Question concerning Light Curtains!

[grege101]

I am installing a set of light curtains on a injection mold machine. I need to see an electrical example of the proper way to mute out the curtains when the machine is at its home position. It's a very common process. Operator loads the die, steps back and starts the machine, if at that point he breaks the light curtain the machine must shut down, which it will at this point but when the machine is home he must be allowed to enter without killing the power. Help would be greatly appreciated!

 

[halosome]

I used light curtain before, and it was quite easy, all you got to do is provide 3 dry contact outputs to its control box, 1 for start, 1 for end and 1 for mute. and wire its dry contact in your control circuit. The trick is the end contact must be closed (opened?) before the start contact, then the mute state is reset to the safe state.

 

[hounddog]

ab light curtain

here is a sample of schematic that i did a couple of years ago for a press type machine. Used the light curtain for point of entry.
Had a prox switch to determine that a shuttle was at home position.
If so it would keep a mcr relay energized.
If shuttle moved to start machine cycle and anything entered the light curtain again, it would drop out the mcr and kill the machine, which was what was wanted.
Sorry, can not go into a lot details, but machine was inspected by a OSHA inspector and they was pleased and gave it the okay to used in production and use the machine as a point of entry device.
Built 10 of the machines.
Hope the schematic helps

 

[grege101]

Houndog: What happens if that sensor becomes faulty and is held high? Curtains are bypassed all of the time. Are you checkin for a change of state?

 

[ndzied1]

I had the same question. Usually control reliable means doubling up on things as you have on K1 & K2.

I'm not sure how a N.O. prox fails but I have had N.C. proxes fail closed.

Now looking at the P/N on the print if it is the 440P that is a mechanical safety switch but you're relying on the spring to open the circuit and the contacts not to weld. The "safe" state with "positive action" or "force guided" contacts is when the contacts are open. Your safe (ignore the light curtain) state is when the contact is closed.

I think ideally, you'd want to use a normally closed safety contact that is mechanically held open (against the spring) during the non-safe part of the cycle. Then if the spring breaks, the contacts will remain open and the light curtain is active. Depending on the results of your safety assement, you may need to duoble up on the switches and do monitoring.

See here: http://www.sti.com/pdf/805.pdf
for more information.

<disclaimer - I work for an Omron-STI distributor>

 

[Jiri Toman]

I have only one thing to say, here in the USA I have never ever seen an injection molding machine without a door !!!

 

[hounddog]

thanks
Yes there are more than one prox switch on the machine and looking for change of state etc., as well as safety doors with ab safety switches mounted on them as well as other light curtains to ensure that nothing can get in the area that is being pressed etc., only shared the one page as a reference tool. Can not share the others, or any other info due to fact that there is a pantent process going on. Thought I would post this sample page since when I was trying to do the same thing and had questions, had a difficult time making sure all of the OSHA requirments etc. where met.
Had our OSHA safety person through our corporate office come in a couple of times, and did have to add some other safety features to meet OSHA requirements, and with some other info I got from this site and mrplc.com, finally got a machine that very safe.

thanks for info

 

[brucechase]

You should not be relying on people on a couple of web sites to make your machine safe. You should be performing a risk analysis of each operation and each part of the machine. From there you will ascertain safety categories needed to minimize the perceived risk and you can select the proper safety device.


If you don't know how to properly do this, you really shouldn't be messing with it without a mentor or training. Realize that what you pick and how you program the mute on a light curtain will open you up to legal and moral responsibility if someone gets hurt or killed. This safety stuff is too important to just look on the internet or listen to what some anonymous person (who might not know what he is talking about) writes on a forum.

You need to step back and rethink your whole safety process.

 

[grege101]

Thanks for all of your replys.

NZIDE1: I've seen countless N.O. Prox's fail. For instance, a small piece of metal gets enlodged in the surface of the inductive prox holding it on at all times. Or the prox getting hit too many times and malfunctioning.

I agree with the normally closed safety contact that is mechanically held open and I also think that it would definitely be a good idea to have two of them in series below the light curtain contacts for redundancy. That was basically what I was thinking about doing just wasn't sure if it would be sufficient enough.

Jiri Toman: These are 30 year old machines with basic safety requirements. Hence the reason we are upgrading. There will be a door on it but there also is an area where the operator has to be able to reach up under the machine. This is where the L/C's come into play. Hard to explain without pics. Which naturally I cannot show you.

Hounddog: I understand that you cannot share any more than what you have and assumed you must have been checking for change of state, more sensors,etc. Thanks for the schematic. It helps people see what I'm trying to get at.

In conclusion, does everyone agree that the safest way to do this would be to use 2 sensors in series below k1 and k2? Any more thoughts would be greatly appreciated.

 

[grege101]

Brucechase: An engineer has to sign off on it. No worries, I would just like to here other peoples opinions on how they've done similiar projects.

 

[JesperMP]

Is the solution not to NOT stop the entire machine, but instead only the dangerous movement that the operator can get in contact with. In that way there is no need for a muting function. In stead it is the other way, if the operator tries to access the dangerous area "too soon", the machine sequence is interrupted and the HMI displays an appropriate error message "light curtain interrupted within the sequence". This part does NOT have to be handled by a safety relay, as long as the safety relay cuts the direct dangerous movement.

Let me ask: In which way do you stop the dangerous movement ?

In my company we use light curtains in a cat 4 configuration in exactly this way. The movement is stopped by hydraulic valves with feedback to the safety relay.

 

[hounddog]

You should not be relying on people on a couple of web sites to make your machine safe. You should be performing a risk analysis of each operation and each part of the machine. From there you will ascertain safety categories needed to minimize the perceived risk and you can select the proper safety device.


If you don't know how to properly do this, you really shouldn't be messing with it without a mentor or training. Realize that what you pick and how you program the mute on a light curtain will open you up to legal and moral responsibility if someone gets hurt or killed. This safety stuff is too important to just look on the internet or listen to what some anonymous person (who might not know what he is talking about) writes on a forum.

You need to step back and rethink your whole safety process.

First of all, I thought this was a forum to share ideas, not to bash others. Second, the relying on others that you are talking about are OSHA inspectors, and yes, this forum is always a good place to ask others about ideas, after all, what do you use it for?, or are you so smart you never have to ask? May be possible, but I doubt.
Third, if you dont know to do this, you should not be messing with this ---- Well, if you consult with OSHA then you find out exactly what needs to be done, unless you just happen to be one of those that knows everything, like maybe yourself.
Also, as far as you being liable for killing or hurting someone - Well that is why OSHA was consulted, as well as our corporate lawyers to ensure that everything was as safe as possible.
And yes, myself and I am sure others use this forum to possibly get advice and maybe even to help someone, but never to bash.
I found out a long time ago that there is always someone out there who knows a better way, and I also found out that it is very easy to bash someone when you do not have to worry about meeting them face to face, after all, there are some folks on this earth that you will run into that it is best not to mess with, but if you never have to run into them, then i guess its okay to talk.

After all, this forum, but the best i have been able to tell is a great place to share info and idea's, seems to me if you not willing to do that, then maybe you should take up golfing or something else.

 

[hounddog]

Thanks for all of your replys.

NZIDE1: I've seen countless N.O. Prox's fail. For instance, a small piece of metal gets enlodged in the surface of the inductive prox holding it on at all times. Or the prox getting hit too many times and malfunctioning.

I agree with the normally closed safety contact that is mechanically held open and I also think that it would definitely be a good idea to have two of them in series below the light curtain contacts for redundancy. That was basically what I was thinking about doing just wasn't sure if it would be sufficient enough.

Jiri Toman: These are 30 year old machines with basic safety requirements. Hence the reason we are upgrading. There will be a door on it but there also is an area where the operator has to be able to reach up under the machine. This is where the L/C's come into play. Hard to explain without pics. Which naturally I cannot show you.

Hounddog: I understand that you cannot share any more than what you have and assumed you must have been checking for change of state, more sensors,etc. Thanks for the schematic. It helps people see what I'm trying to get at.

In conclusion, does everyone agree that the safest way to do this would be to use 2 sensors in series below k1 and k2? Any more thoughts would be greatly appreciated.

Glad the drawing i posted was helpful. was unable to post the final drawings due to some legal issues, it has the company name, the patent info etc. on it. But, i did post one of the first drawings from the first prototype to maybe help give you a idea.
The question as to wheter or not it is safest to do this with 2 sensors in series with k1 or k2, the final machine that we finished has three sensor in series with another light curtain in series with the k1 and k2.
Hope this is helpful info.

 

[panic mode]

first of all you need to realize what standard you need to follow. if this is a press (sounds like it) then general machine safeguarding standard (in Canada Z-432) will not cut it, you also need to follow press standard (ni Canada Z-142).

if you have press and machine Cat4 is the way to go but just Cat4 safety circuit on control side is not going to cut it since this is a press (read standard, there is number of things to consider, there valves also need to be monitored etc. because of inertia, stored energy etc. use of external braking system able to stop the press may be required - such as Zimmerman blocks etc.)

you may need to do muting to allow loading of the parts for example. you need to do risk evaluation and determine if what you want to do is acceptable and what category circuit is needed. muting is bypassing safety function and it is something that should be left as last resort. i sure do my best to avoid muting where possible. muting circuit must at least meet category level of the circuit that is being muted (light curtain). in other words you need Cat4 to mute Cat4.

If this is accepted, you may setup suitable circuit to monitor retracted position of the device (die for example). to get Cat4 you can use number of devices such as Psen sensor (ensure proper rating) with matching controller (you must read which controller is certified to be used with it). you could also use pair of proxy sensors and wire them to Cat4 relay (in this case sensors would need to be of complementary polarity since this is how proper safety releays detects cross channel faults). but if you need to do this, it is better to just used dedicated muting controller. there should be yellow light to indicate when muting is active and good muting controller tests if the light is connected and functional (disables operation if light is disconnected or burned etc.).

i will try to post some circuits later on (no time now)

 

[brucechase]

First of all, I thought this was a forum to share ideas, not to bash others. Second, the relying on others that you are talking about are OSHA inspectors, and yes, this forum is always a good place to ask others about ideas, after all, what do you use it for?, or are you so smart you never have to ask? May be possible, but I doubt.

hounddog - you obviously are the one with a problem. Read the OP. Nothing was said about any inspectors or anyone else. If you don't know how to properly engineer a safety system, then you should not be doing it.
As for the bashing, it is only you that is bashing. Nowhere in my post was anything said that was derogatory - UNLIKE YOU!! Since my comments were originally directed at the question the OP asked and not you, maybe you should mind your own business until I direct my comments to you. Remember, you started this.

I get the feeling that you are the one who really doesn't know anything and are trying to cover it up by attacking others.

Third, if you dont know to do this, you should not be messing with this ---- Well, if you consult with OSHA then you find out exactly what needs to be done, unless you just happen to be one of those that knows everything, like maybe yourself.

Again, you are wrong. OSHA will NOT tell EXACTLY how to do anything. You might want to educate yourself before you speak.

Also, as far as you being liable for killing or hurting someone - Well that is why OSHA was consulted, as well as our corporate lawyers to ensure that everything was as safe as possible.
And yes, myself and I am sure others use this forum to possibly get advice and maybe even to help someone, but never to bash.
I found out a long time ago that there is always someone out there who knows a better way, and I also found out that it is very easy to bash someone when you do not have to worry about meeting them face to face, after all, there are some folks on this earth that you will run into that it is best not to mess with, but if you never have to run into them, then i guess its okay to talk.

WOW! it sounds like you are threatening me. You might want to be careful about what you say.

After all, this forum, but the best i have been able to tell is a great place to share info and idea's, seems to me if you not willing to do that, then maybe you should take up golfing or something else.

You go take up golfing. You obviously have nothing good to add for me. This is a real good reason that a previous member (RIP Ron) showed me the ignore file.