What's new
What's new

Designing production machine: need air activated something to grab 8-32 screws

michiganbuck

Diamond
Joined
Jun 28, 2012
Location
Mt Clemens, Michigan 48035
QT: [He said demand was hundreds a day]
For only 100 per day, I would put a manual adjustable swing arm fixture next to a bench grinder using a plate-mounted AO wheel, so to grind off .015 at a slight angle so to have no burr at a rate on about 6 seconds each.

The slight angle would be only 1/2 degree but could be dead flat if needed to be dead flat.
 

gustafson

Diamond
Joined
Sep 4, 2002
Location
People's Republic
QT: [He said demand was hundreds a day]
For only 100 per day, I would put a manual adjustable swing arm fixture next to a bench grinder using a plate-mounted AO wheel, so to grind off .015 at a slight angle so to have no burr at a rate on about 6 seconds each.

The slight angle would be only 1/2 degree but could be dead flat if needed to be dead flat.

Sometimes we get caught up in the more must be faster when it is not always the case. This might be a solution if it reduces the human handling of the part
 

surplusjohn

Diamond
Joined
Apr 11, 2002
Location
Syracuse, NY USA
I have been thinking about this. Horizontal disc with drilled holes. Second stationary back up disc underneath it. Drilled disc rotates. Set screws are dropped in holes and disc takes them under a grinder then over a hole in the lower plate that lets them drop out. Question is can you get the end square enought without scwewing them in.
I had a similar problem where i had to shorten many thousands of half inch ss bolts by a quarter inch. I set up a fixture on a brlt grinder and screwed the bolt in while grinding it. Then unscrewed it which knocked the burr off. Then gave it a turn in a cradle to champer the ends. This was fast and the results were good.
 

GregSY

Diamond
Joined
Jan 1, 2005
Location
Houston
So....hundreds a day equals (at least) 200 x 365 = 73,000 per year minimum.

If a jeweler's saw uses 6 (?), then that's 12,166 saw per year, or 1,013 per month, or just under 34 saws per day.
 

Alberic

Cast Iron
Joined
Jan 15, 2011
Location
SF Bay
So....hundreds a day equals (at least) 200 x 365 = 73,000 per year minimum.

If a jeweler's saw uses 6 (?), then that's 12,166 saw per year, or 1,013 per month, or just under 34 saws per day.

Yeah, that's about right. Bit low, if memory serves from last year's numbers. Each saw only uses 4 (of these) screws. 2x half inch, 2x .188. So roughly 20K of each size per year is what I'm ballparking on. If I can keep up with that, I should be good.

I've tried dog points, but they're not flat enough. The other issue is the oxide coating. It's slick. Which causes problems when you're using them as a clamp. I've had HoloKrome make flat points without the oxide, (which they can do) but the lead time is +6 months right now.

The rotating disk idea is interesting. I'll see if we can get flat-enough that way. Worth a shot anyway. (If they're clamps, the tips have to be dead flat, and square.)
I wonder if there isn't a way to do some sort of cam-actuated collet-ish thing to get a solid clamp on the screw at a certain point in the disk's rotation......hummmm.....

The Alberic.net saw listed earlier is my personal site, and the blue saw was my personal saw, from when I started out as a jeweler at 15. Tres ergonomic, or at least that's what I thought at the time. I don't think I've used that saw in....15 years now?

This is what we do now:
Knew Concepts Precision Crafted Saws and Tools

Thanks,
Brian
 

Thunderjet

Hot Rolled
Joined
Jun 24, 2019
So, I'm assuming that you have folks that assemble these saws, this is where my idea begins.

There are folks that have a box of screws that are hand picked and place on an allen wrench of some sort to insert into the saw assembly. I'm assuming they do this by hand, to get then started properly.

Here is my idea.

Before they insert them into the assembly, they use a simple table top disk sander and place them on a v-block fixture and sand them until flattened sufficiently, then go directly to assembly.

Handled by hand ONCE. One operation added, assembly complete.

Our company teaches assemblers how to do things like this all the time. It works.
 

Thunderjet

Hot Rolled
Joined
Jun 24, 2019
I actually just remembered this product we use for locking sliding components:

Object moved

I think these would be perfect for you, and no grinding.
 

Alberic

Cast Iron
Joined
Jan 15, 2011
Location
SF Bay
So, I'm assuming that you have folks that assemble these saws, this is where my idea begins.

There are folks that have a box of screws that are hand picked and place on an allen wrench of some sort to insert into the saw assembly. I'm assuming they do this by hand, to get then started properly.

Here is my idea.

Before they insert them into the assembly, they use a simple table top disk sander and place them on a v-block fixture and sand them until flattened sufficiently, then go directly to assembly.

Handled by hand ONCE. One operation added, assembly complete.

Our company teaches assemblers how to do things like this all the time. It works.

That's more-or-less what we're doing now. Not flat enough, unfortunately. Or at least testing has discovered that grinding the tips down 0.020-0.030" widens out the contact patch enough to make a difference in grab strength.

The real problem here is that we're in a race between the strength of the sawframe, and the holding power of the clamps. The original frames could only pull about 45 pounds of tension. (Compared with a max of 25-30# for a traditional design.) So the original clamps, with screws that were just hand sanded were plenty strong enough. Our newer sawframes can pull north of 80# of tension. Which means they're plenty stiff enough to either tear the blades apart, or rip them out of the clamp. But if they slip out of the clamp, even if they do it at more than twice what a traditional sawframe can do, then the user gets upset that the blade is slipping. So I'm working to increase the grab strength of the clamp as much as I can, to keep up with the improvements in the sawframes. There are other styles of clamp that exist, but unfortunately, some of our clamps rotate around their centerline. So we need to make sure that the blade is grabbed very near the CL of the clamp. Which you can't guarantee with any other style of clamp. So.....improve the screws.

Thanks for the input guys, I do appreciate it.

Regards,
Brian
 

Alberic

Cast Iron
Joined
Jan 15, 2011
Location
SF Bay
I don't know if any of you know the backstory, but Lee Marshall (Oldster) invented the original saws as a retirement project. He and I knew each other from his jewelry tool line. I moved up to the bay area after I got married, and agreed to help him out for a couple of weeks as a machinist while I was job hunting. 13 years later.....
He passed away about 5 years back. One of the things I truly miss is that he and I could sit around the office and bat ideas for stuff like this back and forth until we'd figured out a way forward. With him gone, I don't have anybody really to do that with. So I really appreciate you guys filling in for him.

Regards,
Brian
 

GregSY

Diamond
Joined
Jan 1, 2005
Location
Houston
I think if you're not grabbing the set screw with a collet, or using the threads, it'll never stay square with the grinding wheel. It'll cock over. Or fly out.
 

Chip Chester

Diamond
Joined
Dec 13, 2007
Location
Central Ohio USA
Would grinding a small flat on each side of the blade, right at the end, weaken the blade too much? If there was even a slight recess the screws could land in, the face of the screw could be much less important.

Of course, that means you're in the blade business, if you're not already...
 

michiganbuck

Diamond
Joined
Jun 28, 2012
Location
Mt Clemens, Michigan 48035
Holding it from spinning a .188 and .500 length set in a .165 hole would stay straight for a .015 grind off the end. a turn/rotation in that hole would take off the burr, IMHO.
Yes, we would need to know what kind of screw it might be so as to know if it has a head of some sort.

This holding device (just a hole) in position to a parting wheel might suffice. But cutting into the thread would deflect the wheel a little and with .032 per thread would hole
a very close length.

Simple grinding off if there is only .015 of taking (needed stock removal).

Grasping the threads would be difficult because one would then need to turn the thread tight to the head shoulder to make size/length.

For automation a vibrator bowl to line up the parts (screws), A robot of sorts to move the part to the sizing machine (perhaps just a slide ramp), and a sizing machine/device. Very big bucks making this special machine.

I worked in a shop that did/made special machines (Interlakes Engineering), but I would not be capable of designing one.

I could fudge up a simple fixture and run the part on one of my grinders at a couple hundred an hour or so. Two hundred an hour would make the cost about .15 each.

Running them on the mill would take longer and require deburring.
 

Thunderjet

Hot Rolled
Joined
Jun 24, 2019
One of the things I truly miss is that he and I could sit around the office and bat ideas for stuff like this back and forth until we'd figured out a way forward. With him gone, I don't have anybody really to do that with. So I really appreciate you guys filling in for him.

This is what my boss and I do regularly. It spooky how much we think alike.

I'm sorry for your loss, but if you try to channel him thru his toolbox, I'm sure you'll come up with something.

I do design stuff like this every day, so if you need more ammo, just LMK.

If your married to the grinding thingy, I do have more ideas, just too complicated to explain here.
 

surplusjohn

Diamond
Joined
Apr 11, 2002
Location
Syracuse, NY USA
Annother idea. Threaded plate fastened in front of a vertical belt sander. Pick up screw with powered screwdriver. Screw into the threaded hole against belt and then unscrew with a rail to strip screw off of bit allowing the screw to fall into a can. It seems like all the work is handling the part and the actual grinding is presumably minimal. Alternatively. If this is an allen drive.
Pick up the screw with a allen wrench and just push it into a drilled hole
Grind and retract.
 








 
Top