Hiss in Trident 80 Monitor summing amps

[RFSiesta]

WE HAVE A VERY HOT TRACE!

1.) It's almost certainly two different problems. Ground/Hum is the one, it's global in the studio and we have yet to find out,
2.) The "ghost tone" and the crackle come from the same source and it looks like it has to do with THE CORRELATION METER PCB.

I touched the pins of a IC there, resulting in completely different sound/frequencies. We changed the chips for test, nothing changed. When measuring pins 6 of the chips, there is a strong signal - very strong on one IC, almost rail voltage! With around 500 Hz. The frequency we had all over the board. When pushing "remix to monitor", instead of the tone, the crackling comes. And it's all there:  broadband crap and crackling.
We shut down the correlation meter (by removing rail voltages connections) and: the tone IS GONE!

The problem is: We have NO schematic of this part!

SO, PLEASE - LAST CALL: Anyone knows of the schematics for the Trident 80 correlation meter pcb?

Thanks! cheers , Richard

 

[PRR]

> ground loop, but IF the hum came from magnetic fields, you should expect it to be sensitive to movement.

No; a ground Loop would be sensitive to Current and the resistance of the loop between inputs.

Which asks: when you "connect second one to power", are you using a power-strip? Same one? Do its ground contacts really work good? (Some US strips, the ground "contact" is very flimsy and prone to bad contact.)

> adequate ground from a security viewpoint, close to zero

How "close"? A continuity beeper may call 100 Ohms "continuity". Most DMMs will resolve near 1 Ohm. There are tricks to go lower, but sub-Ohm measurements suggest 4-contact techniques.

And "zero" relative to what? Probably not to True Dirt, and we probably do not care for an indoor studio (unless it is concrete floor on dirt). "Low Ohms" from here to there around the many outlets you use is probably the goal. How low is "low"? I do not know. Start by asking if any ground-ground paths look much worse than others. If you find 1, 1, 7, 1, then ask what is different about the 7.

 

[RFSiesta]

No; a ground Loop would be sensitive to Current and the resistance of the loop between inputs.
Which asks: when you "connect second one to power", are you using a power-strip? Same one? Do its ground contacts really work good? (Some US strips, the ground "contact" is very flimsy and prone to bad contact.)

I meant: It would change if you move it to another place, change the angle of the corresponding cables or decrease if we move two cables closer to each other.
Sorry, I didn't write that it was on the same power-strip and the contact is very good.

How "close"? A continuity beeper may call 100 Ohms "continuity". Most DMMs will resolve near 1 Ohm. There are tricks to go lower, but sub-Ohm measurements suggest 4-contact techniques.

REALLY close to zero, less than 1Ω iirc. It was done by the electrician

And "zero" relative to what? Probably not to True Dirt, and we probably do not care for an indoor studio (unless it is concrete floor on dirt). "Low Ohms" from here to there around the many outlets you use is probably the goal. How low is "low"? I do not know. Start by asking if any ground-ground paths look much worse than others. If you find 1, 1, 7, 1, then ask what is different about the 7.

The measurement was the loop resistance, so measuring from disconnected neutral ("Nulleiter" in Austria) to our ground bridge. Since the "Nulleiter" must have very good ground at the transformer station for security reasons, this is the most used kind of measurement. It's almost as good as measuring from a auxiliary ground to the main ground. We're about 50m away from the transformer, but since we have to available grounds (one ground rod and one ground strip in the lawn), we could measure there, too.

 

[radardoug]

50m  from the transformer? How big is the transformer?

 

[RFSiesta]

Sorry for a lot of misinformation now. We had a grounding technician here and the measurements are really embarrassing...
We've been misinformed for a long time, as it seems:

The guy who set the ground rod was sinking it to a mere 1 meter - and the second grounder was not too good as well.
That's what comes out if your mechanic seems to be a fraud....
Since neutral and ground are usually combined at one point, the measurements were faulty from the start: In fact, we have something around 80Ω(!!!) at the ground rod and 20-30Ω at the ground strip.

So on Jan 4. we will get a new ground earth and see what changes by that.
I may change some info in this thread then, as it is very misleading to read these faulty informations if someone will search for similar problems in the future...
:'(

As to the transformer, there's a big office building of an industrial complex between us and the transformer.  I doubt it has any direct influence...

 

[radardoug]

In actual fact if the system is good, it will run with NO ground, as long as all the interlinked grounds are good. It is not uncommon for ground spikes to produce high readings, it all depends on the ground conductivity, and some ground is not a good conductor.
You should start again from square one. Disconnect everything from the console. Then connect a pair of headphones via a small monitoring amp. Verify that this configuration is noise free. If it is not, you first need to find and correct that problem. Then wire up the external devices one at a time, starting with your multitrack/DAW. Again if there are problems, correct them. Use star grounding as far as possible. Wire all mains voltages in via a star system. A separate distribution board for the studio is a good idea.

 

[Monte McGuire]

To further comment on the utility of a ground rod, consider an aircraft at 35,000 feet. Not grounded at all. Or is it?

The point of ground is not some absolute voltage where all electrons go to or flow from, that 'pole in the ground' has to do with the power distribution from the power company into your distribution panel, and the loads attached to it. The typical 2 phase power has the power company using a rod into earth to establish the center tap voltage, which defines the neutral voltage at the two phases on your panel. If you add your own, then you're just messing with the power company's transformer wiring, which may or not be a problem or a help. Quality professional electronics and power wiring that is to modern code should not require "trick" earth schemes to be extremely quiet.

All of that RF you see on the scope could also simply be how you're grounding your scope probe and measuring your circuits. If you're using that dangly clip lead to the side of the probe as your "ground", you could well imagine that it isn't gonna be much of a ground at dozens of MHz. Take the grabber tip off of the scope probe and you'll notice that there's a center conductor surrounded by a 4-5mm metal sleeve. That sleeve is ground, good to hundreds of MHz, not the dangly alligator clip thing off to the side. You can get little spring clips that you can solder to your circuits, or you can rig some stuff up with copper foil to provide a wideband connection from your circuit reference to the probe sleeve. That will allow you to measure your signal relative to this reference, up to a high frequency.

The best way to think about HF "grounding" is not that you're gonna dump it down some rod to earth, but that you make the RF common to everything. If a signal is common to everything, it doesn't exist - it is common to the voltage that you measure all other voltages against - i.e. "ground".

This is the idea behind "ground" in aircraft electronics. The airframe can be floating in the sky, picking up and shedding electrons to randomly charged clouds, but since the airframe is the reference for everything, and it acts like a Faraday cage, then all external fields will be common to the reference and all metal within the Faraday cage, so they do not exist to the circuits. The fact that the Faraday cage is conductive and has no holes is what makes it work - that conductive surface guarantees that the 'skin' will have the same potential around what's inside, so there can be no field gradients within the Faraday cage due to any external field.

This means that you need to ground (at RF) everything to everything else, especially where cables enter and exit a chassis. All chassis should be continuous (or at least overlapping) metal, to provide a Faraday cage. Any holes in the chassis (for ventilation) need to be small relative to the RF wavelengths. If you need to clip a shield at one end for some hum reason, you should re-attach the shield at RF using a small MLCC capacitor from the shield to the cut end. I used some small glass packaged ceramic caps in a studio installation that was close to an AM transmitter, and they worked well. The caps were around 0.1µF, and they kept 1.03MHz on every wire in the studio, so it essentially disappeared.

Best of luck! Once you get your measurement stuff down, then you can tackle any monitor section oscillations, and then hopefully the noise problems in that section. While that console is not a bad one, the modern world has much more RF, and designers have figured out how to get things to work a lot better, despite the ever increasing amount of interference. Do not be surprised if techniques from the '70s fall short today.

 

[PRR]

> aircraft at 35,000 feet. Not grounded at all. Or is it?

Yes, the dirt is not involved.

Distance to power company transformer is usually unimportant.

What the airplane has, and proper house/studio power has, is SOLID connection between all things referencing each other.

A common mystery buzz/hum is when a power strip, ground lift, or bad wall wire is not really connecting "ground" to "ground". I am glad RFS has had his main connections verified. Unlike the US, in UK and I gather in Austria the electrician is required to make verification tests, including with Neutral and PE separated as well as bonded.

> RF you see on the scope

AM/MW RF waves are a hundred meters, police/taxi a few meters, TV and cell even smaller. A 50 meter line back to the power company "ground" is great for 50Hz but terrible for >1MHz. RF grounding is all about local bonding, not to-dirt or to-power bonding.

> something around 80 Ohms(!!!)

That's OK. Not great, but better than a lot of the US.

I drove one new dirt-rod, under the septic field(!), and using 120V excitation measured about 110 Ohms between the rod and my utility power ground-neutral. Utility line had two better rods at my end and multiple pole-grounds along the road, should be under 10 Ohms (if thieves have not been stealing the copper). Now with five rods on my property, I hope I come close to the NEC-suggested 25 Ohms just for my rods. But I know it is higher in drought or when frozen. However due to a too-long unbalanced power feed on the far side of the line, I can have 0.3A or more AC current through dirt, and routinely find 1V-3V of AC at my house between electrical boxes and local dirt. Protective Earth absorbs lightning and keeps stray voltage non-shocking but is NOT a clean technical ground.

 

[RFSiesta]

Yes, the airplane That's why we tested the desk on battery, and with success! The hum was gone completely. It only appears when a device is grounded...

I drove one new dirt-rod, under the septic field(!), and using 120V excitation measured about 110 Ohms between the rod and my utility power ground-neutral. Utility line had two better rods at my end and multiple pole-grounds along the road, should be under 10 Ohms (if thieves have not been stealing the copper). Now with five rods on my property, I hope I come close to the NEC-suggested 25 Ohms just for my rods. But I know it is higher in drought or when frozen. However due to a too-long unbalanced power feed on the far side of the line, I can have 0.3A or more AC current through dirt, and routinely find 1V-3V of AC at my house between electrical boxes and local dirt. Protective Earth absorbs lightning and keeps stray voltage non-shocking but is NOT a clean technical ground.

Very interesting, the fact that we have AC on our ground could not have anything to do with ground rod resistance, if I continue to think about it. If it would, the hum should have changed significally one way or the other by changing from one ground to another or by combining the two... It did not, as far as we could observe.
Still not sure what conclusions we can draw out of this. Is it still a good idea to make the earth rod better?
We hope to get as deep as ground water in our area (6-9m), maybe this makes the dirt go away. I can't imagine AC to be coupled to it.
We would do that on Jan.4th, since the 80Ω seem to be not enough protection, according to our laws. We will also give some feedback to the guys who installed this in the first place, since we paid for more than what we got
Could be, I imagine, that our 50Hz still come from the air, but in theory less voltage will appear over better ground Z - so let's give it a try: We will know better on Jan.5th
Merry Christmas, by the way! And thanks for all the expertise and help,
cheers Richard

 

[PRR]

> as deep as ground water in our area (6-9m)

Conventional Wisdom in US grounding studies is that going deeper is not a lot better. The 8'-10' (3m) rod is selected as a best compromise, though couplers are made for going deeper. Conventional Wisdom is that multiple well-spaced rods are better than any single rod. However if you know you have not reached permanent water level, deeper could help.

> we paid for more than what we got

Dirt resistance, at the same spot, will vary a LOT. Over 10:1 when it freezes. 5:1 from dry to wet. It is possible your rod-banger had a good reading at the time you paid. It is also possible he salted the hole, a trick that will decrease resistance over the first week but tends to rise over the months. Very elaborate grounding at Power Stations and MF transmitters will put down miles of rods and wires and salt-basins which are refilled on schedule.

But unless your studio works on dirt, dirt-rod resistance is probably not the prime problem.

We say "25 Ohms is safe" but note that 120V in 25 Ohms is 5 Amps, will never blow our 15A-20A fuses/breakers. We would need near 3 Ohms to get a quick sure fuse-blow in a ground fault. At ~~100r per 8-foot dirt rod, that is an awful lot of rods for a home or studio to afford.

 

[RFSiesta]

Dirt resistance, at the same spot, will vary a LOT. Over 10:1 when it freezes. 5:1 from dry to wet. It is possible your rod-banger had a good reading at the time you paid. It is also possible he salted the hole, a trick that will decrease resistance over the first week but tends to rise over the months.

I doubt he knew about the salt trick!
The evidence is, actually, as the other tech showed us, there is not enough space in our basement to place a machine for mounting the rod. So he could only have done it with a hammer - there's no way to get more than one rod down through our concrete baseplate with nothing but a hammer...

We say "25 Ohms is safe" but note that 120V in 25 Ohms is 5 Amps, will never blow our 15A-20A fuses/breakers. We would need near 3 Ohms to get a quick sure fuse-blow in a ground fault. At ~~100r per 8-foot dirt rod, that is an awful lot of rods for a home or studio to afford.

For stroke protection, we have so-called FI fuses that react to leaking currents in housings. They react to a mere 30mA, so there would be enough protection with almost any ground resistance. Overload protection is done in reference to neutral, not case earth - ground is not for flowback currents here except if lightning hits the building.

 

[PRR]

I doubt this has anything to do with your audio troubles, but--

> our concrete baseplate

You don't have any outside dirt?

The absolute best building ground is a Ufer ground. When you pour your concrete,bring one of the steel reinforcing rods up above the pour, near your main service panel. Concrete conducts. Weakly, but there is a LOT of it, and far more contact area on dirt than any rod. Yes the multiple re-bars needed to do a floor are not electrically bonded, but they are usually wire-tied and overlapped enough to transfer stress from one to the next, so they are electrically fairly intimate. (Welded wire mesh would provide large areas of metallic conduction.)

I really wish I had done this on my new garage. Sitting below the septic field with 26'x20' area, it might be <10 Ohms added to my ~~30 Ohm present dirt-rods grounding. But I didn't plan to run power to it, and the pour was weather delayed then done in a rush.

And of course if the pour was 1950, it's too late now. (Ufer started in WWII, but didn't publish to 1963, and it was another decade or so before the crazy idea was accepted.)

Another very good ground is a wire buried all around the building. This is especially good for diverting ground current "around" your place, like a moat would divert water running down the hill. Corrosion is a problem so it ends up being a lot of copper (and cost). And if you are so urban that you can't drive an outside dirt-rod, you can't trench around for a surround ground.

> not enough space in our basement to place a machine

Rock drills for underground mines have been around over a century. They can work in less space than a man can stand. True, set-up is major work which may be justified if you plan hundreds of holes over several years, not for one hole.

Multiple rods are better than one. I suppose you don't want a forest of copper rods in your basement.

 

[RFSiesta]

I doubt this has anything to do with your audio troubles, but--

> our concrete baseplate

You don't have any outside dirt?

Well, as for the audio problem - I've run out of ideas. We will try individual large caliber ground cables for audio and see if something changes.

And dirt, yes: part of our building has a basement from concrete, then there is a baseplate with 50cm around the building at ground level. On one side, there's a lawn with trees. And underneath is - dirt. I think mostly loam.
To get a ground rod down there, you must get through the baseplate first.

The absolute best building ground is a Ufer ground. When you pour your concrete,bring one of the steel reinforcing rods up above the pour, near your main service panel. Concrete conducts. Weakly, but there is a LOT of it, and far more contact area on dirt than any rod. Yes the multiple re-bars needed to do a floor are not electrically bonded, but they are usually wire-tied and overlapped enough to transfer stress from one to the next, so they are electrically fairly intimate. (Welded wire mesh would provide large areas of metallic conduction.)

I really wish I had done this on my new garage. Sitting below the septic field with 26'x20' area, it might be <10 Ohms added to my ~~30 Ohm present dirt-rods grounding. But I didn't plan to run power to it, and the pour was weather delayed then done in a rush.

And of course if the pour was 1950, it's too late now. (Ufer started in WWII, but didn't publish to 1963, and it was another decade or so before the crazy idea was accepted.)

I'm almost certain we don't have such a grounding system here. The building was built as a simple utility building at the time.

Another very good ground is a wire buried all around the building. This is especially good for diverting ground current "around" your place, like a moat would divert water running down the hill. Corrosion is a problem so it ends up being a lot of copper (and cost). And if you are so urban that you can't drive an outside dirt-rod, you can't trench around for a surround ground.

> not enough space in our basement to place a machine

Rock drills for underground mines have been around over a century. They can work in less space than a man can stand. True, set-up is major work which may be justified if you plan hundreds of holes over several years, not for one hole.

Multiple rods are better than one. I suppose you don't want a forest of copper rods in your basement.

They don't necessarily drill, the rods get pushed by brute force with an electric hammer. That's how most do it here, if possible.
We will know more soon. At some point in the basement there's enough headroom for the machine, underneath the staircase. There we can add more rods to go much deeper than the last one, the tech said. It's only 40€/rod, galvanized. So it's not too costly to just try reach ground water.

 

[radardoug]

As I have said previously, and others have illustrated with the aircraft example, in a well working system you dont need ground. You do need it for safety. But for the purpose of faultfinding in your audio gear , you could try it with no ground. Just exercise sensible precautions, measure the leakage to ground, and be aware of it. I have run a 30,000 watt p.a. outdoors with no ground, it worked fine.

 

[RFSiesta]

As I have said previously, and others have illustrated with the aircraft example, in a well working system you dont need ground. You do need it for safety. But for the purpose of faultfinding in your audio gear , you could try it with no ground. Just exercise sensible precautions, measure the leakage to ground, and be aware of it. I have run a 30,000 watt p.a. outdoors with no ground, it worked fine.

Of course it's strictly forbidden here to use any 230V eqipment without ground. As I stated, we used the desk on battery and it did not hum! Unfortunately, this is no solution.
Thus we need to find the source, and of course it won't depend on any "absolute" zero-potential, it's always a differential. In our case a differential between two earth cables from the same junction to one device. If I understand correctly, the source could be a 50Hz magnetic field all over the building, but we could not make out differences when moving the corresponding cables.

So I figure it could as well be the other way around, earth is floating with 50Hz and builds a magnetic field that leads to currents because there's a 50Hz field differential to the surrounding area. Does this make sense? Then the current flows not only inside the (quite small) loop we constructed, but inside the whole electric wiring of the building. Because impedances can't be exactly identical in two different cables/connectors, there will be a voltage over the signal cable shield - that affects audio.

Guess what we found: The hum can be clearly influenced by moving one of the power connectors, if they are not perfectly clean. I think if the impedance is the same, then the hum is gone. That's no evidence, though, just a hint imo.

Something else we found: The hum is NOT directly correlated to any load in the building, neither concerning gain nor PHASE - Isn't it weird that what we seem to have is 90° phase shift from reflowing currents in neutral to our hum signal?

 

[nielsk]

Are you are performing these tests with NOTHING but a set of speakers interfaced to the console?
It is a good approach to follow the basic new install sequence when chasing mystery noises, i. e. power up console with ONLY one set of speakers connected to it, when that is clean, connect a 2 track playback source, test, then interface the multi-track, test, then ancillaries, and so on.
If  the console is clean running on batteries, you should look closely at the power supplies.
There are usually grounding options between the power supplies and the console, you should be able to run it (for test purposes) with the console ground connected to the P/S common, with NO technical or safety ground connected to the mains or anywhere else.
It is well within the realm of possibility that both of the supplies you own have issues, scope the power rails at the supply and at the console and determine if you have clean & correct DC on all rails.

 

[radardoug]

Fully agree with Nielsk!!!! You MUST do this first.

 

[pvision]

The upside of all this hard work will be an electrical installation you know is safe and can be confident in. That will be a good position to be in!

Nick Froome

 

[PRR]

> what we seem to have is 90° phase shift

Again, *probably* not related to your audio problems.

But this is a full can of worms.

3-wire center-tap power lines deliver twice the power as 2-wire lines, for same line-to-earth voltage. Even Edison learned this.

In AC systems, 4 wire "wYe" connection with three "phases" 120 degree apart do a little bit better. Because electric profit is a lot about wire-cost, essentially all larger systems are 3-Phase.

3-phase power is not run to most homes. At this lower power level the economics of 3-phase are not so good. You need three transformers, but the smallest single pole-pig (or ground mount) transformer will power several homes. Here, I would have to look like a business (with large power consumption) to get 3-phase. (Even then, while my long street is clearly planned for full 3-phase, only two phases worth of line and transformer are installed. We have the third transformer brackets and insulators, empty until more customers come to this neck of the woods.)

3-phase has been seen in large houses and would be routine in an urban utility business.

Your Main Panel will have a 3-pole Main Breaker if it is 3-phase.
https://images-na.ssl-images-amazon.com/images/I/414vfM1buXL.jpg

This is 120 degree between phases.

Before 3-phase, in the US a 2-phase system existed. The two phases were 90 degrees apart. This is sufficient to start and run very large motors without the starter-tricks needed on single-phase systems. It is marginally less economic than 3-phase. While you can convert 2- to 3-phase, it is ugly. I believe all these systems are long gone.

To add confusion: the US's 240V CT home supply is called "2 phase". This is slang and should be avoided. In electric power the real point of a "phase" is to get an angle on the dead-spot of a 2-pole motor. 90 deg works, so does 120 deg, but 180 deg does not. Picture a bicycle and a rider with bad knees. If the pedals are stopped straight up/down, he can not get started (except by knee action). If pedals were rigged 90 or 120 deg apart, he could press on one or the other and get started. So 240V CT (and single 230V) is single-phase, not 2-phase.

 

[RFSiesta]

Are you are performing these tests with NOTHING but a set of speakers interfaced to the console?

Maybe I should state clearly that this thread has developed to be a bit off-topic now, as the "Trident HISS" issue seems to be in most part solved, and the hum problem is not at all restricted to the Trident!

Let me again describe the easiest possible setup we use for testing the hum:

1.) Any little mixer or electronic device with a non-balanced audio OUT and a power cable
2.) Any guitar/bass etc. amp with input jacks
3.) both devices connected via instrument cable AND
4.) both devices connected to 230V at the same junction of a power distributor.

Only the amp needs to be switched on. As soon as we connect the first device to ground, the amp begins to hum.

We drew these conclusions:

- It doesn't have to be a large ground loop in the house installation, the loop from two power cables and one instrument cable (shield) is enough,

- It has nothing to do with PSUs of any device, we tested with all different amps, mixers etc. The first device isn't even ON in the test, and it makes no difference if it is or not. Hi-Z instr. inputs *seem* to be more sensitive than lo-Z, like mic ins.

- We tried to move the corresponding cables, power and/or instrument; pickup from air would correspond with the m2 of the area the cables form, because that's what an antenna does. So moving the cables close together would result in less hum: It doesn't.
That's why we concluded the hum is propably NOT coming from magnetic fields...

3-phase power is not run to most homes

3 phases are common in Austria and many countries of europe, I think. We even have our kitchen stove running on 3 phases. That said, it does not make any difference if it is on or not, like all the other high current loads.
Apart from the stove, the phases are all dedicated: One phase is ONLY for audio, one is ONLY for lights/heating etc, and one is for digital devices and all the dirty rest

It could be a 120° phase shift, btw. Have to check with Andy for this. But the power company specialist found no correlation with our power consumers in the building, he said it does not come from loads here. We can switch on/off whatever we have, electrical heating, stove, monitors, anything without affecting the amplitude of hum. And since we don't have a  reliable reference whatsoever, we can only estimate the amount of AC we have on our ground...

As radardoug state before, it should make no difference if the AC is everywhere the same and there is no differential voltage anywhere - and that exactly is what puzzles us. The hum is there but doesn't react like pickup hum I had in other locations, rehearsal rooms etc.