white ones, they left less mess.

I'm a retired EE. I went to UC Santa Barbara. They loaded us up with so much electrical and math that I couldn't take an ME class if I wanted to. No strengths of materials, statics, dynamics or other interdiciplinary courses. It was so single disipline it was rediculous. Since the department head was into semiconductors we were required to take statistical thermodynamics - thremodynamics on a quantum level - but no classical thermodynamics. How stupid is that?

Cool to meet another retired EE. I took Statics, Dynamics, Mechanics of Deformable Bodies, Materials, and a classical Thermodynamics class (steam tables). Back in the mid-80s. Then went on to circuit design in radio systems and weather radars. They were worthwhile classes, though, even though I really didn't use a thing out of those classes at work.
 
Thank you for posting those observations. I always suspected that this was going on and one day I will do the same tests but I would expect very similar displacements. This I know because I have never been able to get something absolutely square/cuboid! I usually finish off a precision angle using a file, chalk and a an accurate set square! I know how much the stock moves when one taps it down onto a parallel, it's quite significant. My 2 identical vises are Vertex swivel 4 inch ones, nothing special.
 
One of the hardest things to learn is how to square a block. I do think that you should find a better vice because it is making you work way too hard.
 
One of the hardest things to learn is how to square a block. I do think that you should find a better vice because it is making you work way too hard.

There is a lot more to it than the vise. I did some further measurements. The table flexes as well depending upon the position of the vise relative to the center of the table and position of the table to the spindle. For my Bridgeport clone I presently can get up to 2 thou deflection of the table. That impacts tram significantly. Next step for me is to look at tuning up the gibs to see if that helps. Any vise is going to flex and you rarely have the vise centered for every operation. I'm learning the limitations of my equipment and how to compensate for them. Like I said earlier, everything is rubber. Fortunately, I'm just a hobbyist and not trying to make a living at this.
 
Kurt NC vise will not flex under normal tightening. If the bottom of your vice is clean flat and parallel to the top and you table top is clean and flat with no dings and you clamp it down tight in 4 spots it will significantly be better. Unless your vise back jaw mounting block where it is bolted to the main part of the vise is not flat and clean when bolted together in that case it would be the same. The vise I use is a cheap knock off of a Kurt. The first thing I did with it was to disassemble it. I debured all parts and ground the bottom and top on the surface grinder then ground the solid jaw mounting block. Then cleaned everything up good and bolted the solid jaw mounting to the main vise base and tightened the hell out of it.
And I still get deflection itf it is on a swivel but on the machine table not much at all with 4 clamps. I'm just speaking of my experience that I had I know there are a lot of variables that can change the out come.
 
As an engineer of 40 plus years I feel the topic could be misleading to novices! Most drawings used in industry show tolerances to allow for inaccuracies in the process, to strive for absolute accuracy is needless unless the components eventual use demands such, we are in the position where generally the drawings we use do not show tolerances or even an open tolerance in the drawings legend therefore leading the novice to think that a dimension has to be bang on, before high accuracy and repeatability in a machining process was available could be relied upon an assembly would need some fitting before a satisfactory functional result could be achieved e.g an engines big and main bearings wold be scraped and matched to each other they would not be interchangeable no two assemblies would be the same. To get to the level you are describing for squareness an parallelism you are heading to the component needing grinding which does not need such high clamping forces therefore negating the issue of lift etc. many times magnetic clamping would be used so very little if any distortion would occur. Please excuse poor punctuation etc. and the rant.
 
As an engineer of 40 plus years I feel the topic could be misleading to novices! Most drawings used in industry show tolerances to allow for inaccuracies in the process, to strive for absolute accuracy is needless unless the components eventual use demands such, we are in the position where generally the drawings we use do not show tolerances or even an open tolerance in the drawings legend therefore leading the novice to think that a dimension has to be bang on, before high accuracy and repeatability in a machining process was available could be relied upon an assembly would need some fitting before a satisfactory functional result could be achieved e.g an engines big and main bearings wold be scraped and matched to each other they would not be interchangeable no two assemblies would be the same. To get to the level you are describing for squareness an parallelism you are heading to the component needing grinding which does not need such high clamping forces therefore negating the issue of lift etc. many times magnetic clamping would be used so very little if any distortion would occur. Please excuse poor punctuation etc. and the rant.

I was a Tooling engineer for about 25 years and about 15 years as a tool and die worker. You are very correct there is not anything perfect and for the most part I find that on most things if you are within .005 it is plenty good. But remember the smaller the part and engine the tighter the tolerances.
 

Latest posts

Back
Top