Spikes and Surges

[moonloopsun]

Hi guys.
i hope that someone can help us.
We moced to a new flat and installed an audio system in a room where we have 4 Shutters that are driven by a Shelly 2.5 relay.
the problem is that when the relays are starting or stopping the engine the turntable‘s preamp is picking a spike pop noise and amplify it.
we tried everything from the audio system‘s side (electric filters, surge suppressors, different cables...),
On the relay side the only thing we tried is to to wrap the cables with Ferrite clips.
we don’t know if the spike sound is created by the relay itself or the Shutters engine.
but we can’t use our sound system because those amplified pop sounds can destroy the speakers.
wirth to mention that in this room there is also a motorized beamer screen running by a different type of relay and it creates the same problem..
is there anyone that could possibly help us to understand what we have to do?

thanks a lot for your time.

 

[gophert]

Hi guys.
i hope that someone can help us.
We moced to a new flat and installed an audio system in a room where we have 4 Shutters that are driven by a Shelly 2.5 relay.
the problem is that when the relays are starting or stopping the engine the turntable‘s preamp is picking a spike pop noise and amplify it.
we tried everything from the audio system‘s side (electric filters, surge suppressors, different cables...),
On the relay side the only thing we tried is to to wrap the cables with Ferrite clips.
we don’t know if the spike sound is created by the relay itself or the Shutters engine.
but we can’t use our sound system because those amplified pop sounds can destroy the speakers.
wirth to mention that in this room there is also a motorized beamer screen running by a different type of relay and it creates the same problem..
is there anyone that could possibly help us to understand what we have to do?

thanks a lot for your time.

I am assuming the relays are DC coils and as the relay opens at the end of the cycle, the pop occurs. If so, adding a diode in parallel (actually "anti-parallel") with the coil connections on the relay will reduce the intensity of the pop.

Let us know if the popping is the whole time the motor runs vs just at the end of the movement. Also, post photos of the relay (top with model number and specs and bottom to show wiring connections.

 

[moonloopsun]

Thanks a lot for your reply.
a photo of the connections is attached.
it’s Shelly 2.5 relay. The motors i don’t know. I don’t have access to them as they are in ceiling..
the pops are at the beginning / end of the cycle. Not during the time the motor is turning. Just when it kicks in and stops.
Can you please direct me where i should put the diode and which site or model i should get please?

Thanks a lot for your time

 

[gophert]

The diode won't work with AC operation. The relay can be used with DC but that is not your goal. I'm running off to an appointment but maybe someone can add some capacitors to smooth out your spikes for you. Otherwise, give the nice people at Shelly a call/email and let them know your problem.

 

[simonbramble]

What wires in the hifi have you tried moving? It sounds like the noise is getting into the input wires and not the speaker wires. Loudspeakers are low impedance so it should be quite difficult to put noise onto these wires (but still possible). The input wires to the amplifier (or preamp) are the ones to re-reroute first. Also, are these coax (shielded). An earth shield around the wires will help block out noise from the outside.

It sounds like the spike is coming from the relay action on the motor windings. If you disconnect the current in anything inductive, it produces a spike, so it is the interaction of the relay and the winding, not just one component. Regarding the diode/capacitor mentioned above, this really needs to be placed as close to the motor as possible, but if you dont have access to it, then you need to compromise.

Have you tried wrapping the relay box in tin foil then earthing it somehow? This will provide a crude way of stopping the electrical interference getting out.

Hopefully these give you a few ideas to try. Let us know how you get on..

 

[gophert]

It would have been great if the designer would have thought about the interference caused by connecting/disconnecting inductive loads (both the electromagnetic switching coil of the relay and the motor). They could have used solid state relays with zero-crossing detection circuitry.

 

[KeepItSimpleStupid]

The motor might have a capacitor across 01 and 02.

What's the distance from the "Shelly" and the motor?

A couple of things come to mind:

a) Move the motor cap to close to the shelly.

#1. Twisting eliminates EMI
#2. Shieleding reduces RFI.

3. The voltage across a capacitor can't change instantaneously.
4. The current in an inductor can't change instantaneously.

Suspecting EMI.

a) Move the cap closer to the shelly if there is one.
b) Snubbers across the AC contacts. This is an RC series combination. Calculating the value is somewhat of an art.
But 10-100 ohms and maybe 1 uF AC rated capacitor.
c. Twist the shelly to motor connections.

Edit: add snubber reference: https://en.wikipedia.org/wiki/Snubber

If you disconnect the motor, it likely would not happen so you can try that. Lets also say that twisted pair shielded cable wit the shield connected at one end. The end preferred is the signal generating side is used a lot. Power circuits, not so much.
I specified shielded power cables for heaters for a 13.6 MHz RF plasma chamber. We had no issues.

 

[moonloopsun]

The setup is:
Turntable connected to a preamp using rca cables and a ground cable (it’s not those components as we tried changing them already. Currently an Audio Note TT3 turn table.
pre amp of Whest2, also tried a different one, same spikes.
the pre amp is connected to the speakers via xlr cables (audioquest cinnamon) tried different brand. No luck there.
I tried to change everything from the audio side. Even electric cables.....
It drives me crazy!
I can access the shelly relays as we opened the wall. But can’t access the motor

 

[KeepItSimpleStupid]

The motor MAY be a PSC motor that uses a capacitor. You end up with 3 wires with a common, a CW and CCW directions It might be useful if you can find a wiring diagram. You connect c to either the CW or CCW sides.

You likely have a large antenna. So, when the motor turns on, you get a nice little short briefly at the other end.

What I was really suggesting to do is to:

Shield the wires from the Shelly to the motor and if there is a capacitor, move it closer to the relay.

Try this for fun. If you know the approximate path of the wiring, temporarily put a sheet or sheets of grounded aluminum foil along the path, Just see if it gets better. It's not foolproff, but the radiation could be to a nearby power line as well. In that case, try powering the stereo from an extension cord to another circuit. There might be a change. I am in no way suggesting the problem will go away.

I'm just using this https://www.hurst-motors.com/lyg42geared.html as a wiring diagram. An AC synchoous motor is essentiall constant speed because speed is related to frequency, so as long as the motor has enough torque, speed will be constant.

There are X and Y capacitors or safety capacitors e.g. https://www.digikey.com/en/product-highlight/k/kemet/emi-suppression-capacitors and thats what you need for a snubber. Also Metal oxide resistors.

They should go across the contacts of your relay. Would need some info about the motor to try to size a value.
A ZNR would probably also help.

Another idea that MAY work is to use a delay on make relay and a delay on break relay. Unfortunately, these are usually just control relays and would need another one to handle the current. You can apply power through a metal oxide resistor for a short time and teh short that resistor out. The resistor would burn out if left in the circuit for a long time.

The coolpart about metal oxide resistors is that they pretty much break in half when they are overloaded.

I would suspect that the delay on make would only help the ON spikes and not the OFF ones.

The voltage across an inductor v = is L di/dt. So, when the motor is turned off, di/dt is esentially any you get a nice big voltage spike/

So, shorting out the motor when it's turned off, reduces the voltage. Having that short closer to the motor reduces the antenna length.

di/dt happens when the motor turns on too.

You have a big antenna.

So, the supression has to be close to the contacts AND close to the motor. Close to the contacts should dampen turn on spike. Close to the motor, turn off spikes.

The snubber also "protects" the relay contacts.

At work we had a 1000 W arc lamp that required a 40 kV pulse to start the lamp. Normal power was 22V at 40A or so. there were two 5' cables going to the lamp housing. Whenever we started the lamp the 8" floppy drive would move. We just didn;t start the lamp with the floppy drive running. Moving the ignitor inside the lamp housing fixed the problem. The OEM redesigned the ignitor module.

I don't think it would work, but you could interrupt the pre-amp signal with an optofet or short it for a short time when the motors turn off.
Like I said. Not a good idea.

As you no doubt found out, "Throwing money at the problem" doesn't work either.

 

[Pommie]

If the interference isn't getting into the input cables and is being amplified, could it be via the power supply? What happens if you power the amp via an extension from another room?

Mike.

 

[moonloopsun]

Hi

If the interference isn't getting into the input cables and is being amplified, could it be via the power supply? What happens if you power the amp via an extension from another room?

Mike.

mike. When i plug the whole audio system to another room it helps a lot!
Any guesses?

 

[moonloopsun]

The motor MAY be a PSC motor that uses a capacitor. You end up with 3 wires with a common, a CW and CCW directions It might be useful if you can find a wiring diagram. You connect c to either the CW or CCW sides.

You likely have a large antenna. So, when the motor turns on, you get a nice little short briefly at the other end.

What I was really suggesting to do is to:

Shield the wires from the Shelly to the motor and if there is a capacitor, move it closer to the relay.

Try this for fun. If you know the approximate path of the wiring, temporarily put a sheet or sheets of grounded aluminum foil along the path, Just see if it gets better. It's not foolproff, but the radiation could be to a nearby power line as well. In that case, try powering the stereo from an extension cord to another circuit. There might be a change. I am in no way suggesting the problem will go away.

I'm just using this https://www.hurst-motors.com/lyg42geared.html as a wiring diagram. An AC synchoous motor is essentiall constant speed because speed is related to frequency, so as long as the motor has enough torque, speed will be constant.

There are X and Y capacitors or safety capacitors e.g. https://www.digikey.com/en/product-highlight/k/kemet/emi-suppression-capacitors and thats what you need for a snubber. Also Metal oxide resistors.

They should go across the contacts of your relay. Would need some info about the motor to try to size a value.
A ZNR would probably also help.

Another idea that MAY work is to use a delay on make relay and a delay on break relay. Unfortunately, these are usually just control relays and would need another one to handle the current. You can apply power through a metal oxide resistor for a short time and teh short that resistor out. The resistor would burn out if left in the circuit for a long time.

The coolpart about metal oxide resistors is that they pretty much break in half when they are overloaded.

I would suspect that the delay on make would only help the ON spikes and not the OFF ones.

The voltage across an inductor v = is L di/dt. So, when the motor is turned off, di/dt is esentially any you get a nice big voltage spike/

So, shorting out the motor when it's turned off, reduces the voltage. Having that short closer to the motor reduces the antenna length.

di/dt happens when the motor turns on too.

You have a big antenna.

So, the supression has to be close to the contacts AND close to the motor. Close to the contacts should dampen turn on spike. Close to the motor, turn off spikes.

The snubber also "protects" the relay contacts.

At work we had a 1000 W arc lamp that required a 40 kV pulse to start the lamp. Normal power was 22V at 40A or so. there were two 5' cables going to the lamp housing. Whenever we started the lamp the 8" floppy drive would move. We just didn;t start the lamp with the floppy drive running. Moving the ignitor inside the lamp housing fixed the problem. The OEM redesigned the ignitor module.

I don't think it would work, but you could interrupt the pre-amp signal with an optofet or short it for a short time when the motors turn off.
Like I said. Not a good idea.

As you no doubt found out, "Throwing money at the problem" doesn't work either.

Thanks a lot for your kind answer. I will try a snubber. The problem is that i can’t see the motor and nor the cables (all in celing). I opened the wall where the relays are to access them. But can’t open all the way. It’s built in a very complicated way..
On another note, i tried plugging the whoke audio system to another room in another floor. It helped so much.....

do you think that a snubber would do the job?
I should connect it between the relay and the motor on the power line actually?
Which size u think is good?

Thanks a lot for your kindness

 

[Pommie]

Hi

mike. When i plug the whole audio system to another room it helps a lot!
Any guesses?

It sound like the power to the room is at the end of a long spur or there is a loose connection somewhere along the way. I'm not familiar with German wiring regulations. BTW, I'm assuming the system wasn't moved to another room but an extension lead was used. If you moved the whole system then RF and EMI interference can't be ruled out.

Mike.

 

[KeepItSimpleStupid]

Snubber design: https://sciencing.com/design-rc-snubber-12030206.html and https://daycounter.com/Calculators/Snubbers/Snubber-Design-Calculator.phtml

Capacitors: https://uk.rs-online.com/web/c/?sra=oss&r=t&searchTerm=x1+y1+capacitors Note the X1 and Y1 and the voltage rating. https://www.hunker.com/13408990/differences-in-x1-x2-y1-and-y2-capacitors You want an X class.

Metal oxide resistors: https://uk.rs-online.com/web/c/?sra=oss&r=t&searchTerm=metal+oxide+resistors
These can work too: https://uk.rs-online.com/web/c/?sra=oss&r=t&searchTerm=non+inductive+resistors

The snubber will go across the relay contacts.

ZNR's Search for ZNR's at RS-oniine. You probably want 240V operating. The ZNR across the line in and motor so you would end up with three.

Do, I think it will work 100%? No. Help? yes.

Wires that parallel each other pick up interference by capacitance, Wires that cross don;t so much.

A note on shielded cables: https://www.mouser.com/pdfdocs/alphawire-Understanding-Shielded-Cable.pdf

So, really and short lengths and metal conduit combined with a snubber have the best chance of fixing it.

Power conditioning: I can tell you that these https://www.powervar.com/products/power-conditioners/ground-guard-power-conditioners or similar
combined with https://www.tripplite.com/isobar-8-outlet-surge-protector-12-ft-cord-3840-joules~ISOBAR8ULTRA The connected equipment warranty in the US is hard to beat. We used the warranty once for the ISOBAR in an ISOBAR only surpressed system. The suppressor was covered in black soot. Then someone decided to replace them with plastic supressors from APC.

I took a failed one apart and got really angry, The plastic was compromised and the "protection" if it fails "protects: the surge supressor and not the equipment.

The Triplite and Powervar (Then known as OneAC) made a fantastic power conditioner. I used the combination at work in the 1980's for eventually three critical data acquisition systems where a complete loss of power was OK.

One particular Macintosh computer was in service for 17 years, The SCSI hard drive had no issues. The floppy drive died (mechanical) and the computer accumulated dust, With the computer prior failures were fans due to dust, power supply and 8" floppy drives.

For the most part, the power conditioner is an isolation transformer that re-references the neutral/ground connection of the secondary.
The order probably doesn't matter much. What the transformer does is it attenuates greatly any high frequency spikes because they won;t pass from the primary to the secondary. You also get a clean ground.

Renting a power line disturbance monitor might help. https://www.dranetz.com/dranetz-hdpq-family-of-power-quality-analyzers/

So, I think it's EMI radiated into the power line, so you can prevent it from getting there or prevent or reduce it from getting into the audio system.

Snubbing, shielding, surge suppressing and making the distance between the switch and the motor should be primary objectives.
The re-referenced isolation transformer is another way.

The powering of the stereo from another circuit is going to reduce the capacitive coupling. Powering the motors form another circuits should do the same thing.

In some respects, it more like an EMP event. https://en.wikipedia.org/wiki/Electromagnetic_pulse

 

[KeepItSimpleStupid]

In that case, try powering the stereo from an extension cord to another circuit. There might be a change. I am in no way suggesting the problem will go away.

What happens if you power the amp via an extension from another room?

Funny, how when responses are short they get more attention. Not complaining. It's just an observation that I have seen a lot. Two people with the same idea reinforces the idea. I gave Pommie a like.

 

[moonloopsun]

Snubber design: https://sciencing.com/design-rc-snubber-12030206.html and https://daycounter.com/Calculators/Snubbers/Snubber-Design-Calculator.phtml

Capacitors: https://uk.rs-online.com/web/c/?sra=oss&r=t&searchTerm=x1+y1+capacitors Note the X1 and Y1 and the voltage rating. https://www.hunker.com/13408990/differences-in-x1-x2-y1-and-y2-capacitors You want an X class.

Metal oxide resistors: https://uk.rs-online.com/web/c/?sra=oss&r=t&searchTerm=metal+oxide+resistors
These can work too: https://uk.rs-online.com/web/c/?sra=oss&r=t&searchTerm=non+inductive+resistors

The snubber will go across the relay contacts.

ZNR's Search for ZNR's at RS-oniine. You probably want 240V operating. The ZNR across the line in and motor so you would end up with three.

Do, I think it will work 100%? No. Help? yes.

Wires that parallel each other pick up interference by capacitance, Wires that cross don;t so much.

A note on shielded cables: https://www.mouser.com/pdfdocs/alphawire-Understanding-Shielded-Cable.pdf

So, really and short lengths and metal conduit combined with a snubber have the best chance of fixing it.

Power conditioning: I can tell you that these https://www.powervar.com/products/power-conditioners/ground-guard-power-conditioners or similar
combined with https://www.tripplite.com/isobar-8-outlet-surge-protector-12-ft-cord-3840-joules~ISOBAR8ULTRA The connected equipment warranty in the US is hard to beat. We used the warranty once for the ISOBAR in an ISOBAR only surpressed system. The suppressor was covered in black soot. Then someone decided to replace them with plastic supressors from APC.

I took a failed one apart and got really angry, The plastic was compromised and the "protection" if it fails "protects: the surge supressor and not the equipment.

The Triplite and Powervar (Then known as OneAC) made a fantastic power conditioner. I used the combination at work in the 1980's for eventually three critical data acquisition systems where a complete loss of power was OK.

One particular Macintosh computer was in service for 17 years, The SCSI hard drive had no issues. The floppy drive died (mechanical) and the computer accumulated dust, With the computer prior failures were fans due to dust, power supply and 8" floppy drives.

For the most part, the power conditioner is an isolation transformer that re-references the neutral/ground connection of the secondary.
The order probably doesn't matter much. What the transformer does is it attenuates greatly any high frequency spikes because they won;t pass from the primary to the secondary. You also get a clean ground.

Renting a power line disturbance monitor might help. https://www.dranetz.com/dranetz-hdpq-family-of-power-quality-analyzers/

So, I think it's EMI radiated into the power line, so you can prevent it from getting there or prevent or reduce it from getting into the audio system.

Snubbing, shielding, surge suppressing and making the distance between the switch and the motor should be primary objectives.
The re-referenced isolation transformer is another way.

The powering of the stereo from another circuit is going to reduce the capacitive coupling. Powering the motors form another circuits should do the same thing.

In some respects, it more like an EMP event. https://en.wikipedia.org/wiki/Electromagnetic_pulse

Thanks a lot.
do you think that this will do the job?

Rc absorption snubber circuit module relay contact protection resistance surWP4 | eBay

Entdecken Sie Rc absorption snubber circuit module relay contact protection resistance surWP4 in der großen Auswahl bei eBay. Kostenlose Lieferung für viele Artikel!

www.ebay.de

 

[KeepItSimpleStupid]

The MOV is 10D471K, https://www.mouser.com/datasheet/2/54/mov10d-777448.pdf. Hopefully, R2 is a 10 ohm 2% resistor.

It could be closer to your line voltage.

Some info about your motor and we can calculate possibly better values. The seller doesn't know what he is selling. This https://www.amazon.com/Firiodr-Inductive-Solenoid-Interrupter-Protection/dp/B07VB8FNSK/ref=pd_sbs_5? shows a 22 ohm resistor/
of 22 ohms.

Get the values you needed by the calculator from the motor specs. Your pretty much going to need 3 of them.
The PCB is "sort of nice" No mounting holes. The terminals might not even accept the wire used. Lot's of potential gotcha's. for a simple product.

A large piece of heat shrink tubing to cover the module would be really useful.

I typically hate the following vendors:

Amazon: Product reviewed might be an entirely different one, A product basically looks the same, but has different capabilities and thus price. Amazon could care less. No way to correct easily. No "datasheets" or instructions like I'm used to. e.g. Digikey. Mcmaster-carr. they don;t know what they are selling. No support.

ebay: Similar to above. No packing slips..

Aliexpress: Similar to above, No PDF manuals. No packing slips. No support. Schematics should exist for simple stuff. packing slips ough to exist. The "manual" is the webpage usually. You have to make your own if you want one.

 

[moonloopsun]

That’s exactly my problem. I can’t access the the motor. It’s inside the wall.....

 

[KeepItSimpleStupid]

Well, from the 1st reference:

Updated April 24, 2017

By Mike Wallace

A snubber is an electrical device that prevents voltage spikes due to sudden changes in current. These voltage spikes, or transients, can damage the circuit and cause arcing and sparks. One type of electrical snubber is the RC snubber, which is composed of a resistor in parallel with a capacitor. Transients are usually caused by switches in the circuit. When designing a snubber, you should design it with the characteristics of the switch in mind. The exact switch and its switching frequency must be known before designing the RC snubber.

1. Verify that the electrical switch is in the “Off” position and connected to a power source.
   
   Measure the voltage across the switch by placing the voltmeter probes across the input and output terminals of the switch. Turn the switch to the “On” position and read the value on the voltmeter. This is the voltage across the switch. Write that down

Brought to you by Sciencing

Determine the maximum current that the switch can handle. (on the relay) This data is contained in the data sheet for the switch.

Calculate the minimum value for the resistor in the RC snubber by dividing the voltage across the switch (220, 230, 240?) by the maximum current rating. For example, suppose the voltage measurement is 160 volts and the maximum current is 5 amps. Dividing 160 volts by 5 amps gives you 32 ohms. Your snubber must use a resistor that has a minimum resistance of 32 ohms.

Determine the switching frequency, in switches per second. For example, suppose the switch changes states 50,000 times per second, or 50 KHz. This value is determined by the designers of the circuit, and should be available in the circuit documentation.
(50 or 60 hz - It won;t matter much)

Calculate the capacitance for the snubber by multiplying the switching frequency by the squared value of the voltage measurement acquired in Step 2. Take the inverse of this number (i.e. divide 1 by the value). For example, given a switching frequency of 50 KHz and a voltage of 160 volts, your equation looks like this:

C = 1 / (160^2)*50,000 = 780 pF

--

Values won't be critical.

I think you will effectively need 3 units.

One fro CW and neutral. One from CCW and neutral. And one across the line.
You can change the varistor part number based on your line voltage.

The product of R*C matters more. R in ohms and C in Farads. 780 pf = 780e-12. .1 uf = 0.1e-6.

--

I hope you didn't run audio signal close and in parallel with the power line?