Guitar amp ground lift myth?

[leadbelly]

Mr. leadbelly
If you read my original post above, the suggestion was that both the preamp and the amp chassis were to be connected to the safety ground. And that the preamp circuit ground be tied to the chassis ground. That was the assumption. I was only trying to suggest a way to defeat ground loops without removing the safety ground connections to the chassis. If someone wants to muck around with equipment without safety grounds that's their business. It's a free country. They can roll the dice if they wish.

Sorry, wasn't trying to pick on you, just have a little fun.

In all seriousness, your posts were a little tone deaf though. There's no point pulling out favorite recipes when the audience refuses to cook!

 

[6V6dude]

Just to add, I have a lot of vintage gear, most of it doesn't have a ground originally. The only way for current to go live is if there is some kind of mechanical failure inside. High voltage wire comes off and touches the chassis or part of circuit. Nothing else can do it, the chance is pretty remote but yes it does happen, specially with very old gear. For this reason I keep my old gear grounded by extra hard wire to the chassis which connects to another piece of grounded equipment. Some old equipment even has ground connector on chassis for that reason.

 

[Gnobuddy]

The only way for current to go live is if there is some kind of mechanical failure inside. High voltage wire comes off and touches the chassis or part of circuit. Nothing else can do it...

You're right, I have seen the high voltage wire come off and touch something it shouldn't. But that isn't the only failure mode.

The enamel (varnish) insulation on the copper wire inside the power transformer is incredibly thin. In old transformers, that varnish is even more fragile than modern synthetic varnishes. It's not that unusual to see varnish flaking off the wire wherever it goes around a tight corner, which happens four times for every turn of wire around a transformer bobbin with its square corners. Exposure to humidity, and weathering from decades of temperature and humidity swings, can also make the varnish flake off the wire, leaving exposed copper carrying high voltages.

The only other insulation in some old transformers is a sheet of paper, or maybe cardboard. After fifty years, paper and cardboard deteriorate, too.

So old transformers are often leaky. This is another path for high-voltage electricity to reach the chassis, via the transformer core and transformer mounting bolts.

There are other failure modes, too. Suppose an electrolytic capacitor ages and becomes leaky - now the B+ voltage on the positive terminal can leak to the negative terminal / capacitor can, which is usually grounded to the chassis. Now the chassis is "hot", waiting to shock some unfortunate person.

Even vacuum tubes themselves can fail in dangerous ways. Inside a valve, there are fragile electrodes spaced very close together. What happens if the input valve in a guitar amplifier is damaged by vibration or impact, and the grid wires come into contact with the inside of the anode ("plate")? Now the full anode voltage flows out of the grid, and into the guitar connected to it. And into the guitarist, if the amp chassis is not securely grounded.

We can't anticipate every possible failure mode. Nature is complex, and will constantly come up with new ways to kill us. Statistically, the simple fact is that enough people died or were badly injured in the days of ungrounded two-wire AC cords, to cause new laws to be written. That's why just about every country now insists on either three-wire systems with a solid and separate chassis ground, or double-insulation with two-wire AC plugs.

If it's missing, it's definitely a good idea to add that third ground wire to old electronics!

-Gnobuddy

 

[jneutron]

6V6, you need to run AC extension lead from amp power source to supply the preamp, ie both items are on the same power circuit. This is routinely done in live sound and eliminates such earth noise problems. Disconnecting the safety earth at the preamp is actually illegal and as we know inherently hazardous. My joking about electric shocks is that such shocks are a reminder that safety earth conventions are for good reason and must be followed.

Dan.

While I absolutely hate to agree with Dan ... This is an excellent way to do it. In addition, wrap the extension cord and line level cords together. Never break either ground.

Option "C"
system_ground_layout_options - My Photo Gallery

jn

ps.. Bill, one can be shocked many times, electrocution can only happen twice to an individual...the first time...and the last time. They occur simultanously..

 

[6V6dude]

So old transformers are often leaky. This is another path for high-voltage electricity to reach the chassis, via the transformer core and transformer mounting bolts.

There are other failure modes, too. Suppose an electrolytic capacitor ages and becomes leaky - now the B+ voltage on the positive terminal can leak to the negative terminal / capacitor can, which is usually grounded to the chassis.

I can confirm the old trafo or cap leaking, I have a couple of old reel to reels, when I put voltage tester on chassis it lights up. When I measure the Voltage, it's about 100V (we have 240 here). Not enough current to get a shock but I wouldn't touch it bare foot. I only got those for spare parts so won't use them.

As for capacitor leaking, that can actually kill a circuit even when it's grounded. There are some tube output transformer less preamps out there for bass/guitar that only use capacitor directly from anode for output. If the cap goes short the high voltage can blow what ever is connected to it. But again it is rare and usually starts hum increasing hum over time.

 

[Kay Pirinha]

In the amp, AC ground the circuit to the chassis with a film cap and add a CL-60 thermistor in parallel to the cap. The resistance of the thermistor should take care of the ground loop and still allow a fault to trip the circuit breaker though the safety ground.

Why a thermistor? A NTC thermistor needs some time to get hot enough and become low ohmish. While this time you're getting electrocuted most certainly.

Better do it the usual way and use a thick (>35 A) rectifier bridge with faston's, tie it's + and - together and connect one ~ with the chassis and the other one with signal ground. This bridge allows a potential difference of about +/- 1.5 V to avoid hum, but will be strong enough to trip the circuit breaker in case of a fault, and if it also fails, it's die will melt and cause an even lower ohmish path.

Best regards!

 

[jneutron]

The assumption that a 35 amp bridge will fail short is not guaranteed.The fault currents will depend heavily on how far away the load panel is. In addition, bridges are internally designed in two ways.

First type, the die is oriented parallel to the mounting surface, and this type will be more inclined to fail open. The only thing holding integrity is the potting compound, usually Emerson and Cuming 2850MT or KT in either black or blue (the typical name is "stycast").

Second type, the dies are oriented perpendicular to the mounting surface. in this case, if the aluminum housing is one piece with side walls, the walls will maintain integrity such that the silicon will remain in place and melt into short.

In essence, standard bridges are NOT safety rated, nor are they intended to be used as a safety device. So, I would not recommend it.
jn

 

[thoglette]

The only way for current to go live is if there is some kind of mechanical failure inside.

Unfortunately, that's just not true. There's any number of ways to get a low-enough impedance between active and chassis, with the death cap (Rob Robinett's explanation) being the classic case.

I've seen 65VAC at milliamp levels on "double insulated" equipment (this on an RS485 line) that was not exhibiting any obvious fault beyond the comms not working reliably eek.

For this reason I keep my old gear grounded by extra hard wire to the chassis which connects to another piece of grounded equipment. Some old equipment even has ground connector on chassis for that reason.

This practice of yours is to be commended.

 

[Gnobuddy]

I've seen 65VAC at milliamp levels on "double insulated" equipment (this on an RS485 line) that was not exhibiting any obvious fault beyond the comms not working reliably eek.

A former friend (he passed away) was a licensed electrician and contractor in California, and also taught in the electrical program at a community college. He told me the same thing - he too had encountered dangerous "double insulated" equipment.

He didn't trust double insulated equipment at all, and felt that the concept was invalid, and such equipment should not be legal.

Like most of us, I have small "double insulated" power tools with two-wire AC cords, some with exposed metal. My friend's words still ring in my head sometimes, when I work with these.


-Gnobuddy