PLC guidance. Adding a vibration sensor to a CUP 15

so…the warrantee is over on our CUP 15.
I’d like to add a vibration sensor, because I’ve seen what happens when they are unbalanced and the operator wanders off.
Delta says they can’t help.

seems to me that adding a normally closed relay that can be tuned to open upon receiving a 4-20mA signal from a vibration sensor could simply be added in series with the emergency stop button. If the relay opens, it triggers an emergency stop.

assuming this works, it’s a trivial addition to any of y’alls fuges. best guess, I’m looking at ~$400 for the sensor and relay.

just not quite sure what I’m looking for as far as relays go.

any pointers appreciated.

@greenbuggy @that445guy @Soxhlet @SidViscous

4 Likes

how many g’s are we tripping at dawg? :nerd_face:

2 Likes

Wasn’t planning on measuring in g (but willing to look at it).

Pretty sure this critter or something like it will work for the 4-20mA signal (measures velocity?).

actually looks like STI makes vibration sensors with the relay function built in…

https://www.stiweb.com/product_p/cmcp423vta.htm ($395… love it when I guess right :wink: )

looks like it can be field adjusted (so it panics early enough to save the driveshaft).

guess I should get STI on the phone…

3 Likes

…and then there is the VFD, on the off chance that Delta was incorrect about it having an input that I could use directly.

image

1 Like

Are you looking to have the vibration sensor relay just cut the drive out by installing the vibration sensor relays n/c contacts in the run circuit of the drive? Depending on the drive you should be able to set it up so that when the vibration sensor is tripped it immediately starts putting the breaks on. Depending what you have for a pic and what kind of inputs you have available you could program a control loop that would very the the drive output based proportionately on the analog output of the vibration sensor and then use the adjustable dry contact set on the vibration sensor as a fail safe in the run circuit of the drive.

2 Likes

I know Fluke makes vibration sensors for preventative maintenance but they’re hella expensive. I would imagine omron or Siemens makes an actual safety device for this kind of application, I’ll take a look when I get in front of a computer.

If you did a functional safety assessment you’d probably be in the plC or plD range so redundancy with the drive signal is probably merited. Something like a safety rated relay timer to cut power to the drive after the braking time has been exceeded in case the drive fails.

Also, avoiding any analog devices in safety applications makes life about 1000x easier. I’m not sure a vibration sensor/accelerometer is the easiest/best way to do this but I’m sure someone has had to make a SIL-3/plD centrifuge balance control system before. I’ll take a look through my literature to see if anything looks like an easy retrofit.

Honestly, I can’t believe Delta didn’t build the thing with some sort of functional safety system in the first place but then again industrial equipment in this industry is like 60 years behind any other industry

3 Likes

So I just did a quick look and this looks like the correct solution: https://www.pch-engineering.dk/404/compact-vibration-monitors

It even comes in the ATEX variety.

Unfortunately it’s only SIL-2 but at least it’s got a MTTFd rating. I’d run it through one of the safety rated S7-1200 and send the stop output to the drive, then stop sending signal to the safety rated contactor/motor starter after the braking time has been exceeded. Probably also makes sense to tie in your other devices to keep power isolated to the drive to prevent things like accidental start up

This solution would probably be about $2500 in hardware new but the contactor and PLC could be found a lot cheaper from factory uninstalls on eBay and get down to around half that. A bit more than the $400 option but if you’re going through the effort it’s probably worth it. Like you said, you could probably skip the PLC if all you want is the vibration sensor, but you’ll definitely want to use the drive to stop if possible and then backup with positive power interruption which would require at least a timer relay.

2 Likes

Yeah, that’s probably how it should be implemented… I’d absolutely prefer to slow the machine if it sensed unbalance, maybe give itself a little shake the way the LG washing machines do to deal with unbalanced…but PLC is still pretty new to me and I’m dealing some one else’s PLC programming and HMI.

I’m suggesting putting the NC relay built into the unit above, in the same 24V circuit that the emergency Stop button on the operator’s panel interrupts when the operator panics. I believe this will at least give the machine a fighting chance if the operator is otherwise occupied when things get a little shaky :shushing_face:

I haven’t had a chance to read through the docs for that sensor to figure out how a reset is
Performed.

1 Like

Knew you’d have a better solution. Agreed with the part about using active braking, then cutting the power. I’d like to imagine that’s how an emergency stop triggered by the big red button on the hmi works. I’ve determined that on more that one centrifuge, some do, some don’t. I still need to figure this one out.

Triggering that emergency stop by breaking a contact in series with the actual big red button seems like it should perform the same ritual.

Given that there is actually a second big red button on the “power console”, the addition of a third panic button that the machine can operate seems like the expedient solution.

I promise I’ll try and wrap my head all the way around your response.

It’s not like I have to fabricate a mount, or drill a hole in the HMI panel…

image image

Edit: actually, I might need to rotate that mount. It will give me a chance to address the hinge…

2 Likes

I appreciate the vote of confidence lol.

I’ll try to not nerd out too hard about functional safety but it’s one of my favorite subjects.

Strictly speaking, HMIs should never be used as safety devices because they not only rely on non-safety hardware but they usually run on some sort of windows OS and we’ll, yeah I need not say any more.

The issue with the e-stop button is usually this: there are 3 types of e stop.

Type 0 is just a power cut that latches. It is like 95% of e stop devices.

Type 1 is a controlled stop. This is what we want here. It’s common for things like saws or other things that spin fast and need to be slowed before power can be shut off.

Type 2 is a powered e stop. This is what we run for our CO2 extractor and should really be used on any large cryogenic system. These are used for things like nuclear reactors where you need to maintain instrumentation to keep making safety decisions and allows the system to triage. They’re very expensive lol.

I would imagine the e stop on your CUP is a type 0. What that means is the drive is just going to lose power, which is kind of the right thing because unless you buy a mega expensive drive, it can’t be depended on for critical safety functions. But, without the clever design that we’re going to come up with here, if you just cut power you lose active braking.

It’s pretty cut and dry if you do a risk assessment that the risk of getting electrocuted from relying on the e-stop fallaciously to isolate power inside a sealed panel is WAY less than throwing your fuge across the shop because your wooks decided to dry their boots in it (hopefully not but I like colorful examples).

The good news is converting to type 1 from type 0 (while it will probably void your warranty) is fairly simple in this case. There are two options and I really like the PLC option but it’s maybe overkill here depending on how much work you want to do.

The non PLC option is this:

*Change the wiring through a safety rated delay-after-break relay timer and set the time for the programmed braking time on the drive so that the relay stays powered but the signal is broken when you punch the button. Two channels here (you’ll need to change to a 4-channel e-stop button).

*Use the other two channels to break the e-stop circuit (I’m sure it exists on that Lenze drive, I can help locate it if you’d like).

That handles the button problem. Next, you would need to add the sensor to make it the same as hitting the button. This means using two safety relays, DPDT or DPST NO that are regularly powered by your vibration sensor. Using two separate sensors here is actually one of the few cases where it might turn your SIL-2 up to SIL-3 although working out channel monitoring (which will trip the system if feedback is lost) is a little tougher and might take auxiliary contacts on those relays. This is where the PLC option shines.

The issue though is that those relays need to “latch” otherwise you’re depending on the drive not starting back up which is silly. Doing the physical wiring to make a NC circuit latch isnt tough and a quick Google search would probably spit it out. I’ll probably work up a diagram this weekend because I think this a fairly simple schema that might make a lot of people’s fuges, Chinese or otherwise, a shitload safer.

The PLC option is basically draw the same thing in TIA portal and just plug everything in lol.

I feel like one of the forum vendors could sell a sweet little panel to plug and play this, I can work up the schema and BOM if there’s interest.

Edit: sorry for the book of a post lol

4 Likes

Thank you!

For all the scary shiny spinning things.

2 Likes

Having a bit of a hard time with that photo. I can see that it’s a i550 but that’s about it. A more complete model number would be helpful

that might require demounting it. which might get in the way of production if I get it wrong. I’ll see what I can manage tomorrow. thank you!

Nope, same picture just a little less blurry should be fine

2 Likes

Perfect

sooo…for anyone following along. the STI device listed above resets automagically, and would need to be paired with a latching relay to actually be useful.

STI suggested this (~$700) device instead. https://www.stiweb.com/CMCP_DVS_Digital_Vibration_Switch_p/cmcp-dvs.htm

2 Likes

:cry: Sad but true…

We have all of this standard on our equipment.

Centrifuges can cause WAY more damage than a solvent explosion

2 Likes