Hi all!
I just can't figure this one out. My relay bypass is popping when deenergized.
I have built the ARP Quadra Phaser http://jhaible.heim.at/tonline_stuff/hj_qph_1.gif (http://jhaible.heim.at/tonline_stuff/hj_qph_1.gif). The link is a schem of the in/out section.
It's a stereo effect so I put an opamp buffer first from the main in jack, the buffer is ac coupled on it's output. The output from the buffer goes both to the input of the quadra ( see schem), where the signal has a path to ground via the resistors on the quadra's input, and to the normally open side of a relay. The relays common goes to stereo out's of the entire thing. So the relay switches the outputs between out from quadra in stereo or the same buffered bypass signal on both main outs. The outputs from the quadra also have paths to ground via resistors.
I can't measure any DC difference on the signal lines when switching states.
The power to the relay is via a split on the secondary of the mains ac transformer to it's own dc supply. I figured this was a good way to prevent strange things with the bypass since the quadra's own power supply is balanced. It's a center tapped transformer so the supply for the relay uses the entire winding and the quadra uses the center tap for +/- 12VDC.
There is no relay driver, only a swich (on/off), a led and resistor in series with relay power (on/off indicator), a 10uF cap across the coil and an option to do the switching remotely. So the entire effect is in bypass when deenergized. Maybe I should add that protective diode. I thought I put one in there but now I see I didn't... There's no active driver in there but that 78L12 might be considered a driver in this case...
Anyway. No matter what I get a pop when going to bypass. No matter which output I use and no matter if I use the switch or bypass it via a remote on/off switch. This pop is loud and clear on the clean channel of my amps. The thing that stumps me is that it occurs when going to deenergized. I always thought energizing a relay was the thing to watch out for when it comes to clicks.
Long and tired post, I know but I could really use a little help to think straight here.
Best!
Pelle G
Try adding an anti-EMF diode across the relay coil. When the relay coil de-energizes, it's probably putting a nasty spike into the secondaries. Regulators can only do so much.
Go read all of the articles on relay switching at GEO - http://www.geofex.com - especially the one on relays for audio.
The pop is the inductive kickback of the coil inductance capacitively coupled to the audio contacts. Slow down the coil voltage change or snub it solidly.
Thanks guys!
I have read most of the relay articles on geo but gotten it all a little backwards :icon_redface:
In fact, this is what I was using as a template for my switching http://www.geofex.com/FX_images/relays2.gif (http://www.geofex.com/FX_images/relays2.gif). So I thought that the 10uF across the coil was for slowing the switching down? I've got that one in place at least but will of course add the anti-EMF diode too.
And edit to the last post, my led indicator with series resistor is in fact parallel to the coil.
/Pelle
Update:
Today I added the anti-emf diode. Nothing audibly helpful. Then I tried a 1000uf cap across the coil and that slowed the switching down a lot.
Visible too via the led. Too bad it didn't have any effect on the popping. When the contacts finally return to the nc state, the pop is as loud as before.
It seems as if the contact closing is causing the pop in this case.
Is this a case when I should be looking at a shielded relay? Or maybe do a better layout,
my input buffer and relay are on the same piece of stripboard not too far from each other.
Cheers!
Pelle
Did you mistake the pinout of the relay, or otherwise get DC onto the relay contacts?
I double checked the pinout and wiring and that's ok. The bypass function has all along been working as expected.
But I did find a small DC offset (about -20 mV) when in bypass mode but not in effect mode. Haven't seen this before.
So my buffer (ac coupled) is always feeding the input of the quadra and both nc contacts of the relay. When the nc contacts are in contact with the poles (going to main out), I get this DC offset. No such thing when the poles are in contact with the no contacts (coming from quadra out, ac coupled). Thing is, both the input and outputs of the quadra have resistance to audio ground.
I guess I'll have to divide and conquer here. Disconnect everything and see if I got a leaky cap somewhere, tiny solder brige or something.
Btw, is there a reason for wiring up the bypass with remote led as in your second picture here http://www.geofex.com/FX_images/relays2.gif (http://www.geofex.com/FX_images/relays2.gif) RG?
I mean, wiring with a DPST instead of, as I did, parallel the indicator and coil and then just a SPST? It feels like I'm missing something again...
Best!
Pelle
I'm trying to remember. That was seven years ago. :icon_eek:
A couple of things come to mind. I think maybe I was trying to prevent the relay from reverse-breaking the LED. But a diode reverse biased around the LED/resistor takes care of that. If I were you, I'd try the diode first.