Loud buzzing from power relay - need advice

[atouk]

I powered up my control panel for the first time today and was greeted with a very loud buzzing from the relay that handles main power on/off. We're talking loud like I had to put in earplugs to troubleshoot.

My panel is a 2 element, 30A system. I used Kal's system as a basis for my wiring plan. Both legs of the power line connect in to the relay, one leg runs out directly to the element relays and the other leg to the SSRs. I use a key switch to turn it on and off. A quick sketch is attached.

The whole panel is getting power (PIDs, pumps, etc). The element relays are the same type and when I turn them on, I get the same buzzing. Also, I can see that my element on light flickers a bit.

Seems to me (from my zero electrical experience) that the noise is coming from the contacts inside the relay opening and closing really fast, but I have no idea why this could be.

Here is a link to the relays I am using. I did go cheap, so perhaps I am paying for that...

Any advice is appreciated!

http://www.zoro.com/magnecraft-enclosed-power-relay-30a-120vac-dpst-92s7a22d-120a/i/G2290127/

 

[RufusBrewer]

Throwing out things to check:

Is it possible you ended up with a relay with a DC rated coil?
Is the relay coil rated for the correct AC voltage?
Are you sure the two pins shown on the bottom of your relay the coil input and not relay contacts?
Do you have a diode across the coil as shown in the link?

 

[atouk]

Well crap. They send me relays with 12v DC coils! Not sure how I missed this, but I guess that would cause the buzzing? Is there any way I can still use these without major changes to the panel?

 

[GotPushrods]

Well crap. They send me relays with 12v DC coils! Not sure how I missed this, but I guess that would cause the buzzing? Is there any way I can still use these without major changes to the panel?

Sure, your switch could feed 120v to a 12v wall-wort internally. But that's probably more trouble than just using the correct relay.

This is pretty interesting, I never thought about what would happen feeding 120 VAC to a 12 VDC coil. Maybe the freewheel diode causes a longer pause making the chatter louder? I'd love to hear it!

 

[RufusBrewer]

I'd love to hear it!

ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt ZZZzzztttttt

 

[atouk]

^^^^^^
That's about right. Just imagine taking 10,000 angry bees, placing them in an empty kettle and then placing your head inside said kettle. The noise was that loud and just about as painful (I'd imagine)

Thanks for the help! Time to exchange some relays.

 

[RufusBrewer]

Maybe you can keep them. Never know when you can use a 12 VDC relay.

 

[atouk]

So they're giving me a refund but I don't have to send the relays back. That's pretty good customer service considering I ordered these back in March. New AC relays are on the way!

 

[atouk]

Update. New AC relays arrived, installed this afternoon and now the panel is perfectly quiet. What's more, it works! One step closer to brewing!

Thanks again for the help

 

[Rockn_M]

Update. New AC relays arrived, installed this afternoon and now the panel is perfectly quiet. What's more, it works! One step closer to brewing!

Thanks again for the help

That's an awesome feeling, when you turn something on that you built yourself and it works!

 

[TheMadKing]

Just an answer to "why this happened."

Running AC into anything designed for DC will make a buzzing sound if there's a moving part in there. DC is just a smoothed out "positive" AC current. Think of AC as a sine wave where the peaks are positive and the troughs are negative. DC is just all the negative chopped off and the remaining positive pulses stitched together so that there's no gap between them.

So when you run AC into something looking for DC, it only sees the positive part of the sine wave, and since the wavelength is 60Hz (the distance between peaks), it should vibrate at exactly 60Hz because its essentially getting pulses of DC power at that frequency... if that makes any sense I'll be amazed :cross:

 

[ajdelange]

So when you run AC into something looking for DC, it only sees the positive part of the sine wave,

Not so. A DC relay is the same, broadly speaking, as an AC relay i.e. a magnetic core with wire wound on it (the DC relay may contain a back biased diode to absorb the reverse voltage generated when the coil is deenergzied and the number of turns of wire is different). What is probably causing the loud buzzing here is the magnetosrtrictive effect as the 12V relay is going to have about at least 10 times as many turns as a 120V relay. When you throw 170 V at something designed for 12V lots more current flows. The core certainly saturates reversing direction 120 times per second. Thus the buzz is at 120 Hz (somewhere between B-Flat and B 2 octaves below middle C). I'm surprised the thing didn't just burn out. If I'd sold them I wouldn't take them back knowing what happened to them. Guess the resistance of the coil wire was enough to protect them.

 

[passedpawn]

To continue the pedantry, because I apparently don't want to get to work this AM...

Yep, 120 Hz. Relay is chattering. AC coils have a shading coil that prevents them from "chattering when the AC current goes through zero. This shading coil is a second coil that gains and releases its magnetic field out of phase with the main coil. What this means is that between the two coils, there's always some magnetic field and the relay doesn't chatter.

DC relays don't have a shading coil since the current isn't switching through zero. So if you drive a DC coil with AC, it's going to chatter at 120hz. Though, with a DC relay, I would always add a flyback diode in parallel with the coil to limit high voltage (+/-) transients during switching. These transients create hella RF noise that can affect local electronics, even if they aren't connected to the relay ckt. With this diode in place, you'd have 60Hz clicking if you connected AC.

BTW @TheMadKing, Hz is frequency, not wavelength. The wavelength of a 60Hz sine wave is about 3100 miles ( electricity travels at the speed of light, which is about 186,000 miles per second. So, wl = 186,000 / 60.

 

[TheMadKing]

I was 100% incorrect about the 60hz and saying wavelength instead of frequency. Thanks for the fix guys! I'll be more precise next time, and more accurate

 

[ajdelange]

To continue the pedantry, because I apparently don't want to get to work this AM...

Yep, 120 Hz. Relay is chattering. AC coils have a shading coil that prevents them from "chattering when the AC current goes through zero.

Yes, I think I need to change my mind about magnetostrictive effect (though there will still be some noise from it in the case of a relay - that's why transformers hum with no moving parts).'

This shading coil is a second coil that gains and releases its magnetic field out of phase with the main coil. What this means is that between the two coils, there's always some magnetic field and the relay doesn't chatter.

This is the first time I have heard of this though I've been an electrical engineer all my life (though only involved in industrial controls for a couple of summer jobs in distant youth). Anyway, the important thing here is to realize that the shading coil is, like the shading coil in a shaded pole motor, around a portion of the core. Thus, in a typical solenoid type relay the pole is split into two 'D"s with the shading coil a copper ring around one of them. The current induced in the shorted turn is in the direction which opposes the change in the linking flux. Thus when current in the main coil is increasing during the first quarter cycle and the flux in the main part of the core is increasing with it the flux in the shaded core in increasing at a slower rate and so lags the flux level in the main core. When the the flux in the main core is decreasing, during the second quarter cycle, the current in the shading turn again opposes the main coil current change so that the shading coil boosts the flux in the shaded part of the core and its flux is greater than in the main pole, again lagging it. When the coil current goes through 0 the main pole flux goes through 0 (ignoring hysteresis here) but the shaded pole's flux, lagging, still has finite value. Thus the shaded pole continues to attract the armature though the main pole does not. By the time the shaded pole flux decays to 0 the main pole flux has started to build up in the opposite direction for the third quarter cycle.

With this diode in place, you'd have 60Hz clicking if you connected AC.

Actually 120 as the diode would be removed from the circuit during the first half cycle that forward biased it. Put it in series and you'd get the 60 Hz buzz.

So, wl = 186,000 / 60.

Miles!

Even longer wavelengths have been used to communicate with submerged submarines (and perhaps still are being so used). The bit rates are low.

 

[passedpawn]

Even longer wavelengths have been used to communicate with submerged submarines (and perhaps still are being so used). The bit rates are low.

Talk about low bit rates, you should look up the UTC time broadcast data format (WWVB) from Ft Collins. It's the RF data stream that some clocks sync to. It's 1 bit per second. The signal used to be super-simple AM PWM, but now PSK was added and the new data is packed more efficiently.

[edit] oops on the diode, you're right, no rectification with the parallel diode.

 

[TheMadKing]

Anyway, the important thing here is to realize that the shading coil is, like the shading coil in a shaded pole motor, around a portion of the core. Thus, in a typical solenoid type relay the pole is split into two 'D"s with the shading coil a copper ring around one of them. The current induced in the shorted turn is in the direction which opposes the change in the linking flux. Thus when current in the main coil is increasing during the first quarter cycle and the flux in the main part of the core is increasing with it the flux in the shaded core in increasing at a slower rate and so lags the flux level in the main core. When the the flux in the main core is decreasing, during the second quarter cycle, the current in the shading turn again opposes the main coil current change so that the shading coil boosts the flux in the shaded part of the core and its flux is greater than in the main pole, again lagging it. When the coil current goes through 0 the main pole flux goes through 0 (ignoring hysteresis here) but the shaded pole's flux, lagging, still has finite value. Thus the shaded pole continues to attract the armature though the main pole does not. By the time the shaded pole flux decays to 0 the main pole flux has started to build up in the opposite direction for the third quarter cycle.

I've been working as a technician on a control system that includes relays for years and I've never had this explained better or more clearly. I usually go with the "out of phase" explanation as its easier to visualize. I've never actually understood it fully until now.

Just goes to show that the "how" and the "why" of things aren't always packaged together. I certainly know the how, but the "why" was never quite as clear. Thanks AJ!

If a forum ever needs an explanation on particle physics or neutron kinetics I'm your guy, but until then I'll start keeping my mouth shut and learning

 

[ajdelange]

Talk about low bit rates,

There's always Verizon's 'LTE' service in northern Vermont.

...you should look up the UTC time broadcast data format (WWVB) from Ft Collins. It's the RF data stream that some clocks sync to.

Yes, I've done that as my wristwatch and a couple of wall clocks at home use it. I can't receive the signal here. It's only about 10 kW (IIRC) out of Fort Collins (again IIRC). I thought it would really catch on as the signal has been designed for easy reception but it doesn't seem to be too popular.

 

[ajdelange]

If a forum ever needs an explanation on particle physics..

I do have questions on particle physics or one anyway: "WTF, over?". I wouldn't know how to formulate a second question.

 

[TheMadKing]

I do have questions on particle physics or one anyway: "WTF, over?". I wouldn't know how to formulate a second question.

haha, given your displayed level of knowledge on every other subject, you could get the hang of it in seconds. I work at a nuclear research reactor at a university, and teach reactor theory to undergrads, if I can do it, anyone can!

 

[ajdelange]

Between red, green, blue, up, down, heavy, light, charm, Bosons, Fermions, leptons, mesons, muons, pions..... I get lost pretty quick. There's more of the damn things than Republican presidential hopefuls. Not to mention that my relativistic quantum physics is nothing to speak of.

 

[TheMadKing]

I don't deal with much at the quantum level, anyone in their right mind stays away from that spooky stuff, I keep my feet on good solid tested and proven physics, and about the smallest I get is neutrinos. No exotic particle detectors here!

 

[BrunDog]

Wow, I'm just a dumb ME. Makes me wanna RDWHAHB!

-BD

 

[TheMadKing]

In my experience there's no such thing as dumb people, just people specialized in something different from me, but RDWHAHB is never a bad option :beer:

 

[ajdelange]

As if the gods felt that I didn't really and truly believe in this shaded pole business they granted me an 'opportunity' today when I noticed that the fan on my air handler would not go off no matter what I did (short of killing the breaker). Sure enough there is a relay in there with a welded contact but that's not all. It has D-shaped pole pieces with a little copper ring around one. Now I've seen it and I believe! What I don't believe is that I had never noticed this before.

 

[spiveyb]

I just made this same mistake. This thread help me diagnose the issue... thanks! But now I likewise feel dumb.