[4]

Machine Guarding

[3]
[10] Like what you see?
Click here to donate to this forum and upgrade your account!

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#1
As some of may know I'm building a small punch press for a customer. https://www.hobby-machinist.com/threads/servo-punch-press.72601/ This press is going to be out of my control and operated by unskilled or semi-skilled labor so it must be designed to be as safe as possible short of welding a large steel box around the entire unit and welding the main power switch in the Off position. :) In the die area there are no pinch points, completely finger safe. No reason to be poking around in there, but ya never know. :cautious: It will be guarded anyway with a lexan guard. And that guard will contact a limit switch that is in the E-stop circuit and must be in place to operate the machine. Designed to not be easily defeatable.

I had to add a stiffener to the press frame to keep it squared up, and in doing so created a pinch point at the wrist pin.

Without the stiffener installed
1538280483233.png


And with the stiffener in place. The bolt head moves up & down with the ram.
1538280539594.png

No problem, just build a lexan guard to cover the pocket.
1538280694051.png

But, this guard is WRONG. Mechanically it will work fine, the lexan is well suited for this application, and is completely finger safe. The guard allows easy access for removal of the wrist pin for die maintenance. Die maintenance would be done by skilled personnel, this is not an operator function. I have a fix for the problem and will implement it as the build continues, there are two solutions that I thought of.

But I wanted to post this to see if anybody else can point out the problems (there are more than one) with this design, it took me a day to realize that it wasn't right.

Time to put on your thinking caps. :eagerness: Looking forward to your comments.
 
Last edited:

Norseman C.B.

Active Member
Registered
Joined
Dec 21, 2013
Messages
189
Likes
88
#2
If I'm seein it right it can't cycle down cuz the relief pocket is positioned wrong.......
 
Last edited:

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#3
You are correct in that regard, it needs another 0.125 clearance, but it is nearly at the maximum down stroke in the photo.

But that's not the problem, the problem is in the human factors.
 

Martin W

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Mar 26, 2014
Messages
382
Likes
352
#4
I see no form of lubrication spots. Return spring?

Cheers
Martin
 
Last edited:

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#5
I see no form of lubrication spots. Return spring?

Cheers
Martin
Lubrication will be added later. Spring return is close if I understand correctly :)

The only issue here is the guard itself. I'll post the answers in a day or so if nobody else gets it. :)
 

francist

Active User
H-M Supporter - Gold Member ($25)
Joined
Sep 5, 2013
Messages
1,096
Likes
1,664
#6
It might be a safer guard if the mounting screws came in from the back side of the plate. Less chance for us dumb operators to take it off....

-frank
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#7
It might be a safer guard if the mounting screws came in from the back side of the plate. Less chance for us dumb operators to take it off....

-frank
You're getting warm :)
 

derf

Brass
Registered
Joined
Oct 3, 2015
Messages
661
Likes
770
#8
For those of us that have a mechanical inclination and a rational thought process, it can be quite hard at times to think like an unskilled idiot, thus the need for beta testing to see how the mechanically impaired understands a simple process.
Just spitballin'.....since the cover is transparent, you will be able to see the wrist pin moving up and down, and to an unskilled operator this situation would indicate that something is loose or broken, triggering a brainstorm about investigating the situation by removing the guard.
Curiosity killed the cat....
 

dulltool17

Active User
Registered
Joined
May 9, 2014
Messages
324
Likes
212
#9
If the machine will still operate with the guard removed, the hazard still exists. You need an interlock, either mechanical or electrical.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#10
If the machine will still operate with the guard removed, the hazard still exists. You need an interlock, either mechanical or electrical.
That's one of the issues. :clapping:Still at least one more that I thought of.
 

dulltool17

Active User
Registered
Joined
May 9, 2014
Messages
324
Likes
212
#11
Well, you don't have the actual point-of-operation guarded at this point, either.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#12
Well, you don't have the actual point-of-operation guarded at this point, either.

That is true, just because it hasn't been built yet. It will be before the machine ships.
 

francist

Active User
H-M Supporter - Gold Member ($25)
Joined
Sep 5, 2013
Messages
1,096
Likes
1,664
#13
This may be a stretch, but is the screw pattern in the guard a square? If so, could that lead to the guard being installed incorrectly on the horizontal as opposed to the vertical orientation?

-frank
 
Last edited:

Radials

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Feb 7, 2018
Messages
41
Likes
65
#14
This may be a stretch, but is the screw pattern in the guard a square? If so, could the guard be installed incorrectly on the horizontal as opposed to the vertical orientation?

-frank
Perhaps a small dowel pin pressed into the back of the lexan guard that has a corresponding hole on the stiffener plate so that it could only be installed one way.

Does the wrist pin rub on the guard?
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#15
Perhaps a small dowel pin pressed into the back of the lexan guard that has a corresponding hole on the stiffener plate so that it could only be installed one way.

Does the wrist pin rub on the guard?

You're close, it really needs a shoulder screw permanently installed and locktited in place in the top so it can swing closed by gravity. But there is still the problem that the machine will run with the guard removed. Overall just a bad design and not foolproof or failsafe. No, the bolt head clears the guard by about 0.025 or so.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#16
Guys, thanks for commenting.

I have been designing machine safety systems for years, and one of my thought processes is to be the most pessimistic SOB on the planet. If anything can go wrong, it will do so worse than you ever imagined. :cautious: That little guard looked cool, but it is no good. The human factors are that someone will remove the guard while maintaining the die. Then they will lose it, lose the screws, or for some other reason the guard won't get put back on. Now you're going to say that nobody would stick their finger in there, but you would be amazed at the stupid stuff that people do.:rolleyes:

The guard over the slot will be incorporated in to the main guard over the die area, which will open the E-stop circuit when removed. All I have to do is extend the guard up a bit to cover the slot.

I added some clearance to the slot, now the die can close to the absolute minimum height.
1538454094071.png

There is quite a bit of finger clearance, the die is about 1/8'' above minimum in this pic, and the punches will never come out of the punch guide plate. But....It is still going to get a guard.
1538454546846.png
 

astjp2

Active User
Registered
Joined
Aug 23, 2013
Messages
770
Likes
92
#17
Just buy a Minster. problem solved
 

Reddinr

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Jan 5, 2014
Messages
132
Likes
63
#18
Is there room for a taper on the front and sides of the top die plate? If guards are removed / interlocks bypassed by a large thumbed, unthinking but high initiative operator, risk might be reduced even further.
 

astjp2

Active User
Registered
Joined
Aug 23, 2013
Messages
770
Likes
92
#19
actually you need to look at the osha regs on presses (especially mechanical), mute height is limited to 6mm, 2 hand controls are sufficient guarding for the operator, but you need to consider limiting other into the area and a way to prevent debris if the tool breaks from hitting anyone. I have to deal with them at work. You also need to consider stop time and stop distance and the measured distance from the operator.
 

astjp2

Active User
Registered
Joined
Aug 23, 2013
Messages
770
Likes
92
#20
Our punch presses at work use a bolster plate that we can bolt up a "BT" die, you bring the upper down to match the lower, tighten up the bolts and then set your snap through height with a piece of paper. Carr Lane sells the punches and you can use a surface grinder to sharpen the whole punch and die assy. WAY more easier that building a whole machine. and reinventing the wheel.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
7,272
Likes
5,571
#21
Our punch presses at work use a bolster plate that we can bolt up a "BT" die, you bring the upper down to match the lower, tighten up the bolts and then set your snap through height with a piece of paper. Carr Lane sells the punches and you can use a surface grinder to sharpen the whole punch and die assy. WAY more easier that building a whole machine. and reinventing the wheel.

I agree. I looked around for a high speed punch press with a 3/8 or 1/2 inch stroke and couldn't find anything in the 1 to 3 ton range. The design was to run at up to 600 hits/min, but it really seems happiest at about 280 to 300. I accidently ran the ram at 2500 hits/min for a few seconds so I know it will do it. The limitation on speed is really the inability to feed any faster with the current set up, the ram actually strokes at an equivalent speed of about 1000 hits/min, but the ram has to wait for the feed to position the material before it can stroke. The feed is also variable, on-the-fly, from 0.75 to 6 inches so that adds another layer of complexity to the mix. The feed is actually pulling the material off of the roll and any faster I start getting miss-feeds. The feed rolls are already marking the material so I really can't up the pinch pressure any more. The material handling is outside the scope of this project, so it is what it is. The snap through height is also variable depending on the product they want to run, it punches 3 different patterns depending on the shut height.

I bought the punches and die buttons from Dayton Lamina. To reduce mass I decided to not use a normal die set, and to attach the punch holder directly to the ram base. I can have the punch holder and die holder out and on the surface grinder in about 15 minutes. In this case, the die holder assembly actually sits on what would be a bolster plate in a normal press system.

This press has to be completely guarded because it is one hand (more like 1 finger) operation. https://www.hobby-machinist.com/threads/servo-punch-press.72601/page-2#post-618498

Bottom line is that this ain't a normal press, so that's the reason it was purpose designed and built from the ground up. :)
 
Last edited:
[6]
[5] [7]
Top