Should phantom power be supplied by a linear supply?

[ruffrecords]

For the 100th reply, can I ask why you want to use SMPS for phantom power at all? Why not use a voltage tripler like the one used in Soundcraft consoles or similar?

One reason would be the voltage tripler connects the phantom power 0V to analogue 0V at the power supply when what you need is a floating phantom supply with its zero volts connected to the chassis rather than analogue 0V.

Cheers

Ian

 

[moamps]

One reason would be the voltage tripler connects the phantom power 0V to analogue 0V at the power supply when what you need is a floating phantom supply with its zero volts connected to the chassis rather than analogue 0V.

Really? Why? So almost all 500 type racks are faulty? And all Soundcraft power supplies IIRC? There are other examples where the phantom power supply reference is connected to analog 0V.

 

[ruffrecords]

Really? Why? So almost all 500 type racks are faulty? And all Soundcraft power supplies IIRC? There are other examples where the phantom power supply reference is connected to analog 0V.

You can often get away with it on a small scale item like a 500 rack but it is not best practice.

Cheers

Ian

 

[moamps]

I thought this was a thread about small scale rack, OP uses 7815/7915 power regulators for rails, it can't be big scale console .

Also this concept of “the chassis connection” is quite imprecise. There are many examples where the chassis is not made of an ideally conductive material. It can be made of painted iron or anodized aluminum, it can also be made of several plates (sometimes a combination of aluminum and iron plates) and not of one piece of material, so these plates are sometimes unintentionally electrically insulated. Also, after some time, corrosion can occur at the joints of such materials (initiated by the current flowing through the plates).
So the idea that "the chassis connection" will unconditionally make the ideal electrical contact of pin 1 of the XLR with 0V from the phantom power supply is wrong, that is, to assume in advance that the chassis is at all its points at one electrical reference potential is completely wrong, IMHO.

 

[ruffrecords]

So the idea that "the chassis connection" will unconditionally make the ideal electrical contact of pin 1 of the XLR with 0V from the phantom power supply is wrong, that is, to assume in advance that the chassis is at all its points at one electrical reference potential is completely wrong, IMHO.

I did not say that. Even in my relatively small 6 channel tube mixers, I habitually bus mic inputs XLR pins 1 together with thick copper and connect the floating phantom. 0V to its centre. Each XLR pin 1 is also connected to chassis at the connector supply. The analog 0V is connected to chassis next to the mains IEC connector. All exposed metal MUST be electrically connected to safety ground otherwise the device could be lethal. This is just normal good engineering practice. As I said, you can get away with it on a small item but why compromise?

Cheers

Ian

 

[sahib]

One reason would be the voltage tripler connects the phantom power 0V to analogue 0V at the power supply when what you need is a floating phantom supply with its zero volts connected to the chassis rather than analogue 0V.

Cheers

Ian

Hi Ian,

This is no different than connecting the phantom 0V to analogue 0V at the chassis stud point.

.....I habitually bus mic inputs XLR pins 1 together with thick copper and connect the floating phantom. 0V to its centre. Each XLR pin 1 is also connected to chassis at the connector supply. The analog 0V is connected to chassis next to the mains IEC connector. All exposed metal MUST be electrically connected to safety ground otherwise the device could be lethal. This is just normal good engineering practice. As I said, you can get away with it on a small item but why compromise?

Cheers

Ian

This again becomes same as the previous one.

This time your floating phantom 0V is connected to the analogue 0V through the thick copper wire and the metal case chassis itself.

What is most important is how you treat +48 ground on the module/backplane pcb.

I tend to side with moamps on the phantom psu issue.

Edit:
Ian, I get what you mean. By running a copper wire for the phantom 0V you prevent phantom return currents running through the case chassis.

In the case of 500 racks, as the XLRs are mounted onto the backplane the phantom 0V is taken care of at the backplane. So the phantom return currents do not flow through the chassis.

 

[ruffrecords]

Edit:
Ian, I get what you mean. By running a copper wire for the phantom 0V you prevent phantom return currents running through the case chassis.

That is exactly what I mean because I use metal XLRs bolted to the metal chassis

In the case of 500 racks, as the XLRs are mounted onto the backplane the phantom 0V is taken care of at the backplane. So the phantom return currents do not flow through the chassis.

Which also works because AFAIK 500 series racks use plastic bodied XLRs.

Cheers

Ian

 

[Bo Deadly]

In the case of 500 racks, as the XLRs are mounted onto the backplane the phantom 0V is taken care of at the backplane. So the phantom return currents do not flow through the chassis.

Pin 1 is the shield and body of the mic. So it should be connected to the chassis. Period. The fact that it's also used as phantom return is incidental. Why exactly do you think it would better for phantom power return currents to not run through the chassis? As long as the wire connecting the chassis bolt to the supply 0V at the last filter cap is short and extra thick, that should not be a problem even for a lot of phantom loads.

 

[moamps]

Pin 1 is the shield and body of the mic. So it should be connected to the chassis. Period. The fact that it's also used as phantom return is incidental. Why exactly do you think it would better for phantom power return currents to not run through the chassis? As long as the wire connecting the chassis bolt to the supply 0V at the last filter cap is short and extra thick, that should not be a problem even for a lot of phantom loads.

The phantom power 0V should be connected to pin 1 of the XLR (and pin 1 should then be connected to the chassis) and not somewhere to the chassis just because according to the pin 1 problem/solution philosophy, currents induced by EM interference flow through the chassis. And no one wants the phantom power currents and all the other interferences to mix.

 

[Bo Deadly]

The phantom power 0V should be connected to pin 1 of the XLR (and pin 1 should then be connected to the chassis) and not somewhere to the chassis just because according to the pin 1 problem/solution philosophy, currents induced by EM interference flow through the chassis. And no one wants the phantom power currents and all the other interferences to mix.

I'm not sure I understand exactly. Are you suggesting that there should be a wire or trace from pin 1 to 0V running within the enclosure?

 

[moamps]

I'm not sure I understand exactly. Are you suggesting that there should be a wire or trace from pin 1 to 0V running within the enclosure?

Yes, and please read my reply #104 about what I think about chassis electrical imperfection.

 

[Bo Deadly]

Yes, and please read my reply #104 about what I think about chassis electrical imperfection.

Ok but if the chassis panels have imperfect contact, that's a separate issue. Generally I don't advise people to do something in a non-ideal way to compensate for the possibility that they're incompetent.

 

[sahib]

Why exactly do you think it would better for phantom power return currents to not run through the chassis? As long as the wire connecting the chassis bolt to the supply 0V at the last filter cap is short and extra thick, that should not be a problem even for a lot of phantom loads.

I can see that moamp has replied.

However, because running any circuit return current, let alone phantom, through the case chassis is a lousy practice, particularly from compliance point of view. No serious designer would do a thing like that.

The case is the shield to the circuitry placed inside. The only current that is allowed to run through it to the chassis stud point is the noise current that is coupled to the case chassis.

 

[Bo Deadly]

However, because running any circuit return current, let alone phantom, through the case chassis is a lousy practice, particularly from compliance point of view. No serious designer would do a thing like that.

That is just false. Running phantom return through the chassis is the correct thing to do. The only case were it is not done is if the chassis is simply not metal or for practical manufacturing reasons.

The case is the shield to the circuitry placed inside. The only current that is allowed to run to the chassis stud point is the noise current that is coupled onto the case chassis.

Pin 1 of the XLR is connected to the chassis. So how can current NOT return through the chassis? If you add a wire from the XLR pin 1 to 0V at the supply there's a strong possibility that you're just radiating RF into your enclosure. Currents always follow the path of lowest impedance and unless the panel screws are corroded or loose (which is not an impossibility but highly unlikely and not something I would design around), then the chassis is going to be the path of lowest impedance.

 

[sahib]

That is just false. Running phantom return through the chassis is the correct thing to do. The only case were it is not done is if the chassis is simply not metal or for practical manufacturing reasons.

It is not false. It is true. But you are free to do it your way.

No serious designer would mix noise and circuit currents at any point, let alone on a case chassis. At least now, in modern days. As some serious designers made that mistake in the distant past.

Pin 1 of the XLR is connected to the chassis. So how can current NOT return through the chassis? If you add a wire from the XLR pin 1 to 0V at the supply there's a strong possibility that you're just radiating RF into your enclosure. Currents always follow the path of lowest impedance and unless the panel screws are corroded or loose (which is not an impossibility but highly unlikely and not something I would design around), then the chassis is going to be the path of lowest impedance.

This would be true if you connected phantom 0V to chassis earth at the chassis stud point inside your equipment. Consider Ian's scheme. He does not connect the phantom 0V to the chassis. Instead he connects it directly to XLR pin 1. Now, the phantom return current runs through that wire into the phantom 0V. The noise current coupled onto the case chassis runs to the chassis stud point and finally into the mains earth as his IEC earth tab is connected to the chassis stud point.

 

[ruffrecords]

Pin 1 is the shield and body of the mic. So it should be connected to the chassis. Period. The fact that it's also used as phantom return is incidental. Why exactly do you think it would better for phantom power return currents to not run through the chassis? As long as the wire connecting the chassis bolt to the supply 0V at the last filter cap is short and extra thick, that should not be a problem even for a lot of phantom loads.

Because interference currents also flow in the chassis and hence could appear on the mics 0V supply.

Cheers

Ian

 

[My3gger]

This Jensen schematic looks very clear about it; trafo's phantom tap goes to psu, then both wires (+-) connect to R1,C1 filter on a switch without touching ground before it. Ground wire from C1 minus pole is connected to XLR pin 1 which is bolted to chasis at the screw this XLR. Then + wire from joint of R1,C1 go to joint at 6k8 resistors which are then connecting pins 2,3 of the same XLR.
I just confirmed it on 1ch preamp. It has always worked well and is probably quietest tube preamp (2 triodes) i ever measured, with phantom on or off.
https://www.jensen-transformers.com/wp-content/uploads/2014/08/as016.pdf

 

[Bo Deadly]

Because interference currents also flow in the chassis and hence could appear on the mics 0V supply.

So do you not connect pin 1 of XLR to the chassis directly at the entry point?

 

[ruffrecords]

So do you not connect pin 1 of XLR to the chassis directly at the entry point?

No, I DO connect pin 1 directly to the chassis. I also connect the phantom 0V to pin1 (not to the analogue 0V).

Attached is my approach documented.

Cheers

Ian

 

[sahib]

@ My3gger. You have to be extremely careful when you are interpreting schematics. In isolation, that can also be interpreted as a bad practice.

As Ian mentioned before in smaller distances it will be fine, as it is the case in your equipment, but what happens if you have a steel chassis of say over 1m and you have the first XLR 1m away from the chassis stud point (earth). Now, you inserted a considerable resistance between the phantom 0V (as it is connected to chassis stud point) and the pin1 of the first XLR. Also think about the future when, say, screw joints get corroded etc.

I do not mean to pretend as an expert console designer. I am certainly not and I have never designed anything like that, but as a principal, letting a circuit current depend on a metal chassis alone does not sound like a good engineering practice, let alone circulating it with noise currents across the equipment chassis.

In the case of 500 racks, on our rack we have a separate solid earth plane on the backplane PCB. Yes, at the end of the day the earth connection to the metal chassis, and the connection to the earth plane on the backplane join at the DC power connector fitted on the rack rear panel, but we do not rely on XLR pin 1 to get its phantom 0V through the metal case. This a bit like Ian's scheme except that at external power supply we tie the phantom 0V to chassis stud point. Schematic attached.