Employees getting it wrong

[donek]

We have turned on the vacuum tables on our machines manually forever without any problem until now. I have a guy who keeps failing to turn it or or turning on the wrong vacuum table and ruining material because the part gets pushed around the table.

I'm contemplating incorporating a contactor to turn on the correct vacuum motor. I'm looking for the best approach. The controller is a PRT. The PC supply in the box delivers 3.3V, 5V, and 12V. The motor power supply is 44V. I'm guessing the outputs on the controller are 5V.

The two solutions I would be familiar with would be to use an SSR (have some on hand), but I see there is a concern of over amperage even when off :https://www.omron.com/ecb/products/pdf/precautions_ssr.pdf

The other solution I could implement is a 5V relay board (http://www.amazon.com/SunFounder-Channel-Shield-Arduino-Raspberry/dp/B00E0NTPP4/ref=sr_1_3?ie=UTF8&qid=1446258177&sr=8-3&keywords=relay+board) to control a 110 voltage and then drive this contactor (http://www.amazon.com/gp/product/B001KGSJ74?psc=1&redirect=true&ref_=oh_aui_search_detailpage) . I'm thinking there is likely another way that I'm unaware of. Please let me know if there is a better cost effective solution.

[Burkhardt]

Not sure what the concern with amperage is, provided the SSR is spec'd properly and mounted on a heat sink if necessary.
On my machine I don't run the vacuum but my dust collector (about 10 amps) is controlled using a 20 amp Opto22 SSR and that has been working just fine for several years. I hate contactors due to the solenoid drive power requirements, the contact wear and the arcing EMI. Nowadays it is kind of steampunk technology.

[donek]

I woke up with what I think is a simpler solution. I'll swap out the switch turning on the vacuum with one of these (http://www.homedepot.com/p/Leviton-20-Amp-Double-Pole-Toggle-Switch-Light-Almond-R56-0CSB2-2TS/202027030) and connect the other pole to an input on the control board and add a custom header that stops the programs if it is not turned on.

[feinddj]

I think your problem is much simpler. Replace the loose nut behind the machine. If the loose nut can't turn on the vac before the toolpath start.

[MogulTx]

Yeah. If a guy can't learn to follow a procedure he is not going to rise very far and do much for you.... culling the problem early- while you have less invested ( in him as well as lost work) may be a decent answer. You could pay a couple more dollars an hour and get a better class of employee. Hate to be mercenary- but that is a pretty dang simple set of processes. Does he remember to set the speed on the spindle and hit the button to activate it?? Have you lost bits with him that way?? Just does not bode well for his intellect and ability to absorb this process and more complex processes to come....

[myxpykalix]

"I think your problem is much simpler. Replace the loose nut behind the machine"

No, just get rid of the employee if they continue to cost you money over stupid mistakes, or take it out of their pay...THEN they won't forget to do it properly.

Actually, it is simple to incorporate a popup reminder screen, although i have forgotten how. But recall when you do a Z level with the plate and after the procedure it pops up a message asking if you removed the clip and you have to answer yes/no/cancel ?

Well i'm sure you can incorporate a popup screen regarding turning on your vacumn that the operator has to click on before he can start the cutting process. Others can tell you how to do it, i have forgotten but that seems simple and you can log your files to see it is done.

[jerry_stanek]

you could use an output switch to toggle a 220 volt relay for your vac. just like the spindle switch. My air compressor has an air valve so when I shut my lights off it shuts the air down at the compressor.

[shilala]

If you want the vac to come on before the table can start winding up, some ice cube relays run out to contactors would be great.
But then you have the problem "does this employee have the part set correctly in the machine?"
There are so many ways to control the vac and machine to come on, it's hard to suggest one without seeing what you're doing in the shop, and going over your process quickly.

What I WOULD NOT do is take power for the control circuit from the psu on your computer.
I'd isolate a low voltage circuit to go through the switches (for safety's sake) and out to the contactors you'll be adding for the vaccuums.
It's not a lot of money, can be easily done, and will put your vacuum starter right at your shopbot near your input device.

You could also easily start the vacuum and create a timed lockout for the shopbot to spin up. Or use a vac proving end switch.
That's all easy, too.

Once again, I'm not looking at what you have there, and it's been about 10 years since I quit designing control systems.
I wish I was handy to take a look for you, brother.

[tri4sale]

My first thought was "fire the guy" if he keeps on making the same mistake like that over and over.

[knight_toolworks]

true. myself I have physical switches with lights at the front of the machine this makes it easy to see if the vac is one and it is easy to do.

[donek]

I agree 100% on firing him, but we have some difficulties there. We're in a rural area and have not found a suitable replacement yet. He did the job right all last season and now seems incapable of doing so. It's rather frustrating.

I tend to look at things from a bit of a different perspective though. Anytime we see a repeated mistake occurring, it tends to represent a weakness in the process. While the greatest weakness is likely the employee, finding a way to work around that employees ineptitude results in a more robust process.

Having spent a fair bit of time thinking on the issue, I think that having the CNC turn on the vacuum doesn't resolve all the problems. I think the best solution is to sense the vacuum pressure and prevent the machines motion if the pressure is below a threshold. This technique would cover failure to turn on the vacuum as well as loss of vacuum pressure mid cut. As we use a table with holes that are taped off, we sometimes see the router strip tape off the holes resulting in loss of vacuum. We've also had the vacuums filter become clogged to a point where it no longer generates adequate pressure. A vacuum sensor makes more sense to me. I've done some searching and found some adjustable vac switches, but given that I'm not currently certain of the correct pressure range, I have ordered a couple BMP180 digital barometric sensors for $3 each. I have a number of arduino nanos that should be capable of working with this sensor and relaying an on off signal to the shopbot controller. I may come up with other sensors that will be useful as well.

[coryatjohn]

I hate to say this but it might be he's started doing something that recently became legal in your state. It sounds like symptoms of self medication to me. An otherwise good employee suddenly forgetting things they have done well in the past? Pretty classic. You might have a talking to him and explain things in plain language.

[rb99]

I agree 100% on firing him, but we have some difficulties there. We're in a rural area and have not found a suitable replacement yet. He did the job right all last season and now seems incapable of doing so. It's rather frustrating.

I tend to look at things from a bit of a different perspective though. Anytime we see a repeated mistake occurring, it tends to represent a weakness in the process. While the greatest weakness is likely the employee, finding a way to work around that employees ineptitude results in a more robust process.

Having spent a fair bit of time thinking on the issue, I think that having the CNC turn on the vacuum doesn't resolve all the problems. I think the best solution is to sense the vacuum pressure and prevent the machines motion if the pressure is below a threshold. This technique would cover failure to turn on the vacuum as well as loss of vacuum pressure mid cut. As we use a table with holes that are taped off, we sometimes see the router strip tape off the holes resulting in loss of vacuum. We've also had the vacuums filter become clogged to a point where it no longer generates adequate pressure. A vacuum sensor makes more sense to me. I've done some searching and found some adjustable vac switches, but given that I'm not currently certain of the correct pressure range, I have ordered a couple BMP180 digital barometric sensors for $3 each. I have a number of arduino nanos that should be capable of working with this sensor and relaying an on off signal to the shopbot controller. I may come up with other sensors that will be useful as well.




Crazy that you have to use your super brain and a very highly technical solution to assist a deficient brain that can't remember to turn on a simple switch.

[tri4sale]

I agree 100% on firing him, but we have some difficulties there. We're in a rural area and have not found a suitable replacement yet. He did the job right all last season and now seems incapable of doing so. It's rather frustrating.

I tend to look at things from a bit of a different perspective though. Anytime we see a repeated mistake occurring, it tends to represent a weakness in the process. While the greatest weakness is likely the employee, finding a way to work around that employees ineptitude results in a more robust process.

Having spent a fair bit of time thinking on the issue, I think that having the CNC turn on the vacuum doesn't resolve all the problems. I think the best solution is to sense the vacuum pressure and prevent the machines motion if the pressure is below a threshold. This technique would cover failure to turn on the vacuum as well as loss of vacuum pressure mid cut. As we use a table with holes that are taped off, we sometimes see the router strip tape off the holes resulting in loss of vacuum. We've also had the vacuums filter become clogged to a point where it no longer generates adequate pressure. A vacuum sensor makes more sense to me. I've done some searching and found some adjustable vac switches, but given that I'm not currently certain of the correct pressure range, I have ordered a couple BMP180 digital barometric sensors for $3 each. I have a number of arduino nanos that should be capable of working with this sensor and relaying an on off signal to the shopbot controller. I may come up with other sensors that will be useful as well.

Sounds like you are close to developing a potentially marketable product. Run and go get your patents so you can sell your idea to the manufacturers!



I hate to say this but it might be he's started doing something that recently became legal in your state. It sounds like symptoms of self medication to me. An otherwise good employee suddenly forgetting things they have done well in the past? Pretty classic. You might have a talking to him and explain things in plain language.

When they started passing those laws I wondered how issues like this would be addressed. Are there any legal protections from firing someone for "self medicating".

[bobmoore]

The big iron machines that I am familiar with all have a vacuum switch that will shut the machine down if it doesn't show vacuum above a preset level. I think that level is up to you and the vacuum switch you buy. Employee forgetting or vacuum leak or any reason you lose vacuum will shut the program down until vacuum is restored.
Sometimes employees develop poor attitudes for various reasons. I have been successful sometimes by demoting that person to a less technical position for a couple weeks. Seems to shake them out of a rut. Good luck and don't let the situation get too out of control before taking action.

[Gary Campbell]

Sean...
If you haven't reached a decision and want to discuss electrical-mechanical options and programming give me a bump. Contact info in signature
GC

[myxpykalix]

It seems pretty simple to me...figure out what he is costing you in salary, loss of time and material and if it is more then you are making on the product itself, it's time to get rid of him and do it yourself.

One thing i always ask people who have this problem, and it applies to many other situations is...when is enough, enough? Do you get rid of him when he costs you $100 or $1,000.? Sometimes it's a tough thing to do but you need to know when to cut your losses.
Give him an ultimatum and an incentive not to screw up. The next time he does it, the cost comes out of his pay or he loses his job. As long as there is no punishment for screwing up, there is no incentive to correcting the behavior. Frankly that is what is wrong with society today.

[MogulTx]

I popped off and said he should get rid of him too... but I understand NOT getting rid of him. I have had guys who I KNOW will not amount to a whole lot, and I keep them anyway. (NOT ALL OF THEM!!) I had a guy, who, in 2 weeks, broke two of our most custom --- "CUSTOM!" (at the time) items- and that had a retail value of near 10K each. ( Different business. Expensive materials. Lots of labor involved....) However: I kept him. The error he made has never been repeated since. He is not my fastest employee. He is not my brightest employee. He is a slow mover- but he is consistent. He does not "reach really high"... but he shows up every day. He performs consistently. He is on time, and he will do ANYTHING I ask of him. He is not a rocket scientist... but he is a team player. He will not earn the major amounts of $ that an aerospace engineer would.... but he is a good machine operator.... I am sure donek has a rough idea in the back of his mind as to what he should do and what he can do, and what he would like to do--- and that he will blend those three things and come up with what he will do. If it is automation- no problem. If it is a new hire, no problem AFTER the hire, training, shake out and assessment of the new guy... There are a lot of answers to the problem. None of them is right or wrong, per se. All of the decisions come with potential upside and downside.... and someone has to decide what the course will be... i wish him the best of luck and insight in making the decision.... he'll need it.

[Gary Campbell]

Monty....
There are few ways that you can eliminate stupid errors by employees that are not "of the best quality". That said, when you are using a programmable machine you can add checks, steps and verification by those employees to reduce them, hopefully to a tolerable level.

Not saying that complete "idiot proof" is achievable, because as we know, a new class of idiot can be spawned on a daily basis.

[curtiss]

Not quite sure how many you cut at one time... but could you have some sort of vinyl template on top of the bed that would match the shape of the final cut ???

[donek]

Sean...
If you haven't reached a decision and want to discuss electrical-mechanical options and programming give me a bump. Contact info in signature
GC

Thanks Gary. I spent about an hour today running a wire and installing a double pole switch on the offending vac table. It works fine. I created a new custom cut file. Code is below. I'll just add the custom file at the top of each program.


'Test Vacuum

test:
on inp(2,1) goto success
'Vacuum is not turned on
pause
goto test

success:

end

[BrandanS]

donek,

You could use:

'Waiting to Activate Vacuum
PAUSE UNTIL(2,1) 'Make sure there is no Space between UNTIL and (2,1)

[Remainder of Code Here]

This would work even better if your relay was 'latched', Input 2 to initiate coil(in parallel to SB Control Board), N/C Switch from SB to kill latch at the end of your program...

You could also just save yourself the hassle, and control your double pole, outputting automatically with SB at the beginning of the file?

eg.

SO, 3,1 'Turn VAC on

[Remainder of code here]

SO 3,0 'Turn VAC off


Hope this Helps!

[MogulTx]

You are a good guy, Sean. I hope this solves your troubles.

Monty

[scottp55]

+1 to the above Sean!

[shilala]

I think you went the right way with the vac proving switch, brother.
It hadn't occurred to me that you guys use the same cut files time and again and could code it in.
I seldom (if ever) use a cut file more than a couple times. I'd have to add the code, and I'd forget.
I'd need to do a lot more control work in my environment (if I ran a table vac all the time)

I solved my "forgetting to turn on the dust collector" by putting end switched gates at all my machines.
Then I don't have to run to the dust collector umpteen times a day.
I just needed to go to the trouble because I'm stupid and lazy. :D

[myxpykalix]

"I just needed to go to the trouble because I'm stupid and lazy. :D"

.....​and ugly too......LOL

[bleeth]

Naw-he ain't ugly; just misinformed like all Steeler fans!

[shilala]

But I mean well. :D

[donek]

I think you went the right way with the vac proving switch, brother.
It hadn't occurred to me that you guys use the same cut files time and again and could code it in.
I seldom (if ever) use a cut file more than a couple times. I'd have to add the code, and I'd forget.
I'd need to do a lot more control work in my environment (if I ran a table vac all the time)

I solved my "forgetting to turn on the dust collector" by putting end switched gates at all my machines.
Then I don't have to run to the dust collector umpteen times a day.
I just needed to go to the trouble because I'm stupid and lazy. :D

We almost never execute the same file more than once.

I have written a piece of software (it's all in excel) that allows me to design a custom snowboard based on a set of parameters or variables. This program actually generates all of the CNC code for each board, so adding a line of code to this program is actually quite simple.

The video below gives a bit more background on that.

https://www.youtube.com/watch?v=X3urRYdDGdg&feature=youtu.be