Muting relay problem

[decky]

I built recently a DoZ preamp based on Rod Elliott kit (P37A) powered by his P-05-rev B power supply.

http://sound.westhost.com/project05b.htm

The PSU is a standard LM317/397 dual supply but with a twist - it has a AUX voltage output to drive a muting relay.

I had a turn ON thump so the relay was a must. However I did not have any turn OFF noise. I included an output relay where the output is normally shorted and the relay has a delay of around 0.5 sec, just enough to avoid switching noises.

The turn ON thump is gone but I have a sharp thump after switching preamp OFF - looks like from the relay itself. CAn I tweak my relay setup somehow to get rid of this? MAybe a capacitor across the relay supply to slow down its reaction.

 

[Frank Berry]

Did you put a diode across the relay's coil to suppress the inductive "kickback"?

 

[decky]

That diode is on the relay supply board already, together with a 10uF cap in parallel with the diode.

It seams to me that the shorting of the preamp output causes that - like there is significant level of signal at the output even after the relay delay time. Should I increase the delay? However I am not sure how to do it.

 

[AndrewT]

the relay should be delayed ON and instant OFF.
That capacitor across the coil is keeping the relay powered and preventing instant OFF.
Post the circuit in case there are other delays built into your relay operation.

 

[decky]

Here is the PSU - AUX out is the relay supply

 

[AndrewT]

take your relay power from d2. That will discharge c11 more quickly.
Is c13 & r10 necessary?

 

[decky]

I am not the designer - its Rod Elliott's kit - here is his explanation:

"When power is first applied, Q3 is turned on by the charging current of C13. This shorts the base of Q2 and prevents it from turning on, so there is no output. After C13 is charged, Q3 turns off, Q1 and Q2 turn on, and DC is available at the aux output. As soon as AC is removed, C11 discharges rapidly, removing base current from Q1, so Q1 and Q2 turn off, removing the DC. An attached relay will promptly drop out, activating the mute function."

I hope it makes things clearer.

But why there is that noise on power off? CAn it be the relay itself?

 

[Burnedfingers]

I am assuming that you have a seperate power amplifier.

The correct turn on and turn off sequence is as follows:

Turn the preamp on first and after it is stable then turn on the power amplifier and there will be no noise.

Turn off the power amplifier first and allow a few seconds for the power supply to die down and then turn off the preamplifier and no noise will result.

Not being familiar with what you have there I can only make a guess. No need for a relay when the units are powered on and off as I mentioned.

 

[decky]

I know about the powering sequence and you are right I have a power amp (gainclone) after the preamp. I just wanted to do things "properly"

 

[Mooly]

All sounds a bit weird this, just to recap then, you say you have a relay connected across the output of preamp that shorts this output to ground and the relay is in this state with no power supplied (the output reads s/c to ground with all off). You switch on and after delay relay powers up removing short on output. No noise there ? On switch off if the relay instantly shorts output I can't see it's possible to get noise.Possibilties include, 1. A conflict off time constants, does the relay drop out and then pull back in again perhaps for a fraction of second. 2. Relays often have significant contact resistance, is it directly across a low impedance output, try a 100 ohm resistor in series with output before the relay. As Andrew points out it has to be "instant disconnect" which in this case means contacts to close. 3. There is a D.C. offset somewhere and the sudden short perhaps discharges a coupling cap somewhere causing thump.
Regards Karl

 

[decky]

You are right about the relay arrangement and regarding the noise- there is a shot delay on the power On sequence and no noise. There is a short delay on power OFF and there is that thump noise. So, the relay does NOT react instantly. The output impedance of the gain stage is only 300ohm - and that can be defined as low.

 

[Mooly]

O.K. What I would try is this, first connect a LED + resistor across relay so you can see when it mutes and if it switches cleanly. Without component values, it's a bit hard to say but I think C11 is keeping Q1 turned on to long. Try a smaller cap, perhaps even less than 1mfd depending on the values of R7/8 . You should aim to keep Q1 only just on. If you could tell us values of R7/8 and C11 it would help.You could also try adding a resistor across C11 to speed up discharge.
Karl

 

[Mooly]

Me again Thinking about this, it's not good practice to define the discharge rate by just dumping the charge into a B-E junction. The transistor will linger in no mans land as the base current reduces slowly. Without major mods look at this idea. Not shown on sketch but add resistor across C11.

 

[john blackburn]

As I understand it, the relay Diode spoken of by Frank Berry isn't the one shown on the schematic.

Scroll down a way to potection diodes on this page...........
http://www.kpsec.freeuk.com/components/relay.htm

John

 

[Burnedfingers]

How about putting a relay on the input of the Gainclown that shorts the input to ground when the amp is powered off?

Power it off the AC going to the bridge. Have a delay upon turn on. On turn off it will turn off right away.

 

[decky]

OK thanks guys, I will try with the resistor across C11 which is only 1uF - is 100ohm enough?

 

[Mooly]

Sounds low to me, what's value of R7/8 ?

 

[decky]

22k and 2k2 respectively. I hope Rod does not mind this little info disclosure of mine

 

[Mooly]

Add a resistor across C11, try 56Kohm to 120Kohm. Made some assumptions, AC from transformer 18-0-18 volts. R9 around 4K7. Also add a resistor across the Base Emmiter junction of Q2 the BD140. Use a value similar to R9.

 

[AndrewT]

have you removed C12 yet?