Question about Apogee Symphony PSU

[iomegaman]

So my Apogee Symphony Chassis got doinked by a power failure here at this old building...

Natually Apogee  being the Jaguars of the audio world want an arm and a leg to repair it...

I pulled the psu and oddly enough its a small 2"x4" Astrodyne ASL-150

(here)...

According to the data sheet its pretty straight forward but as it is a MEDICAL psu, I cannot simply buy it and before I do I want to check to see that this is actually the problem, I have several 24v psu's besides a bench supply that can give me the two voltages (24v for mainboard, 12v for unit fan)...

At any rate there is an 8 header Molex connector that connects the psu to the mainboard...looking at the data sheet pins 1-4 are "DC returns" and pins 5-8 are "V1"...

Without a schematic for ANY of this I am shooting in the dark but would like to at least trouble shoot to see if the mainboard is only sleeping because it gots no power?>>

My thinking is DC returns = DC ground and V1= 24v+...

Its a crap shoot of course but I cannot see how it would be anything else, but lately my brain is foggy because my sleep habits are waaay out of normal...so I'm sending out a query to see if maybe I'm missing the most obvious before I jigger some alligator clips to the molex ribbon and a 24v PSU...

found one that matches MOST specs on Jameco...but because this is a medical grade psu Astrodyne does not sell to the public and under the current situation I would not want to grab a medical psu away form the market anyway...

Thoughts?

Feedback appreciated-Io

 

[iomegaman]

Never mind I sucked it up and tried it...

I'm kinda surprised its only a 150W psu for this...and only 2"x4"...I get that it's probably only "mostly" doing ADDA chips but man for such an expensive product its pretty "not grand" for a psu...

I don't know about the rest of you but this virus has me all kinds of over-reacting...I guess its better safe than sorry, but man second guessing yourself gets old especially when you feel old.

 

[Khron]

How did you determine that was a "medical" PSU? I don't seem to see any mention of that on the product page or the datasheet. But regardless...

"Medical" PSU with CapXon capacitors? If anything, that's a death sentence right there  (At least CapXons are pictured in the photo in the datasheet, so...)

And what do you mean, "only 150W", for a piddly little converter? I'd be surprised if that thing drew anything over 15-20W at full tilt, it's not (even) a power amplifier How much can a handful of opamps and converter chips draw?

PS: If you read the datasheet a bit more carefully, that 150W is specified for a 200LFM airflow; convection-only it's more like 100W (but i'm sure that depends what the switching MOSFET is bolted to, for heatsinking). Not that that's not humongous overkill either, though...

But yeah, wasn't the first tenant of making money, something like not spending it in the first place? Even though these PSUs are quite pricey in their own right, but considering a unit like this had (and might still have?) a 4-figure MSRP...

 

[iomegaman]

How did you determine that was a "medical" PSU? I don't seem to see any mention of that on the product page or the datasheet. But regardless...

"Medical" PSU with CapXon capacitors? If anything, that's a death sentence right there  (At least CapXons are pictured in the photo in the datasheet, so...)

And what do you mean, "only 150W", for a piddly little converter? I'd be surprised if that thing drew anything over 15-20W at full tilt, it's not (even) a power amplifier How much can a handful of opamps and converter chips draw?

PS: If you read the datasheet a bit more carefully, that 150W is specified for a 200LFM airflow; convection-only it's more like 100W (but i'm sure that depends what the switching MOSFET is bolted to, for heatsinking). Not that that's not humongous overkill either, though...

But yeah, wasn't the first tenant of making money, something like not spending it in the first place? Even though these PSUs are quite pricey in their own right, but considering a unit like this had (and might still have?) a 4-figure MSRP...

Well the different searches kept bringing up medical stuff from these guys and I know they are in the bay area and DO provide medical psu's for lots of stuff...nonetheless they won't let an average joe like me order one (yet)...

It sits inside the case and also provides 12v for the case fan...but its on the other end of the mainboard...

I guess 100 watts is ok, but you can easily rack up 16 analog I/O and 8 analog preamps plus digital outs plus two headphone amps (which are quite good) in one of these units...but definitely not for cheap...hell a 2x6 analog is running about $800.00 used...

Everyone's gone Thunderbolt these days even Apogee...

 

[Khron]

If the form factor and mounting holes fit, i'm sure just about any good 24V supply would be just fine - Mean Well make some nice ones, AND they use Japanese capacitors (which are a by-word for "lasts virtually forever" ;D ). Some of their units also have a second aux output, and they're quite decently priced to boot. But on the other hand, they're not made in 'Murricah, so pick your poison, i suppose.

And no, even 100W is faaaaaaar, far overkill for a line-level unit, seriously now. I mean, just look at the heatsinking on a linear (class-AB) 100W amplifier [later edit: which doesn't even need to burn off the 100W going into the speaker / load, "only" the losses] - i don't see anything like that happening in a converter / an audio interface

 

[scott2000]

OT..

The Rosetta has a little Meanwell smps and has cheap caps (maybe capxon iirc), aside from the large 400v which is usually quality,  and can go way off.....

I replaced them instead of springing the $14 + shipping for a new supply in the past bc I had some .... ....Just sharing

 

[Bo Deadly]

Go to Mouser and use their search for dual 24V 12V. Medical doesn't matter. That just means it's been approved by whatever official red-tape organization for use in a medical device. The Mean Well supplies are known to be really good. A quick search turned up RID-125-1244.

However, I do see one potential issue. I'm a little skeptical that that thing needs 150W. Is it a power amplifier? It is actually not good to use an SMPS that is too large because it could down-shift in "hiccup mode" and emit low frequency noise. I would measure the current that it needs, add 50% and then try to find a supply based on that. It could be that they picked the 150W supply because they simply could not source a stock one with less power (24V + 12V is a little unusual). In that case, you could just use RID-125-1244 and accept that it's as good as what was in there. Maybe there's a power resistor load in there? Or maybe they regulated after but that would still not be ideal. I personally would not feel good about using an SMPS in a pro-audio unit running in hiccup mode. You could add a load resistor.

 

[Khron]

I did a teardown and a repair on an Apogee Ensemble last year (the older, silver, Firewire version), and the mains PSU was a single-voltage deal, and the mainboard had a number of power supplies (flyback, buck, boost) to derive all the power rails required.

Sounds like "company policy", with some measure of cost-effectiveness / profit-maximizing. Easier to buy a cheaper, simpler mains PSU, and then derive the voltages on-board, than custom-ordering a 3-4-5 output mains supply.

 

[iomegaman]

Go to Mouser and use their search for dual 24V 12V. Medical doesn't matter. That just means it's been approved by whatever official red-tape organization for use in a medical device. The Mean Well supplies are known to be really good. A quick search turned up RID-125-1244.

However, I do see one potential issue. I'm a little skeptical that that thing needs 150W. Is it a power amplifier? It is actually not good to use an SMPS that is too large because it could down-shift in "hiccup mode" and emit low frequency noise. I would measure the current that it needs, add 50% and then try to find a supply based on that. It could be that they picked the 150W supply because they simply could not source a stock one with less power (24V + 12V is a little unusual). In that case, you could just use RID-125-1244 and accept that it's as good as what was in there. Maybe there's a power resistor load in there? Or maybe they regulated after but that would still not be ideal. I personally would not feel good about using an SMPS in a pro-audio unit running in hiccup mode. You could add a load resistor.

I know its not "science" to think in terms of actual physical size here but in searching my first perimeter was to find something in a similar package and this psu is 2"x4" open frame...found one on Jameco for $21 that also has the 12v for the fan and is 120watts...and yes it is a Mean Well that seems to have a better ratings all around...

Here

 

[Bo Deadly]

I know its not "science" to think in terms of actual physical size here but in searching my first perimeter was to find something in a similar package and this psu is 2"x4" open frame...found one on Jameco for $21 that also has the 12v for the fan and is 120watts...and yes it is a Mean Well that seems to have a better ratings all around...

Here

Yup. RPS-120 is perfect. It's pretty much an exact match. It's listed in Mouser as only 1 output. But clearly it has a second output specifically (and only) for a fan. Case closed.

 

[JohnRoberts]

The medical stuff is double insulated. A commercial version may work and be cheaper.

JR

 

[Bo Deadly]

The medical stuff is double insulated.

Just curious but what does "double insulated" mean?

 

[RuudNL]

Better insulated than 'single insulated'  ;D

 

[JohnRoberts]

Just curious but what does "double insulated" mean?

Pretty much just what it sounds like, two layers of insulation so a failure in one insulation layer is still protected.

Most consumer products that lack a safety ground plug, use double insulation (AKA IEC class II).

JR

 

[Bo Deadly]

OK. IEC class II. I Goolged it but I must admit the "double" part still confuses me. One layer of plastic seems about as good as two in this case. It sounds like the first layer makes it physically impossible for the hot source to touch any electronics inside (other than the specific part / rectifier / transformer windings) and such that no ground plug is necessary. Then maybe all of the mounting points and exposed metal of the casing has no conductivity to anything electrical inside? The point being that if by some miracle the hot source did touch something inside or if something poked inside and touched something electrical, there still would not be a path between the hot source and the outside. Both insulation mechanisms would have to fail before that could happen. So it all probably boils down to wrapping the AC input terminal strip, rectifier and transformer primary wires in a piece of plastic as opposed to running those nets to the main PCB?

 

[JohnRoberts]

OK. IEC class II. I Goolged it but I must admit the "double" part still confuses me. One layer of plastic seems about as good as two in this case. It sounds like the first layer makes it physically impossible for the hot source to touch any electronics inside (other than the specific part / rectifier / transformer windings) and such that no ground plug is necessary. Then maybe all of the mounting points and exposed metal of the casing has no conductivity to anything electrical inside? The point being that if by some miracle the hot source did touch something inside or if something poked inside and touched something electrical, there still would not be a path between the hot source and the outside. Both insulation mechanisms would have to fail before that could happen. So it all probably boils down to wrapping the AC input terminal strip, rectifier and transformer primary wires in a piece of plastic as opposed to running those nets to the main PCB?

It's statistical... two simultaneous failures are much less than half as likely, so more than 2x safer.

You can research the details but I think they consider reinforced insulation as two layers. For mains safety there are also spacing requirements from exposed mains nodes.

I recall one problematic SKU (last century), a class D amp with switching supply. The audio output was considered only effectively single insulated from the mains (intrinsic to the design... don't ask).  The expensive solution (but worthwhile to avoid weeks/months more delay from resubmitting for elevated agency temperature testing) was to use a special transformer winding wire coated with its own insulation rated for high temperature (luckily it all fit on the same transformer core).  This was only stopping my modest EU sales so I bit the bullet and spec'd the more expensive transformer just for that market

JR

 

[Khron]

I'm not sure the "double insulated" part, in this case, means the same thing. The IEC Class II "double insulated" needs no earth.

The "medical" definition may well only mean the insulation between primary and secondary is reinforced, but i'd be very surprised if that forewent earthing. The Astrodyne power supply linked in the first post is most definitely meant to use an earth connection (via two of the mounting holes, as well as the spade terminal near the AC input connector.

https://en.wikipedia.org/wiki/Appliance_classes

 

[iomegaman]

Just a quick follow up, new psu arrived perfect footprint, had to mod the molex connectors from the Symphony and wire in the fan but that was it...works fine...$21.00

And not dissing Apogee but they wanted minimum $150-$500 for the repair...which is not outrageous if you are guaranteeing a product, but I can't imagine it would take more than 15 minutes to slap in a direct replacement...one thing I will note...there was a spade connector on the original psu that called for ground...there is no matching ground wire in this box at all...I suppose they are grounding it via the screws that secure it in but its seems they just grabbed the first psu that fit and went their way...for such a flagship product it raises a few questions for me...generally love Apogee stuff but it feels like they and pretty much every other vendor who builds outboard boxes connected via PCIe or Thunderbolt kinda rushed to market to compete with Avid...

And speaking of Avid the fact that they want to charge me $299.00 for a piece of code (digilink) so I can run my Accel card as host for this Symphony is borderline highway robbery...they are as much a part of the problem as bad electrical design.

 

[Khron]

Well, profit margins gotta come from SOMEwhere...

Or, thinking backwards, if you can get similarly-featured units from other manufacturers... Where does the price difference go? :

(Not talking about "sound quality", though - at least half of that is confirmation bias and whatnot, anyway.)

 

[iomegaman]

Well, profit margins gotta come from SOMEwhere...

Or, thinking backwards, if you can get similarly-featured units from other manufacturers... Where does the price difference go? :

(Not talking about "sound quality", though - at least half of that is confirmation bias and whatnot, anyway.)

I had a friend who ran the service department for a Major Brand car dealership in Tucson 20 years ago...he told me that if customers knew how much money these car dealerships made from the service department (basically repairing new cars and older cars still under warranty) they would never buy another car from that maker ever again...

I once got a quote for my Jags transmission from the Jag dealership...they charged me $300.00 just to hook a machine up to my cars OBDII port and give me a diagnostic code (which was Jag specific and not available on a standard reader)...said the transmission was bad it would cost $2500 for replacement and another $2000 for labor...I drove the car home.

I went on Youtube found the exact transmission (actually was a Ford transmission)and rebuilt it myself for less than $100.00 and about 20 hours of my time...the car has run without a hitch for 5 years since then...

Still...I do love Jags...we like what we like...in spite of knowing they are ripping us off our bias and vanity runs past our logic units warning system.