Relay termination

[ccaudle]

I probed the COM with the two input pins floating (A) and with input pins connected to ground (B).

Where is the scope ground connected? And I don't just mean circuit node, I mean physically.
I (and others) pointed this out before, perhaps you missed the significance:

that will be sensitive to physical layout

This will remove an inductive pulse from the scope ground.

Can you take a pix of your layout

When the coil is energized or de-energized there is a pulse of magnetic field that spreads out from the coil, and every wire around it is an antenna. How well those antennas convert the magnetic flux into a voltage depends primarily on the loop area of the circuit. That is all about wiring layout, so how the scope is connected is very important.

 

[Mr. Moscode]

The inductive kick of the coil is showing up in the coil return circuit, aka your ground.

Don't share an audio ground with the coil. The relay is not inducing anything in the contacts and that's the business end of the relay.

Simple- isolate the relay coil power supply or if you have to share audio power run a shielded wire for the coil ckt directly to the power supply. You may also want a small cap and resistor across the switch, say .01 mfd in series with a 10 ohm R.

 

[Newmarket]

So, I did a quick test last evening...

I disconnected everything (transistor, voltage dividers, etc...) and I just energised the coil straight from the 24V supply. I probed the COM with the two input pins floating (A) and with input pins connected to ground (B).

View attachment 104142

With the pins connected to ground there are no spikes so no point in uploading the graph.

With the input pins floating (A) the results are not very charming as you can see below.

When the relay is "ON":

View attachment 104143



Does this mean that my measuring method was wrong since the contacts will be connected to some audio signals anyway?

Yes. The input pins should be at a defined potential. Otherwise they are floating at "any old voltage" and picking up noise. As has been hinted at - look at what you can get from a scope probe when you "wave it around" in the air. Near some source of noise. Or touch it without attaching the ground lead. (Basically a distorted wavecat mains frequency in most cases as most locations are swimming in 50 (60 ) Hz and harmonics of them.
Now remove the input pins from direct ground and put a resistor to ground. Se how high you can go before you measure / hear a problem.
Or run a sine wave into the pins and look if you get the clicks (apart from the inevitable) fast edge caused by switching signal to ground. That will sound as a click so need to look at scope trace to se if any other click / glitch.

 

[Mr. Moscode]

I think every one has covered all the possibilities. Every inductor will resonate and your coil is an inductor. The diode dissipates 1/2 of the resonant ringing. If it is capacitive pickup, putting the contacts in a circuit with resistors will probably kill it. But from the scope trace it appears to be inductive kick.

 

[warpie]

Thanks, I will do some further tests.

Sorry I missed the comments about the probe. The probe's ground is connected directly to the circuit's ground at the diode's anode leg. I use a breadboard but all the gnd connections are very close to the "- / gnd" pin.

 

[Admiral-dk]

Sorry if I don't get it - but you example A isn't doing anything .... no signal / impedances on the contacts ...!
This is equivalent to having a Toggle Switch only connected to the Scope and nothing else ....

Are you trying to compare Relais out of the circuit they should be used in ?

 

[FIX]

The coil will leak onto the contacts, our friend Eddy Current decided that. The coil collapsing and energizing created a big inductive spike. You can minimize it by putting a small resistor in series with he relay, 50 ohms or so, depending on the coil impedance. That slows the charging time. Make sure you don't drop more than the minimum that the relay needs to energize. Secondly, if you have a ground plane under the relay, it can also induce into that.

Having the transistor does allow you to slow the rise and fall time down, which can minimize it too. With a series resistor, you can also put a small cap across the coil which will slow it's on and off curves.

If you don't have any load on the contacts, they will pick up everything. If it's under some kind of load it will diminish. My console uses relays on every switch, and thats what I do to minimize it.

 

[Bo Deadly]

The coil will leak onto the contacts, our friend Eddy Current decided that.

I'm surprised by that but I'll take you're word for it.

With a series resistor, you can also put a small cap across the coil which will slow it's on and off curves.

If you're going in that direction, one might consider putting the cap across the series resistor instead.

More specifically, choose a series resistor that provides just enough holding current (with a fair margin of course). Then put a large MLCC capacitor across the resistor. MLCCs can be had in small packages but with high C, they're cheap and they don't dry out like electrolytics. The cap will allow sufficient energizing current to pass but then drop to the holding current. This will also reduce power consumption (by probably 20% or so). The size of the cap should be chosen to provide a wide pulse of energizing current for good reliability.

 

[warpie]

Sorry if I don't get it - but you example A isn't doing anything .... no signal / impedances on the contacts ...!
This is equivalent to having a Toggle Switch only connected to the Scope and nothing else ....

Are you trying to compare Relais out of the circuit they should be used in ?

The discussion is on whether the coil can leak onto the contacts. Toggle switches don't have coils. I hope now it's more clear

 

[Admiral-dk]

OK - but you still don't use a relay if the contacts aren't used - and none of the circuits you show are connected on both sides (the Scope do not count here) .... So you still have to make a 'Real Life Circuit' with the Relay contacts connected to representative Impedances or better yet a Live circuit (not Mains) like a PreAmp or maybe a EQ Bypass - etc .... (in my book).

But if I try to see it from another perspective .... :
Unless the Contacts and their wires are incapsulated inside a magnetic Screen (or vise versa) -> you will get the Inducted signal (can't cheat Physics) into your Signal Chain.
Many ways to reduce the problem (and many are already mentioned) - but I know of no one that removes it completely by working on the Coil + Driver side.
Peavey was the first I saw having a simple and efficient sulution the the Switch side (or more correct Signal Chain).
In their RockMaster Tube PreAmp, theres is a small SCR that is triggered everytime a Relay is switched and that gives an Impulse of some mS. that in turn have a number off strategicly placed JFets go On and Short the Signal Chain to GND => Muting All while the Relais are Noisy.

Best wishes

Per