Should phantom power be supplied by a linear supply?

[JohnRoberts]

I kinda gotta wonder too. How many SMPS enthusiasts advocating on here are concerned, or even _aware_ about complying with regulatory agency emissions requirements? This is a question of target audience then. If you want to build _one,_ the FCC probably ain't pulling up with the cat detector van to razz you about it. 'More than one,' maybe different story. But I'm still following here. Meanwell LED supplies, capaciatnce multipliers, desks that run electric bills up. Things change. We'll see.

Hello... This is old news to major manufacturers. Several here have real industry experience. At Peavey we had our own in house test lab to check SKUs before submitting them for agency approval.

CE has different emissions standards and some degree of self certification so may be fast and loose with small manufacturers. (IIRC CE included a stipulation that SKU inputs would resist interference, pretty much opposite the US requirement that puts responsibility on the source of interference.)

JR

 

[Matt Syson]

Studer have used switchmode supplies for a long time to get a split rail and phantom from a single 24 (or thereabouts) DC input so as ever things can be done but just bunging a cheap 'wallwart' or equivalent supply into a piece of gear you want to tote as 'high quality' requires a bit of thought and the ability to check what is going on well above the nominal audio band. While -80dBu at 180KHz on the audio output is not really a big deal, it isn't over clever either.

 

[Cranehazard]

Eminate was what I typed, dang that spell check! I also think building linear supplies is cheaper smaller and easier than buying switchers. And you can recycle transformers from various gear and use an adjustable regulator to get your voltage. Dont let those transformers go to the landfill!

@Cranehazard - Are you sure "eliminate" is the word you want? Perhaps "emanate" instead?

 

[moamps]

I also think building linear supplies is cheaper smaller and easier than buying switchers. And you can recycle transformers from various gear and use an adjustable regulator to get your voltage. Dont let those transformers go to the landfill!

And you will have the opportunity to learn far more about power supplies than just using already made SMPS.

The image on the right is an example of how SMPS should not be implemented in audio (LA610-LA610MK2), especially in terms of reliability.

 

[abbey road d enfer]

Eminate was what I typed, dang that spell check!

https://whichiscorrect.com/emanate-vs-eminate/

 

[Cranehazard]

https://whichiscorrect.com/emanate-vs-eminate/

Emanate! Thank you.

 

[Matador]

Both sides in this thread are in fact correct:

Linear Supplies Pros:
- Easy to implement (especially for DIY'er)
- Generally stable if datasheet recommendations are followed

Cons:
- Incredibly inefficient (tons of wasted heat)
- Require tight coupling between transformer selection and output voltage
- Not generally tolerant of overvoltage/current stress (especially shorts)
- Require large bulk capacitance to handle ~100Hz/120Hz ripple

SMPS Pros:
- High efficiency (can be over 90%)
- Small form factor (minimal bulk capacitance generally needed)
- Can handle a wide variety of input / output voltages and current with a fixed power transformer
- Generally noise is above audio band so easier to filter with smaller components

Cons:
- Complicated designs (especially for DIY'er)
- Can radiate noise if precautions aren't taken
- Can be tricky to make stable across all operating points

 

[FIX]

if we haven't beat this topic to death, heres another design anecdote that might confuse the topic.

Early this century I was supporting a specialized DSP vocal processor. It already used a switching supply. I needed to come up with an additional +48V supply for its one mic input. I decided to KISS and think outside of the box at the same time.

I decided to make a cap doubler, that usually grabs AC from the transformer secondary. Instead I grabbed the HF square wave from the DC-DC switching supply. Switching at 100kHz instead of 60Hz allowed me to use tiny SMD caps for the doubling circuit. The parts were cheap and the PC real estate used was modest. I used simple zener pass regulator for the 48V regulated.

Worked like a charm.

JR

Currently working on a 48 supply the same way with a charge pump at 100K oscillator, smooth as silk. Come on kids, it works. But, it deserved one more beating...

 

[Cranehazard]

Didnt

Both sides in this thread are in fact correct:

Linear Supplies Pros:
- Easy to implement (especially for DIY'er)
- Generally stable if datasheet recommendations are followed

Cons:
- Incredibly inefficient (tons of wasted heat)
- Require tight coupling between transformer selection and output voltage
- Not generally tolerant of overvoltage/current stress (especially shorts)
- Require large bulk capacitance to handle ~100Hz/120Hz ripple

SMPS Pros:
- High efficiency (can be over 90%)
- Small form factor (minimal bulk capacitance generally needed)
- Can handle a wide variety of input / output voltages and current with a fixed power transformer
- Generally noise is above audio band so easier to filter with smaller components

Cons:
- Complicated designs (especially for DIY'er)
- Can radiate noise if precautions aren't taken
- Can be tricky to make stable across all operating points

For phantom I haven't had much trouble with heat and there is definitely not enough current for the transformer voltage to matter much in my experience. with switchers a batch of 100 nicely designed audio supplies 20 radiate high freqency garbage that gets in the circuit and causes high thd+n on an AP system even when every component was changed out. Stray capacitance? I dont know. Bad experience with switchers maybe. Linear supplies are extremely inefficient. Some distance inside the case can keep the thd+n in check with switching supplies.

 

[Newmarket]

CE has different emissions standards and some degree of self certification so may be fast and loose with small manufacturers. (IIRC CE included a stipulation that SKU inputs would resist interference, pretty much opposite the US requirement that puts responsibility on the source of interference.)

Yes. CE compliance for EMC requires passing both emission and immunity to pass the relevant product standard.

 

[Bo Deadly]

Bear in mind that 48V SMPS for phantom power requires special circuitry. In particular, because the load is highly variable (because the loads get switched on / off, especially in a desk with lots of mic inputs) you need to use a circuit on the output that will shunt current when there is no load to keep the SMPS sufficiently loaded so that it does not switch into an underload modulation [1]. To do this, you could use a circuit as simple as the following TL431 based shunt regulator but I would almost certainly figure out how to weave this into a capacitance multiplier filter as well for reasons previously mentioned:



[1] I have previously referred to this mode as "hiccup mode" but I believe this is incorrect. Hiccup mode is when the output is modulated on and off at a very low frequency for overload protection and not underload.

 

[dogears]

Some modern supplies have no minimum load requirements to avoid under load modulation. Which is nice.

 

[ccaudle]

I have previously referred to this mode as "hiccup mode"

Usually referred to as cycle skipping. It is done for efficiency improvement, there is no technical reason it must be done if you don't mind the efficiency loss at low output current. A lot of equipment needs to be rated for idle load power dissipation, either for market or regulatory reasons, so most off the shelf supplies will do that. If you design your own you could just set it for constant frequency operation.

For phantom power you could use a push-pull transformer driver (not just phantom, but it makes a lot of sense there since you will likely have other regulated rails in the system). The LT device line that Analog Devices acquired with Linear Technology have a line of switching regulators that let you set the slew rate on the mosfet drive so you can trade off efficiency for lower switching noise.
Something like this device:
LT3439 transformer driver

That one is open loop so needs a regulated input if you want a well defined output voltage. This model has feedback for regulation:
LT1683 regulated transformer driver

There is an old Linear Tech app note that decribes the principles:
Low noise switching supplies app note
And for a single inductor style regulator rather than transformer driver if you don't need isolation:
LT1534 low noise switching regulator app note

So it just depends on how much you have to rely on finding cheap off the shelf supplies versus being willing to do some design yourself.

 

[Bo Deadly]

So it just depends on how much you have to rely on finding cheap off the shelf supplies versus being willing to do some design yourself.

Yeah, I would almost definitely not try to make my own SMPS. I think I'm pretty good at building (probably better than at theory) but rolling my own SMPS would be challenging even for me. I think "stacking" off-the-shelf supplies is a perfectly valid thing to do. And if one does crap-out, you just swap in a new one for $20 USD. The only scenario were I could see making my own SMPS would be for tube supplies. But even then there are high voltage SMPS that you could run in CC mode (with a shunt regulator incidentally) and dial up whatever voltage you want. So pretty much never I guess.

 

[Trouble Report]

On my VR frames I spread a significant amount of copper across the frame to seriously drop the impedance and distribute about 170,000uf across each rail across the frame. With the Kingshill supply or the Atomic I get the same great noise sweeps. experience has shown poor distribution inside the units are the first cause of power related noise (unless your cables etc truly suck). On Robs MCI I did the 'over copper, and cap', but kept the power distribution on the modules anemic. Get great noise floor, but when you push signal you can see the 2nd order rise significantly around -5 on up till +15 where the 3rd order stuff starts to take over. The R then C then R then C of the distro sorta creates a filter and as Paul says it filters the hashy stuff. With the Atomic, my frame owners save a pile of cash and grief...

Except for the cash they pay you to convert their consoles. Worth every penny, in my estimation, by the way. Your version of the VR performs in a class by itself. But the collective amount of effort you have invested in creating this discipline is, I think, Herculean. Most proponents of the VR are looking for a one-stop-shopping sort of solution, and it's obvious from the replies to this thread that you can't just buy a 3rd party supply; wham it in, and expect it to make your world. No matter what the manufacturer claims.

 

[FIX]

You HAVE TO put your own additional filtering (chokes and caps) and you are able to get it to clean DC. Caps don't do it alone. Think if it as designing a notch filter or a LP filter. phantom is so little current that you can use pretty high value chokes and then regulate after.

 

[abbey road d enfer]

And you will have the opportunity to learn far more about power supplies than just using already made SMPS.

The image on the right is an example of how SMPS should not be implemented in audio (LA610-LA610MK2), especially in terms of reliability.

Would you care to elaborate? For me it's not obvious (actually I don't see the PSU - unless it's that rectangular black box... but I rather think it's the phenolic board on the right)

 

[abbey road d enfer]

you can't just buy a 3rd party supply; wham it in, and expect it to make your world. No matter what the manufacturer claims.

You HAVE TO put your own additional filtering (chokes and caps) and you are able to get it to clean DC.

Gospel!

 

[moamps]

Would you care to elaborate? For me it's not obvious (actually I don't see the PSU - unless it's that rectangular black box... but I rather think it's the phenolic board on the right)

LA610mk2 (right part of the picture I posted) uses four SMPSs, first for getting 24VDC from mains, which supplies two step-up (+B and phantom) high frequency SMPSs and one step-down high frequency SMPS for heaters (better picture attached). I serviced two of LA610mk2, PCB was seriously damaged around active devices on one SMPS so I was forced to place another separate SMPS to fix it. Changing whole original board was out of question.
I know for few older ones 610 (left side of the picture I already posted) with linear PS, which work without any problem.
P.S. Rectangular black box is opto attenuator.

 

[abbey road d enfer]

LA610mk2 (right part of the picture I posted) uses four SMPSs, first for getting 24VDC from mains, which supplies two step-up (+B and phantom) high frequency SMPSs and one step-down high frequency SMPS for heaters (better picture attached). I serviced two of LA610mk2, PCB was seriously damaged around active devices on one SMPS so I was forced to place another separate SMPS to fix it.

OK, got it. So they use an industrial primary SMPS (probably because it helps with UL/CE cert), and all sorts of secondary switchers. It's remarkable that the primary SMPS was not the cause of failure, it's the poorly designe/spec'd secondaries.

I know for few older ones 610 (left side of the picture I already posted) with linear PS, which work without any problem.

You mean the older ones had no switchers at all?

P.S. Rectangular black box is opto attenuator.

Thanks for the info.