Mechanical Relay Quality

[evilkitty]

Re: Mechanical Relay Quality

I have ~600 state changes in my logs 11/26/21 - 1/31/22
this include both relays, i have sanity checks to detect bad data to prevent random state changes, eg if the temp sensor goes from reporting say 20c to 80C in 2 seconds i call bs and read the senor again

edit: just updated the git repo with my current script reversion

[eccerr0r]

Re: Mechanical Relay Quality

600 seems quite low before failure. So it did only ~10 states per day, i.e. 5 on and 5 off per day, for 2 months?
Hmm...
how are you driving your coil? DC? Full rated current?

[stj]

Re: Mechanical Relay Quality

maybe it's time to think about cheap SSR's or triacs, but i dont know if triacs will work at that low voltage.

[evilkitty]

Re: Mechanical Relay Quality

600 seems quite low before failure. So it did only ~10 states per day, i.e. 5 on and 5 off per day, for 2 months?
Hmm...
how are you driving your coil? DC? Full rated current?

the contact failed on the coil powered off state 1st, so i am not sure that the could voltage it the issue, but it is running with 5v -transistor vdrop, but you can see how it was done in this image


"600 seems quite low before failure. So it did only ~10 states per day, i.e. 5 on and 5 off per day, for 2 months?"
yes, though we need to extrapolate time cause it was close to 4 or 4.5 years IRC, but that would be under 15k changes or 7500 full cycles and that includes cycles for both relays

also worth noting that this relay did not have the snubber for probably a week while i wanted on parts

the bottom relay has around 6 years on it, 2 years without a snubber and is still working, but only has 1 side of the relay (coil powered on) that does anything and it gets far less yearly cycles as it does not get used with the AC is active

[evilkitty]

Re: Mechanical Relay Quality

@stj, are double throw SSRs a thing?

[stj]

Re: Mechanical Relay Quality

no - atleast i dont think so.
using 2 and inverting the control signal on the second will work though.

[evilkitty]

Re: Mechanical Relay Quality

i guess i could use a NPN and a PNP transistor and control them both with the same gpio pin each getting there own resistor, guess i should make use new pcb for this design, id need to take some parts off the old pcb, but they are easy ones to remove

[eccerr0r]

Re: Mechanical Relay Quality

Actually using a TIP120?! This might explain the failure.

How much drive current can a rpi source/sink? If it can drive more, you're better off not using a Darlington there, at least for 5V. Maybe a 2N4401 would be better.

[redwire]

Re: Mechanical Relay Quality

The old relay contacts are not pitted from arcing, but the plating has vanished and gone somewhere. There is no black soot so I'm puzzled what is going on. It might be chinesium plating.
Songle SRD-05VDC-SL-C coil is 70R or 71mA and they offer three different contact platings and current ratings 7,10,15A 250VAC as well as basic 10,000 cycle lifetime into resistive loads.
UL/CSA rate them only 3A 125VAC inductive loads, and there is a secret long life 100,000 cycle model. Thne specs are a bit deceptive and I've seen plenty of Songle knockoffs too.

Driving the relay with a TIP120, the base resistor is large 6k8 so the TIP120 sees about 0.2mA drive with hFE 500 so it is on a bit weak.
Changing to 2k gives 1mA drive from the RPi. The TIP120 loses at least a volt so the relay at best will see 4V.
If you drive the relay "weak" it pulls in slow and wimpy which would wear the N.O. contact only. The N.C. relies on the spring for force.

This is why I think the software might extra pulse the relay, or the snubber is simply too small for whatever the load is, or the relay quality is poor.

[stj]

Re: Mechanical Relay Quality

you dont need a transistor for an ssr, they use an optoisolator and take under 5mA usually.

[evilkitty]

Re: Mechanical Relay Quality

you dont need a transistor for an ssr, they use an optoisolator and take under 5mA usually.

if it uses a 5v control i do need a transistor, my GPIO pins run at 3.3v

[evilkitty]

Re: Mechanical Relay Quality

Actually using a TIP120?! This might explain the failure.

How much drive current can a rpi source/sink? If it can drive more, you're better off not using a Darlington there, at least for 5V. Maybe a 2N4401 would be better.

The old relay contacts are not pitted from arcing, but the plating has vanished and gone somewhere. There is no black soot so I'm puzzled what is going on. It might be chinesium plating.
Songle SRD-05VDC-SL-C coil is 70R or 71mA and they offer three different contact platings and current ratings 7,10,15A 250VAC as well as basic 10,000 cycle lifetime into resistive loads.
UL/CSA rate them only 3A 125VAC inductive loads, and there is a secret long life 100,000 cycle model. Thne specs are a bit deceptive and I've seen plenty of Songle knockoffs too.

Driving the relay with a TIP120, the base resistor is large 6k8 so the TIP120 sees about 0.2mA drive with hFE 500 so it is on a bit weak.
Changing to 2k gives 1mA drive from the RPi. The TIP120 loses at least a volt so the relay at best will see 4V.
If you drive the relay "weak" it pulls in slow and wimpy which would wear the N.O. contact only. The N.C. relies on the spring for force.

This is why I think the software might extra pulse the relay, or the snubber is simply too small for whatever the load is, or the relay quality is poor.

i got 10 of these relays in 2015 for $3.92 on ebay, so about 0.40 each, not sure if that is a realistic price for them

I have some 2N2222A NPN and 2N3906 PNP transistors on hand and a wide assortment of 1% 0.25W metal film resistors

rpi gpio is specked at 16ma max, however there is also a shared total max as well, i think that is 50ma, not sure if sinking and outputting both count towards this total
my LCD panel uses alot of pins, but the total ma should be silly low as i was trying to make the led a bit dim
front panel:

edit: i checked the voltage across the coil and got 4.14v on the bottom relay and 4.13 on the top relay
EDIT: took some pictures of the labels on the outdoor unit: https://imgur.com/a/809clTp they are not in great shape, but here they are

[eccerr0r]

Re: Mechanical Relay Quality

The NO contacts get a weak make but the break works fine due to the spring.
The NC contacts get a weak break but the make works fine due to the spring.

Well it's something to consider in any case. You could use 2/6 of a 74HCT04 to make a buffer so the RPi output isn't as loaded to drive something you can use as a saturated switch. Or even drive a logic level MOSFET,... Would be nice to direct drive a MOSFET but a 3.3V signal limits the number of MOSFETs available without some sort of buffer.

I really think that TIP120 is a bad idea for this situation, if you were using a 12V relay and supply then the TIP120 may do much better.

[evilkitty]

Re: Mechanical Relay Quality

What about using this relay, it calls for 4.5v and says it will work as low as 3.2v
https://www.mouser.com/ProductDetail/817-SY-4.5-K

If this issue is the TIP120 would you expect both sides to wear like this? the NC (coil on state) wouldn't that one not look like this assuming the spring was not worn out, this contact does not look good, but it was not dead

i do not know for sure how much voltage the contacts are dealing with, i just know it is low amps based on wire gauge

[eccerr0r]

Re: Mechanical Relay Quality

404

[sam_sam_sam]

Re: Mechanical Relay Quality

Those relays ( if they are 24 ac ones ) are cheaply made in the first place and if your thermostat cycle to many times will wear them out in 3 to 5 years of use any way unless you get the heavy duty ones ( cost a lot more money ) which are hard to find only some Air Conditioning Supply House carry them and most of them require a journeymen license to buy from in the first place

These relays are mostly used to control fan speed between heating and cooling and sometimes used for heating element relays that are made for switching the heating element on and off if there is more than one heating element bank

[evilkitty]

Re: Mechanical Relay Quality

https://www.mouser.com/ProductDetail/817-SY-4.5-K

[sam_sam_sam]

Re: Mechanical Relay Quality

https://www.mouser.com/ProductDetail/817-SY-4.5-K

You can buy 3.3 volt solid state relays that run at this low of voltage you just have to search for them and if I was doing this I would highly recommend going this route

[eccerr0r]

Re: Mechanical Relay Quality

Well, it seems a full amp can be passed through the circuit, so a 1A relay is pushing it.

Hmm... have to see what my thermostat is using now... I'm surprised it's not using a SSR, it too uses 3V logic (uses two AA batteries and probably residual power from the 24V control line)...

[evilkitty]

Re: Mechanical Relay Quality

Using my meter it showed ~32ma at ~27v AC

if i were to use 3 SSRs i would have to redo the pcb layout, but if i can sick with 2 relays it i can just swap them out and garage some wires at the relay for a new pin out/location
maybe i could use this relay: https://www.digikey.com/en/products/...1Z-4-5V/648227