Beiträge von N20net

Hello,

I have previously been impressed by the over-power shutdown feature that I've used on the Shelly 1PM. My understanding is that the Pro 4PM also has this feature. However, I can't find it in the settings for the device. Am I missing something?

When the unit has crashed (see my other thread, about inductive load switching), I have had spurious alert emails, telling me that an overload has been detected, and the unit switched off, so there's definitely something going on. I've also had the same spurious alerts when there has been no crash:

Shelly Cloud - Overpower consumption detected by 84cca87f9380!

84cca87f9380 detected overpower consumption and turned off!

But, there are no log entries.

Thank you for reading.

Excellent. Thank you so much for trying this.

So, we have two solutions to get people out of trouble:

1. RC snubber across the coil.

2. MOV across the Shelly contact.

I'd say the snubber is the better solution, particularly for heavier loads that switch often.

My preference when using controllers with internal relays for inductive, or heavier, loads is to always (where practical) use an auxiliary relay, to protect the relay in the controller. It's much easier to swap a failed relay in a socket than to replace/repair the controller.

Thank you again.

Thank you. Understood. The component values in the Shelly snubber must allow sufficient current to flow to energise the relay. Not hugely surprising, given the low current needed by a relay.

Did you connect at the Shelly, or at the relay? I imagine it would work in either case, but might be easier and neater for people to connect directly across the relay coil.

Thanks again.

These are really useful resources for relay switching and suppression:

https://m.littelfuse.com/~/media/electronics/application_notes/reed_switches/littelfuse_magnetic_sensors_and_reed_switches_inductive_load_arc_suppression_application_note.pdf.pdf

How (not) to destroy a relay | HW-group.com

Hope it helps someone.

Ok. Confession time. What I thought was a 230V snubber (because someone had put them in the drawer labelled 'mains 100 Ohm + 100nF snubbers') is in fact this

https://uk.rs-online.com/web/p/panel-meter-accessories/1959148

Which, on further investigation, is a MOV.

It definitely didn't help connected from N to O, but maybe it would have, if connected directly across the relay coil, at the relay.

However, it definitely did solve the problem across L_x and O_x, at the Shelly.

I'm aware that MOVs deteriorate with every 'firing', but this application has a duty cycle of about once per day for one of them, and about three times a day for the other. So they should last long enough

I will order some Shelly snubbers, next time I buy a Pro 4 PM. But, the MOV across the Shelly output terminals makes for neater wiring!

Now there are two solutions to the crashing with inductive loads. But, I think the Shelly Pro 4 PM would benefit from improved protection in this regard - Industrial environments can be very electrically noisy.

Apologies for posting some slightly misleading information. I'll edit my earlier posts, directing people to look at this one

Zitat

I have just made a test setup with a similar configuration and finder relay on O.

Timer Auto Off and Auto On for 0.5s each for loop

With the original Shelly RC snubber and connection between O and L, it is not possible to switch the external relay. With the RC link between O and N as recommended in the instructions, it works as expected. Without the RC snubber, the Pro 4PM sporadically reboots and freezes

That's very interesting indeed. I wonder why this is. I will investigate further. I am yet to return to my workshop, where I have the details on the snubber I used.

[Edit - Please see the next post in this thread for an update, and more clarity - what I thought was a snubber is actually a MOV]

It's very good that the Shelly snubber works as intended. Clearly it is a different component from the one I used. I assumed (always a bad idea) that it would be very similar.

One question - where you say it's not possible to switch the relay, do you mean that the relay stays energised? That would make some sense, if the current through the Shelly snubber is high enough when the Shelly output is off.

Thank you so much for continuing the investigation.

Hello,

I'll do a sketch later. For now, here's a picture. The snubbers are inside the yellow heatshrink. They are connected between L_x and O_x for any channel driving a relay or contactor. Apologies for the messy wiring.

I'll also post the snubber spec later. They aren't the Shelly snubbers. They're what I had in stock. [Edit - Please see the 13th post in this thread for an update, and more clarity - what I thought was a snubber is actually a MOV]

I'm not sure why the 'official' advice is to connect from N to the output. I've only ever used snubbers directly across relay contacts (or triacs). I did try the official advice. It didn't solve the problem.

Edit: I now realise that I posted this thread in the wrong section. It should be in the Pro 4 PM section, not the 4 Pro section. In my defence, the name's a bit too similar! Mods - could I trouble someone to move this to the right section? Thank you.

Screenshot_20211117-112122_WhatsApp.jpg

Another update. Connecting the snubber across L and the Pro 4 PM output (directly - straight into the screw terminals on the Shelly) seems to have fixed it. I've now performed hundreds of switching operations, with no crashes. So, my theory is EMC from the arcing contacts in the Shelly internal relay was causing the crash. [Edit - Please see the 13th post in this thread for an update, and more clarity - what I thought was a snubber is actually a MOV]

I'll think more about it, and do a (hopefully) final update later.

Hello,

Thank you for your reply. It's the latest stable release. Not the beta.

It very much 'feels' like a back EMF issue. It only happens when the external relay switches off.

1st Edit - Out of desperation (in 15 hours I'm supposed to be installing it) I cut the power measurement tracks. Sadly, no improvement (but random figures for power - usually about 7.4W). I'll now try shorting the cut power measurement tracks, on the side not connected to the shunt.

2nd Edit - shorted the power measurement tracks. Still crashing.. But, now reads 0.0W, so that part works.

This is terrible. If the Shelly Pro can't reliably switch a small 230V relay (Finder Type 40.52) then I'm not sure what use it is. At my customer site there is a 25A contactor waiting to be connected tomorrow. That will probably also cause crashes. The reboots aren't so bad - it comes back pretty quickly - but the white screen crashes don't self-rectify - it stays crashed until a power cycle.

Any suggestions as to how I can solve this? I'm pretty desperate now!

Thank you again.

Thank you again for reading.

Hello,

I have a problem with my Shelly Pro 4. I don't have another to compare it with.

The problem is when switching an external relay. It frequently causes the Pro 4 PM to crash. Sometimes it reboots. Sometimes it just shows a solid white display, and doesn't reboot.

I've tried a snubber connected either at the Shelly end, or at the external relay end. It makes little, if any, difference. [Edit - Please see the 13th post in this thread for an update, and more clarity - what I thought was a snubber is actually a MOV]

I need to install this unit tomorrow. But it needs to be reliable. I've been very impressed with the reliability of Shelly devices, but this unit isn't good.

One thing that occurs to me - could it be caused by the back emf spikes propagating via the power measurement circuitry across the shunt resistors? If so, I would be happy to cut these tracks - I'm only driving relays, so the power measurement is irrelevant.

Thank you for reading. I really hope someone has a solution!

Hello,

I'd like to have a widget that can be tapped to open a gate. I've set an auto off timer of 1s. This works fine, but the widget continues to indicate that the Shelly is 'ON', with the circle blue. To open the gate again I need to tap the widget twice - once to toggle it off (nothing happens at the Shelly) and then again to close the contact for one second.

Is there something I'm doing wrong?

Alternatively, is there a way to send an http request? I need this to operate via the Cloud, though, as the gate needs to be opened when I am away from the local network.

Thank you for reading.

Can the EM show an on/off button at the room level of the app, like the Shelly 1, etc? I ask as I have used the EM to control some external lighting, with a contactor. I want my client to be able to easily turn the lights on and off from her phone, from the same screen as other 'basic' Shellys, without needing to tap on the EM, then press the button. The default behaviour is that 'basic' Shelly relays show an on/off button, but the EM shows the power consumption, and another tap is required to get to the button. I hope this makes sense.

I used the EM, as the load is quite heavy, with a big switch on current surge. A contactor is much better suited to a long life doing this, rather than the little relay inside a Shelly. But, I wanted power monitoring. In this configuration, a standard PM unit would show only the contactor coil load.

Wow! Apologies for the thread archaeology. But wow! I had no idea that the dry contact can't be considered for SELV use, because of that tiny gap. What a massive shame.

I suspect that very few people are aware of this.

Thank you for your great work.

Hello,

Thank you so much for that. A great weight off my mind.

I appreciate your taking the trouble to let me know.

Hello,

I would like to use one channel of the Pro 4 PM to open an electric gate. I hope to connect it to a relay, with the relay contact connected to the gate controller.

On the Shelly 1 that I have previously used for this, I could set it to switch off automatically after one second. Is the same function available in the Pro 4 PM. I really hope so, as I just ordered one.

I really want to get away from WiFi.

Thanks a lot for reading.