Should phantom power be supplied by a linear supply?

[moamps]

..You mean the older ones had no switchers at all?

Correct.

SMPSs are great, but IMO are sometimes used not to improve device functionality, but only to reduce production and transportation costs.

 

[Winston OBoogie]

Correct.

Indeed.
Although I didn't do the LA-610, I did do the design stuff on the separate UA 610 devices & the first version of the 610 + 1176 box whatever that was called.

I didn't have the first idea about how to design switchers so it wasn't really an option.

Looking at the layout that does have switchers, I'm a little concerned that the main PCB looks to be heading in the direction of being a disposable item.
Probably not easy for the average tech to repair it if something goes wrong. Especially if UA have a "no schematics available" policy. Which they do.

Which then brings into play whether the efficiency of a switching power supply ends up actually being "greener" for the planet if a bunch of PCB's end up in a landfill.

That said, with correct implementation, I don't see a problem with switchers.

 

[abbey road d enfer]

SMPSs are great, but IMO are sometimes used not to improve device functionality, but only to reduce production and transportation costs.

Do you know any other motivation than profit for manufacturers?

 

[Blackfinger]

Hi, designing a power supply for a modular preamp box I'm putting together - I'm currently running +/-15vdc out of a center tapped transformer for power for the pres themselves, and was a little curious as to your guys' thoughts on the impact running phantom power through a cheaper switching supply would have on the sound - I'm assuming any small sonic difference from a switching supply on a mic's internal circuit would be amplified once it hits the pre stage, but I know Douglas Self's small signal audio design recommended a voltage tripler hooked up to the positive rail.

Any thoughts?

My 2c:

I would go with a simple linear supply and utilize a TL 783 regulator. The current required for phantom power is minimal with respect to the amount of ripple that might appear at the supply output. Follow your raw rail voltage with a TL783 3 terminal regulator (with 10 – 100uf at it’s output pin) and you will benefit additional ripple reduction of the regulator circuit itself. Add to that – the output of your supply should be split across two 1% 6k8 resistors to the balanced pins 2&3 of the mic input connections – phase cancelling any supply noise across your input even further.

Unless you are looking to feed a ton of mics, a small compact 48vac secondary power transformer feeding a FWBR with 6800uF/80V on the raw rail should provide plenty of pre regulator filtering. That should provide about 67Vdc raw rail to the Vreg input. Two diodes (1 between Reg input and output, one at Reg output) will provide short circuit protection. The 783 is good to 700mA. You will want to fix a minimal load resistor (15mA) on the Vreg output for stability or a 20ish mA status LED will do the trick nicely.

Load/no load conditions nor heat dissipation should be a problem with the 783. This reg does have a huge differential (125v) between in & out to work with, so it’s not hard to feed. The only issue you may run into is if you choose an input rail that is very high – relying on the regulator to dissipate a lot of power. In that case you will need a robust heatsink. I would certainly heatsink the reg anyhow. Go linear! Good luck

 

[JohnRoberts]

Do you know any other motivation than profit for manufacturers?

saving the planet? Channel your inner Greta.

JR

 

[abbey road d enfer]

My 2c:

You're correct. This thread has drift into a anti vs. pro SMPS tirade; and lost both the original subject and the original poster.
I'm certainly not against drifting, for the sake of openmindedness, but when the captain has lost sight of the harbour, it's a good thing someone with a fresh eye put us back on track.

 

[Matt Syson]

6800uF 80 Volt is a massive overkill for this where a couple of say 2200 with a resistor of 47 Ohms between them would give a cleaner supply into the TL783, which needs to have it's minimum current catered for which realistically requires a 1 watt (minimum) resistor to ground. (about 10milliamps of 'bleed should be allowed for. You could do this with two resistors and have a LED in one of them to show the phantom is on if you wished.
Besides, the startup current for 6800uF is getting rather 'fierce' for a small transformer and rectifier.

 

[Newmarket]

You will want to fix a minimal load resistor (15mA) on the Vreg output for stability or a 20ish mA status LED will do the trick nicely.

That's some LED current @ 20mA

 

[Matt Syson]

Around 8milliamps is 'acceptable' for a 'not high efficiency' LED so you would calculate two resistors to give around that current.

 

[JohnRoberts]

I'm old enough to remembers when LEDs only came in red, and needed 20 mA.

These days high efficiency red needs more like 2 mA,

JR

 

[ruffrecords]

6800uF 80 Volt is a massive overkill for this where a couple of say 2200 with a resistor of 47 Ohms between them would give a cleaner supply into the TL783, which needs to have it's minimum current catered for which realistically requires a 1 watt (minimum) resistor to ground. (about 10milliamps of 'bleed should be allowed for. You could do this with two resistors and have a LED in one of them to show the phantom is on if you wished.
Besides, the startup current for 6800uF is getting rather 'fierce' for a small transformer and rectifier.

I agree 6800uF is overkill. My 2011 design used 4700uF but that was only required if you needed the maximum of 100mA of phantom current ( for a 16 channel mixer ofr example). For most designs 2200uF is plenty. You need at least 48VAC supply to the bridge so you get close to 70V of raw dc feeding the TL783 ( it needs about 20V across it to regulate and ripple reduce properly). Schematic attached (the 1N5407 diodes are sever overkill).

Cheers

Ian

 

[Matt Syson]

The Cap/Res/cap/res and even a third cap/res can remove a fair amount of ripple in itself and more effectively than one 'big' cap as proven with valve gear many years ago.
Although you are unlikely to have a problem, the datasheets for most adjustable regulators suggest a diode (in parallel with) the 120 Ohm resistor as shown by Ian helps protect the regulator against some possible abuse. Peak currents from even small capacitors can be quite large.

 

[abbey road d enfer]

Actually my first version of active mic splitter used an unregulated phantom power, with plenty of caps. Nobody ever complained about noise.
For the second version, I used a TL783 for each set of 24 channels, just because I wanted to be in accordance with the standard that specified +/- 4V. Some splitter manufacturers used this tolerance to remotely control a pad, a somewhat risky option when one knows how phantom is (badly) handled when there are a FOH mixer, a stage monitor mixer, an OB van and a mobile studio.

 

[Bo Deadly]

I wanted to be in accordance with the standard that specified +/- 4V. Some splitter manufacturers used this tolerance to remotely control a pad

It seems to me that it would very plausible to send (and receive) high frequency pulses of data over the signal lines and then you could control all sorts of things. A PIC10 micro comes in a tiny 2x3 DFN or 6 pin SOT-23 and costs $0.50 USD. With a little bit if LC filtering (which you probably want anyway) you wouldn't hear anything in the audio range. The biggest challenge would probably be balancing the impedances on the signal lines so that you don't ruin CMR.

 

[JohnRoberts]

Stop me if I've shared this one before...

Back in the early 80s I kicked around the idea of transmitting gain control instructions to remote phantom powered preamp stages using very slow changing voltages on the + and - XLR wires, that could be easily filtered out of the audio signal.

I abandoned this idea as not commercial.

JR

 

[abbey road d enfer]

For a brief period (about 10 years) there was an interest in stage mounted active mic splitters and remote control of a few parameters (pad, rough gain, HPF) from one of the mixers. Some tentatives were made using ultrasonic sinewaves and PLL detectors. Since the commands were set-and-forget, there were no problems with beating frequencies. However, the added complexity, particularly regarding maintaining performance AND the fact that the remote ontrol unit needed to be interfaced with the (proprietary) wiring system made it not enough attractive.

 

[Bo Deadly]

This is drifting off topic but I think the key challenge would be finding a good frequency where the various LC filters work well. At low frequencies it might be difficult to make a "trap" to isolate the load of the digital driver. At high frequencies, transmission line effects become a problem with the relatively high capacitance and unpredictable balance of a conventional analog cable with an unknown device attached. Sounds like a fun problem though.

 

[JohnRoberts]

The challenge is finding customers willing to pay for it...

JR

 

[ruffrecords]

Stop me if I've shared this one before...

Back in the early 80s I kicked around the idea of transmitting gain control instructions to remote phantom powered preamp stages using very slow changing voltages on the + and - XLR wires, that could be easily filtered out of the audio signal.

I abandoned this idea as not commercial.

JR

Very many years ago mixers were built with all the active electronics in an external rack and the board itself was 100% passive but that was in the days when all the amps were fixed gain. In the late 70s? Neve developed several consoles ( I think they were for George Martin) which had remote mic pres but the gain controls were still on the console and which remotely controlled mic pre gain.

Cheers

Ian

 

[moamps]

Hi, designing a power supply for a modular preamp box I'm putting together - I'm currently running +/-15vdc out of a center tapped transformer for power for the pres themselves....

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?