First relay build.. NOISY! Need help!

[LordOVchaoS]

Ok guys.  I built a tubescreamer clone with remote switching capabilities.  I used an Omron G6H-2-DC9 DPDT 9VDC relay with the negative of the coil connected to a jack that that shorts tip to ground.  The setup works great for bypassing the effect!  The problem I'm having is that it's REALLY noisy!  A couple of things I thought of were that the first few times I used it I forgot to put a diode across the coil to avoid voltage spikes going to the input of the effect.  Could I have damaged my input transistor?  I've never dealt with a bad transistor before so I don't know the symptoms.  It did seem to get worse the more I used it but that may be a mental thing.  Another thing I thought of is the enclosure is powdercoated inside and out.  Do I need to strip the powdercoat off where the jacks and pots are mounted to get a better ground? 

One more question that may be a bit unrelated.  On Geo there is a 10uf electrolytic cap shown in parallel with the relay coil.  What's this for? 

Sorry for all the questions!  I know I could just try all this stuff out but I have VERY little spare time and want to know where you experts think I should look first!  Thanks!!!

[R.G.]

When you're using a relay, you have delicate high impedance low voltage signals a fraction of an inch away from a signal that swings 9V or so instantly. It's the "instantly" that really hurts. There is always capacitance between any two electrical conductors - like the coil and the signal contacts - and the faster the edges on the signal in one, the easier it is for the capacitances to couple the signal between them.

The secret of quiet relay use for guitar signals is to get low capacitance between coil and signal contacts by buying a good relay, and then to slow down the sudden change in voltage on the coil. That capacitor across the coil in my stuff at Geofex? That's there to slow down the voltage change across the coil. The active relay driver circuit at GEO is also to slow down the rate of change of voltage across the coil.

The diode across a relay coil is there because the coil of a relay is an inductor. When you open the relay coil, the stored magnetic field energy tries to keep the current flowing, so it ...instantly... inverts the coil voltage and causes the voltage to rise as high as it needs to to keep the same current flowing. This is the familiar inductive kickback phenomena. That's what the diode across the coil prevents by clamping the reverse voltage on the coil to one diode drop.

Usually what this damages is a driver transistor on the coil itself. You never specifically say, but it sounds like there is only the remote switch, and no relay coil driver transistor. So there may be no damage from coil kickback, since the effect input is only connected to the coil through small parasitic capacitances.

There are no "the symptoms" for a damaged transistor, other than it acts differently from what it once did, all the way from lower gain to completely dead.

You do need to use a toothed washer on the jacks to ensure the enclosure is grounded, but that's an RF and AC power line hum statement more than anything involved with your relay. 

I think you need to figure out how to slow down the coil voltage as it turns on and off.

[Jaicen_solo]

Damn R.G, I knew that one!! (Yeah, you explained it to me once so I was going to save you the bother this time ).
I don't know if it comes across in that message, but noise can also be caused by a sudden drop in power supply current, especially if it's near the rated specs.

[LordOVchaoS]

Thanks so much R.G.!  I was hoping you'd be the guy to answer this as I learned how to do this from your site.  So do you think the cap across the coil may be the answer?  I'll try it tonight if I get time!  I'm going to go check out the relay driver you were talking about now.  I'll get a better ground on my jacks as well.  It may be AC noise, my DC jack is grounded to the jacks so if they're not getting a good ground...  I guess I didn't specify but the noise is a constant hum/fuzz.

[Jaicen_solo]

Sounds to me like a badly regulated power supply.

[LordOVchaoS]

Ok, so I stripped all the paint of the enclosure where my jacks are mounted, used star washers on all of them, and put a 10uf cap across the coil.  No better   I'm going to build this: http://www.geofex.com/article_folders/rmtswtch/rmtsw.htm (the one labeled "Remote Switched Relay Bypass with LED")  Today and see if it works.  Since I'm using 9 volts instead of 12 do I need to change any component values?