Piezo switch problems?

[Western]

Hi, I am trying to remotely fault find an issue for work.

Apparently it’s been going on for a couple years and lots of effort applied to solve ... but no results yet and is getting serious.

There are some simple control boxes that are switched by a piezo switch ... and work well ... but occasionally the devices start up by themselves shortly after going through their normal operation.

Seems to be no rhyme or reason ... maybe 6 out of the 50 will do it ... other times only one or two ... and not always the same ones.

Can piezo switches ever trigger themselves? I don’t believe so ... but something’s switching these boxes back on!!!

The switches have to pull a gate to ground (from 4.2V), with a pulse of at least 100mSec (set by software) ... so I’m assuming they have some sort of silicon in them.

They are not overly sensitive to press ... you have to apply a reasonable amount of pressure to make them operate.

They are rebranded with a generic part number, so I don’t have a datasheet ... but have 24V AC/DC 0.2A printed on a tag.

Thanks for any help.

 

[rjenkinsgb]

My first thought would be some form of grounding or static / interference problem.

 

[Western]

Thanks for that suggestion ... I hadn't really thought along the lines of grounding or static at all. I'd certainly considered interference ... but had ground to a halt with that.

The units are inside plastic boxes with two of these piezo switches on the front. The boxes are screwed to a metal plate which is clamped to a common metal frame.

They have a 24V DC supply and switch a couple of vacuum solenoids. None of it really needs to be earthed ...

... so I suppose a charge could build up during use or brushing past any of the fixtures.

I think the main reason I hadn't considered it was because they're in a very damp environment ... but it's definitely worth considering.

I'll work out some questions to ask.

Thank you very much again for the suggestion.

 

[Western]

I've had some more thoughts on the static side of things ... and been playing around trying to generate plenty

I could see static resetting the micro perhaps ... which would put it through the obvious start-up sequence ... which is not happening here.

If static was getting in through the piezo switch ... the switching pulse needs to be at least 100mSec for the board to 'read' the pulse. I've been monitoring the switch input and can get some fairly decent pushes on the switch to not read because they are not long enough.

I had considered asking the guys to change the settings in the software to increase the required switch pulse to the maximum of 250mSec ... which may eliminate any spurious pulses ... but they're unlikely to be anywhere near 100mSec let alone 250. Still worth a try on the worst offending units though in case it does improve things.

The operators will hate it though ... because it will require a much longer press than what it is now.

 

[Nigel Goodwin]

Can piezo switches ever trigger themselves? I don’t believe so ... but something’s switching these boxes back on!!!

Without knowing EXACTLY how they work it's impossible to tell, but any complicated, high gain, electronic switching method is never going to be as reliable as a mechanical switch. Inadvertent triggering could occur for load's of different reasons.

 

[Western]

I do agree Nigel, though in this setup the moisture and dampness kill normal switches very quickly ... that's why they use the piezos.

The guys who installed this setup also did another one quite close by ... and that has had zero of these issues, despite using the same gear.

As rjenkinsgb suggested, it sounds more like interference ... but no-one can pinpoint anything.

 

[alec_t]

in this setup the moisture and dampness kill normal switches very quickly

In view of the very high impedances involved in the switch, moisture could also be a problem for a piezo switch if the IP65 protection is compromised at all.

 

[Beau Schwabe]

"the moisture and dampness kill normal switches very quickly ... that's why they use the piezos. " ... Why not use an inductive proximity sensor that is completely sealed ?

 

[Western]

In view of the very high impedances involved in the switch, moisture could also be a problem for a piezo switch if the IP65 protection is compromised at all.

Yeah, I could understand moisture getting into them and breaking down either the insulation or the semiconductor sections ... but I would have thought they would simply become harder and harder to press ... or simply not fire at all.

... and as above ... there's a second (supposedly) identical plant not that far away that has never seen this issue. Certainly, they have the same control boxes, piezo switches and solenoids etc.y

But having said that ... it's helped me think of a couple more questions to ask the people on the ground over there.

Thanks for your input Alec.

 

[Western]

"the moisture and dampness kill normal switches very quickly ... that's why they use the piezos. " ... Why not use an inductive proximity sensor that is completely sealed ?

I'm not familiar with them at all ... I'll do some research. Thank you.

Having said that, I'm not at all sure that switches are our issue here ... but I just got off the phone ... and they are going to fit normal mechanical switches to a couple of the worst offending units temporarily ... to see if the issue goes away.

I am starting to think that rjenkinsgb was on the ball, when he suggested static ... but I hope to get some feedback in another 24 hours or so.