Maida BIAS regulation grounding question

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm looking to set up regulated B+ and bias supplies on my power amplifier. On Page 41 of the 21st Century Maida Regulator thread, Tom states:

You should be able to generate a positive and a negative voltage with two Maida Regulators in series if the conditions below are met.

Conditions: The two Maida Regulators must be fed from separate transformer windings, separate rectifiers, and separate reservoir caps. Basically, the input circuits must be completely separate. You then put the two regulators in series on their outputs and ground the centre point between the two.

My understanding is that when the signal ground is taken to be the point between the two regulators connected in series (and not connected to anything else), the signal ground is at a positive potential with respect to an earthing rod (or human skin, under realistic conditions).

Given that fact, I would assume safety ground (earth/chassis) should be kept separate, i.e. disconnected, from signal ground in this configuration. However, I'm wondering what the ramifications would be when connecting other devices to the tube amplifier. Wouldn't that raise their signal ground voltage to that of the bias supply (not only causing a ground loop but also causing a safety hazard if their chassis is connected to signal ground)?

If my understanding is correct, this would be very dangerous when using a preamplifier without safety ground, for example. And if the signal ground is to be connected to mains safety ground, the chassis would be live at a DC potential if the safety ground connection fails.
 
Last edited:
I understand that the bias supply and B+ supplies are isolated from each other. My question pertains to the function and safe use of the signal ground point developed where those two supplies are connected in series.

Essentially, if you build an amplifier around this power supply and the signal ground is not connected to a proper earthing rod, when you touch an RCA connector on the amplifier, you will be touching the output of the bias regulator.
 
Last edited:
Essentially, if you build an amplifier around this power supply and the signal ground is not connected to a proper earthing rod, when you touch an RCA connector on the amplifier, you will be touching the output of the bias regulator.

Whenever you have two independent electrical components connected together where signal grounds on each component are at different potentials, you create conditions for a ground loop. In other words, current will flow through the connecting cables from the one ground potential to the other. This situation is not specific to just this type of pos/neg power supply scheme.
 
Sure. But what I'm really wondering about is safety to the end-user. I just don't see how anyone would consider this scheme of attaching the chassis to a high voltage source safe when the only thing preventing an electric shock to the user is the mains ground. That would never pass in a commercial product. Am I missing something?
 
ahanuban said:
My question pertains to the function and safe use of the signal ground point developed where those two supplies are connected in series.
No signal ground point is established when two separate floating DC supplies are connected in series. You may, of course, choose to regard the junction of the two supplies as a signal ground point - but it is you who makes this happen (e.g. by using this as the reference point for signal voltages), not the PSU circuits.

Essentially, if you build an amplifier around this power supply and the signal ground is not connected to a proper earthing rod, when you touch an RCA connector on the amplifier, you will be touching the output of the bias regulator.
Not if the supplies are floating. You seem not to understand what is meant by 'floating'. It may be that you do not understand that there is no such thing as a voltage: all voltages (with no exceptions) are between two points.

Right, but what I am asking is not whether they can be connected; rather, I am asking what will happen if they become disconnected, for example in the event of faulty mains wiring.
Faulty mains wiring cannot disconnect two things which it never connected - it is the DC outputs of the PSUs which are connected (by you - not by the mains wiring). However, I suppose that really badly faulty mains wiring could connect things which are not supposed to be connected - but this possibly means that the fault was built in by the builder not understanding his wiring.

I was really hoping for some constructive help in understanding and mitigating risk here, not vague and unexplained answers.
I think some people may be puzzled by your apparent lack of understanding of basic electronics. That makes it difficult to explain things. You need to start again, wiping from your mind the assumptions you stated in your first post. First, understand what is meant by voltage (and that it always means two points); then think about what is meant by an floating supply (and remember that a mains transformer isolates the mains from the DC side).
 
You may, of course, choose to regard the junction of the two supplies as a signal ground point - but it is you who makes this happen

I really am aware of this. Perhaps my question was simply not clear. When I refer to the signal ground, of course I am referencing the point between the two regulators as signal ground. Isn't that the whole point of "grounding" that point as tomchr described-- so that the rest of the circuit is referenced to that point and the negative input of the first regulator becomes the bias supply?

You seem not to understand what is meant by 'floating'.

Something is floating when it is not connected to anything. My understanding is that the rectifier and filter circuits for the two maida regulators are completely independent, and the only interaction between the two regulators occurs at the point at which they are connected in series, which is taken to be the signal ground reference point of the circuit. This is the whole premise of the bias regulation, is it not? Once that point between the regulators is chosen as signal ground, it is connected in the circuit to various other points which should be at signal ground potential, including the ground of the RCA inputs when using unbalanced audio, as well as the common point on the speaker outputs. If we are working with a power amplifier, how would the common point between the two Maida regulators be floating with respect to an external unbalanced preamplifier?

all voltages (with no exceptions) are between two points

Of course. While I appreciate your effort to simplify things, I'm really not that new to this. Again, perhaps I was not being clear. I hope this post clears up any confusion.

My whole question is regarding the fact that the signal ground is chosen to be the point between the two regulators, and that this puts the ground of the RCA jacks at a positive DC potential with respect to human skin (or an earthing rod). If something is wrong with that assumption, then I am misunderstanding the premise of connecting the two regulators together to make a regulated bias supply, not lacking basic electronics knowledge.
 
Last edited:
Here is some further clarification. If my initial question makes no sense, let's start here.

Tomchr said that you could make a regulated positive supply (B+) and negative supply (C-) using two Maida regulators in series. This makes sense... You run one regulator on positive DC rectified off a transformer, and the second regulator takes the output of the first regulator voltage to be its negative input and the positive input is a separate rectified DC supply for B+. The C- supply is taken as the voltage between the positive output of the first regulator and the negative input to the first regulator. The B+ is taken as the voltage between the positive output of the first regulator and the positive output of the second regulator.

Now, in order for the negative supply to be used to bias the output stage, it is my understanding that we must connect that centre point between the regulators to signal ground. Thus, the cathodes of the output tubes are at signal ground potential, and the grids are fed with the negative bias voltage.

If our signal ground is coming from the output of a voltage regulator, it's reasonable to assume that when it is not connected to any other reference point, it is at a positive potential with respect to a grounding rod or human skin.

The signal ground is connected to the ground of RCA cables when using unbalanced audio. As someone mentioned previously, the use of two different ground potentials creates a potential ground loop. Also, touching the RCA cables or anything using them as a ground could cause an electric shock.

If we then choose to connect the signal ground to chassis and safety ground (from the mains plug), the safety of the device rests entirely on proper earthing of the mains outlet.

Please tell me if anything here is incorrect. This is what I have been trying to ask; I'm sorry if it was unclear.
 
Last edited:
Tom replied to my emails and explained that the situation I described is correct (assuming signal ground is tied to chassis/earth), so a good mains ground is absolutely critical and any failure in the ground will present risk to the user. Of course, the ground point could be taken at the negative input to the first regulator, but there would still be the issue of the RCA jacks being at an elevated potential.

Unless anyone has something to add, no further replies are needed.
 
ahanuban said:
If our signal ground is coming from the output of a voltage regulator
That is your basic mistake. The signal ground does not come from a voltage regulator; you apply it to one side of a voltage regulator (ar anything else).

If we then choose to connect the signal ground to chassis and safety ground (from the mains plug), the safety of the device rests entirely on proper earthing of the mains outlet.
No. If the supplies are floating, as you say, then there is no shock hazard from the DC. You actually create a DC shock hazard by grounding some point in the PSU, but on balance this is still worth doing because it reduces the risk of an AC shock from loose mains wiring. Current, including shock current through a person, always (no exceptions) flows in a loop. That loop may include stray capacitance, but stray capacitance can have high impedance at 60Hz.

Either you are not describing what we think you are describing, or there is some fundamental confusion.

You say you understand basic electronics, yet your confusion suggests otherwise.
 
That is your basic mistake. The signal ground does not come from a voltage regulator; you apply it to one side of a voltage regulator

True. Semantics, really. I think the meaning (that the signal ground reference is taken at that point) was clear.

Current, including shock current through a person, always (no exceptions) flows in a loop. That loop may include stray capacitance, but stray capacitance can have high impedance at 60Hz.

Either you are not describing what we think you are describing, or there is some fundamental confusion.

You're completely right. I was neglecting the isolation of the bias supply (through the power transformer) from ground. I guess when talking about the supplies "floating" I was thinking about the lack of connection between the two supplies, not the supplies and ground. This is why I should not stay up all night working on PCB layouts. Brain stops working.

I appreciate your patience and clear explanation! :cheers:
 
Forget the term Ground, ground is where you plant potatoes...
Think instead of signal returns, especially if you are going to do your own PCB's.
Finally the third mains connection in the UK is the Protective Earth, used for fault currents only.
A signal return paths (0Vs) does not have to be connected to protective earth to function.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.