Help calm the relay down

[alkuz1961]

Hi guys,
I just ran into a problem that I can't solve on my own. It's a very loud pop when the relay is triggered. The device in which this happens is a 1U preamp, I built using the schematic of HBE-100. Problem with the input switching relay between the first and second stages.
Just in case, here is a diagram of this place:

2021-06-06_10-52-42.jpg

2021-06-06_11-32-21.jpg

2017-08-21 14-24-15.JPG

As you can see, the relay has leakage resistors on all three pins. And in theory, it should never clack. But it doesn't work on my preamp.
https://youtu.be/ZK5Wveg461g

I took many different actions. But none was successful.
- the first thing I checked for leaks from C3 (2200pF). Everything is fine in this place.
- then I checked for leaks from the V2 grid. And is fine too.
- I also checked for an oscillation. There is nothing even when the V2 input is open.
- I covered the relay with grounded steel shield - nothing has changed.
- I powered the relay from an external power supply - still clack.
- I replaced the relay.
- I cut the connection between the R5 and C3.
- I short-circuited the R6 (68K) resistor to ground. But pop volume did not decrease.
- I short-circuited the R11 (10M) resistor to ground. But even in this case, you can hear a little pops. The volume has decreased, but the clicks are still there.
https://youtu.be/yaBEP0C9CGE

I don't think I know what else to check. Maybe someone will give me some new ideas?

[xtian]

Seems to me, you need "DC leak" resistors between pins 6 and 4 of the relay, and another between pins 8 and 4. Or maybe just a single 22M resistor between pins 6 and 8 will do the job.

[alkuz1961]

Thank you xtian, this is a good solution, and I hope this should fix the problem. But I'm still interested in the cause of this effect. There is no potential on the contacts, but it clicks. And besides, a similar solution is used in the prototype and this problem does not arise there.

[xtian]

I agree, it looks like there should be no potential at the relay. I'm only offering a Hail Mary.

[R.G.]

You guys are doing the right thing, chasing sudden changes in DC levels. That's the right place to look first.

In case the suggested additional resistors don't work, there are a few other possibilities. Take this as experimental, as most of my experience is with JFETs and JFET opamps, or MOSFETs and MOSFET opamps. JFETs are not as high an impedance as tubes, but MOSFETs are much higher impedance. An inch or two of air is a fine leakage path - and antenna! - for a MOSFET.

FIrst possibility is that the coil voltage change is being capacitively coupled in from the coil pins on the relay. This is one common issue in guitar pedals where someone just has to have a super-hyper-galactic input impedance and uses MOSFET or MOSFET opamps for inputs. Some relays are better in this regard, some worse. The best actually include a shield around the coil inside the package to stop it.
Second possibility is that the click is a "squark!" of RF caused when the relay spends some time in the middle of travels, or when make-before-break intrudes. Modern relays are not really good at specifying this.
There are a few others that are darker horses, like coil current transients pushing a click in through the cathode if the cathode resistor ground is in a different place than the grid leak resistor ground, things like that. I hope you don't have to go there.

[pdf64]

What’s the circuit around the relay coil, ie how is it being suppressed?

[alkuz1961]

What’s the circuit around the relay coil, ie how is it being suppressed?

I use rectified heating, and relays are connected from this power supply. A diode is connected to the coil as standard. I tried adding a 10uF capacitor, parallel to the coil, but it didn't improve anything. I have already written that I tried to power the relay from an external source, but it did not give the expected result.

[pdf64]

How about the switch, eg cap across it?

[alkuz1961]

How about the switch, eg cap across it?

This is a togle switch with three positions. It controls two relays. The first position control the clean/crunch relay. And this relay behaves absolutely quietly. The second relay switches the BE/HBE modes and it is this relay that pops.

[pdf64]

How about if there’s a cap across the switch, eg 100nF?

[alkuz1961]

I wrote that I tried to install a capacitor (10uF) parallel to the relay coil. And it doesn't help with the problem. Isn't that the same thing?

[pdf64]

Dunno, don’t think so but not sure, a picture (in this case a schematic) would speak a 1000 words.
Surely it would be quicker to try it than debate its potential merits or otherwise?

[alkuz1961]

Dunno, don’t think so but not sure, a picture (in this case a schematic) would speak a 1000 words.
Surely it would be quicker to try it than debate its potential merits or otherwise?

Yes, of course, I will check all possible options for correcting this problem. Thank you for the additional ideas! But unfortunately, I can only do all these experiments on weekends. But I will definitely report the results.
In addition, I hope that during this time someone will offer some more solutions, and I will be able to check everything at once.

[Gaz]

The real amps do this too. It's a bad design tbh. Freidman changed later models to have a 2.2M in place of the 10M pull down resistor. There is also 10uF cap in series with a 100R resistor across the relay coil, which helps slow down the relay a little. I don't think you'll ever get rid of the pop completely with this circuit.

[alkuz1961]

The real amps do this too. It's a bad design tbh. Freidman changed later models to have a 2.2M in place of the 10M pull down resistor. There is also 10uF cap in series with a 100R resistor across the relay coil, which helps slow down the relay a little. I don't think you'll ever get rid of the pop completely with this circuit.

Oh! This is an unexpected fact. I always thought that such errors are unacceptable in factory production.

I once did a similar use of the first clean channel cascade as a booster for the OD channel. I used the relay in a similar way, and I had no problem with the claps. So I was surprised by this behavior of the switch. But I no longer remember what resistor values I used around the relay.