AC Induction motor relay RC snubber values of R and C

[BlackICE]

I want to protect the contacts of a rated 30A relay that will be switching an 120VAC induction motor with a nominal current draw of 10A RMS at rated load. I measured the starting current of the motor and it was 48A RMS for 0.28 sec. Would that overload a 30A rated relay? I have read about these RC snubbers that limit the arcing damage from opening the contacts. The R value is pretty strait forward, but what is the best value for the cap and the type of cap to use. I have some left over 5uf aluminum electrolytic caps used for starting caps on HVAC fans would they work. Pluses I already have them but they are very big, or would a film cap like this one be better choice?

https://www.digikey.com/en/products/detail/panasonic-electronic-components/ECQ-E2475KB/295894

https://www.homemade-circuits.com/prevent-relay-arcing-using-rc-snubber-circuits/

Would this relay work given the higher starting current for 0.28 secs?
https://www.amazon.com/ABXLNIU-Miniature-Power-240VAC-NT90-DC12V/dp/B09JKFDPMV/ref=sr_1_30

or this relay
https://www.amazon.com/Relay-Module-Optocoupler-Isolation-Trigger/dp/B07TVDD3VT/ref=cm_cr_arp_d_product_top?ie=UTF8

I looked up the datasheet for this one and it is motor rated the 1HP motor I am using, but only for N/O, but only 0.25 HP for a N/C contact. Photo included.  I hope a snubber would raise the HP rating on for N/C use. I would be cycling on and off with 2 different 1HP motors on either side of the SPDT relay about once every 5 mins.


I have a larger 80amp relay but that much bigger and trying to keep it small.

[johansen]

https://www.eevblog.com/forum/projects/stored-energy-induction-motor/msg4769549/

For the relay those cubic inch sized 30amp, 1hp rated relays should last a long time.

[Benta]

There's normally no need to snubber AC loads on a relay, arcing is short and self-extinguishing and does little damage.
DC switching is a different story, snubbering makes sense there.

[floobydust]

OP I think you need two relays because the N/C contact is always lower current-rated due to the lower contact pressure (spring) and speed.
No amount of snubber will change that. 5uF is way too big for that, toss them aside. The T9A family might work though, pic of it's ratings. Also, on your PC board withstanding 50A inrush does require extra copper and fat traces so they don't fuse.
I always incorporate snubbers with relays switching inductive loads, theory aside you can simply see the reduced arcing with the naked eye.
Less metal getting evaporated off the contacts is always a good thing for product lifetime, as well as reduced EMI for any nearby microcontroller.

Motor ratings for relays are under UL 508/873 and they are far from the resistive load specs. Be careful about the lifetime specs for your 5min. cycling.

[BlackICE]

I ordered 2 of these from Amazon. Claims to support 1.5HP 250V motor on the NC, but I'm using a 1 HP 120V (=2HP @ 240) motor so not quite a fit.

Both sides of the relay are powering about 1HP induction motors that are running mutually exclusive. The NC motor runs a 1HP air compressor, the NO motor runs a 1HP motor connected to a 40:1 gear reduction drive moving a high pressure bicycle style pump up to 300 bar! The pump runs a about 43 RPM and for less than 90 degrees of the pumps rotation is under load. Most of the time the NO motor has minimal load on it since the pump has high resistance for less than 1/4 (probably only 1/8) of its rotation, or 0.35 sec out of 1.4 secs. The reason I monitor the current is to get most output from the pump and motor, I vary the input pressure to the bicycle style pump until I reach over than motors rated current of 9.8A (got to 14.8A) without stalling the motor. In the beginning when the output pressure is low the input pressure can be high, as the output pressure rises the input pressure is reduced to keep the peak motor current below where the motor would stall.

Since the project is already using a micro controller to monitor the current going through the relay, I could add a zero detect on the AC line and try to time closing the contacts at a zero crossing when closing a contact. Likewise I can use the current level to time the opening of the contacts to when the current is approaching 0. Off course this assumes when I measure the time it takes to open and close the contacts, that time remain reasonably constant.

When switching from NC to the NO, the NC motor will always be off, but when switching from the NO to NC motor I can't insure that the NC motor won't be active. So it is impossible to do both closing on zero voltage detection and opening on zero current detection. Unless I use 2 relays. Which is more important for relay lifespan opening the contacts at near zero current, or closing the contacts at zero voltage?

Worst case this will cycle in 10 min. intervals up to 20 times twice per week. So contact lifespan of over 10,000 (5 years) would be enough for me to call it OK.

https://www.electrokit.com/uploads/productfile/41015/NT90.pdf

[floobydust]

I've seen data (with power transformers) that switching on at zero-cross gives the highest inrush current possible. It's because dV/dt is highest at sine zero-cross, compared to switching in at peak voltage. Synchronous switching - that relay's switching times are around 15-25msec and I find not consistent with temperature, age. It would complicate F/W so I would not bother. If you use two relays, leave some deadtime.

AC motors are not simple inductors, they have energy storage in the moving rotor, run cap which can really bite back. So I always use a RC snubber starting at 0.1uF/33R 1/2W and working from there. Simply build it and see what what you get verses theory which is too complex for the armchair.
I'd read that the arc heat and duration are what wears contact metals. There are many different contact alloys.

Placing a back-EMF diode across the relay coil slows its release time 2-5x. It takes much longer for the field to decay, detrimental to contact life and not mentioned on the datasheets. Consider using a diode+zener 2-3x the relay's coil voltage spec, most driver transistors can take that.
I'd put the N/C contact on the larger load.