Dual Output Transformer : both parallel and series current.

[rojer64]

Hi all

I am a musician, currently building a rack for live gigs. Some items in the rack come with wall warts. These are the ones I want to get rid off.

A couple of preamps operate at 9 VAC, 1Amp max; I ditched the two wall warts and replaced them with an overspec'ed transformer :

240 VAC > 9 VAC 4A X 2(dual). Total 72 VA. This is a heavy piece of metal, can be bolted or screwed on the wooden chassis. So far so good. Seems I could add 6 pres and use the same power source.

Problem is I'd liketo add a new (dual) pre which operates at 18VAC, 1 Amp max. Makes sense, since it has two channels...

The replacement transfo, being dual out, could provide 18 VAC/4A with a different wiring. But is there a way, maybe through isolation, to get both 9VAC and 18 VAC from this single transformer, ending up supplying AC power to all three preamps ?

Or whould you advise that I open the dual pre and see if I can manage to provide 9V to each channel ?

I must say I have no knowledge whatsoever in such matters : only thing I know is that I hate wall warts in my racks. I could get a second transfo but the first one is like having it kind of easy right now, at 2Ax9V max when it can do 8...

Appreciate any input, even discouraging or ironic... Maybe I'll have a clue afterwards

Thanks for reading.
Roger.
Biarritz, France.

 

[Diver300]

It's very difficult to tell from your description whether it would be possible to supply the various items from one transformer.

If you need to isolate, you might as well use separate transformers.

You should start off by finding whether the wall warts are regulated, and which connection, if any, is connected to signal ground. The AC input items are unlikely to need a regulated supply, but you can check by measuring the output of the wall wart with no load. If it is well above the nominal voltage then you don't need a regulator.

As for isolation, that is where it will probably fall down. It probably isn't going to work to connect multiple items to one transformer, because the ground connections will be connected though the audio leads, and the items have been designed around wall warts that are isolated.

 

[rojer64]

Hi Diver, Thanks for the input.

I'm sorry that i didn't provide a complette description. I don't have a multi meter handy but I'll get one tomorrow and measure the outputs.

I had no idea about the audio leads gonnecting grounds... This sounds noisy... Does this mean that I'd be better off using the wall warts rather that a shared transformer, even on the two first preamps? Or would using a separate transformer output for each pre be satisfying ?
Seems like ditching the wall warts is not doable after all... Silly me

 

[Reloadron]

Just to be sure. Look closely at the wall warts. Do they output AC or DC?

Ron

 

[rojer64]

That's AC. Pre's are from ART (model=Tube MP). The dual pre I'm planning to add is from Alesis and runs on 18V AC. Oddly enough, I also have an Axon guitar synth that wants 12 VAC. And the RME MULTIFACE could'nt care less whether you give her 12V AC or DC. I'm suspecting a pattern with bands heavy on the guitar here

Now I'm worried about that isolation thing.... Do you suspect I'm biting more than I can chew and would be better off reverting to the warts ?

 

[alec_t]

I don't see any way of using two 9V windings to give both 9VAC and 18VAC without problems when ground leads are commoned.
Separate supplies could provide backup for a failed supply, whereas everything dies if a single power supply goes down.

 

[rojer64]

Hi Alec,

You're right about backups. But I'd rather have a single backup for a single and unlikely failure point. Besides, I can lose all those pre's altogether and stilll be at my gig on alternate channels with Di's... I have them in my bag

I had the idea of getting an overspec'ed common Transformer when I took apart that of the Axon. So tiny and getting hot in it's plastic wrapping... I figured a huge, open transfo would run cooler and probably be safer. I suppose this 18V + 9V idea was silly now...

i'll probably end up pulling the Art's from the rack and let them end their life as DI's. The alesis will probably love the 72VA transformer which is still more convenient than a wall wart.
Gosh do I hate those !

In any case, I'll check what voltages come out of the warts and see what I can come up with. Thanks for the inputs !

Roger.

 

[KeepItSimpleStupid]

OK, 9VAC seems fishy. There is a CHANCE that your 9 VAC equipment can operate off of 12 VDC just like my DSL modem can. Now, that will be tough for you to figure out, but it is a possible option.

These **broken link removed** are cheap for a DC supply. You don't need and don't necessarily want a BIG 9 VAC power supply. The devices likely need isolation. Lots of 50 or 60 Hz lying around also can introduce hum.

I'd get a bunch of these torroidal xformers and mount them in a 1RU or 2RU rack case. **broken link removed** and your done.

You can parallel secondaries to get a higher current and you can series them to get a higher voltage. The VA (Watts) rating you care about. VA is Volt-Amps which is slightly different than a Watt, but for all intents and purposes here, they are the same.

Parallel windings must be phased properly. If they are not, you will let out the magic smoke. To phase them, wire the windings in series until the voltages double. Now, you can mark the phase. The transformers should be marked with a dot. When connecting in parallel, the dots on the secondary windings should be together.

Put jacks on the back and a few on the front. The front ones, just have as spares. Fuse the transformers or at least use PTC fuses on the output. LED indicators may or may not be OK especially if you want to be able to change the brightness of them. PTC fuses don't blow. When they overheat, they limit the output current to a safe level. It's common to use these in fire alarm systems.

Power distribution could be done in a radically different way too: A huge easy to use Circular connectors, up to 37 pins, to a breakout box. The box, then splits the smaller wall wart plugs inside.

Right now, I am just throwing out ideas. Nothing more.

Right now, I'm having a similar issue with my network equipment. The US wall warts anyway don't like to stay plugged in when hanging on the ceiling joists and there are like 10 of them. 5V, 12V and 9 VAC.

 

[alec_t]

I suppose this 18V + 9V idea was silly now...

Not so. If you had another of those trannies you could use one for normal 9VAC and one for normal 18VAC, each with switchable outputs such that the 9V one could provide 18V backup and the 18V one could provide 9VAC backup. You would just have to ensure that you didn't try to get both voltages simultaneously from one tranny.
Pity your kit doesn't have DC input (the dual-voltage from one tranny could then have worked) .

 

[tcmtech]

You should be able to get both 9 and 18 VAC off the transformer a the same time. Just stack the windings in series for the 18 volt 4 amp output and use the center tap at 9 VAC for the 9 volt power. Just make sure you use the same end of the windings sets for your common referance point.

As far as most audio gear goes That I am aware of they usually all share a common earth or common ground on all of the preamp level audio I/O connections any way.

I say give it a test run with a few set of 2 amp fuses in your two supply leads and see if it works. If everything works and neither fuse pops and you don't have any excessive hum introduced you are good to go.

If not then it back to individual power packs for each device or at least individual power supplies for the 9 VAC and the 18 VAC power.

 

[rojer64]

Both the "mean well" and Toroidal look impressive ! I will definitely bookmark and look into them. I love the quieter/low field side of the toroidals. A gem resource for a rack designer... I have some room inside of the 1 U computer case that could be able to handle one or even a couple of those : one 12VAC and one 18VAC would be terrific ! They would benefit from the computer airflow and could even be shielded for interferences. Ideas open suddenly... Thanks !

As for 9VAC sounding fishy, I totally understand. And I did do my homework : the Axon synth actually came (second hand) with a 12V DC wart and would boot but not operate. I almost returned the unit until I found the very undocumented "AC current" thing (found a pic of the original wart!). The rest of the bunch is a couple of preamps with heavy but not roadworthy warts. These pres would sell for £149 back in 1995. Now, amazingly, they are still building them but how could they afford a decent PSU at £20 street price ?

Dual-pre newcomer wants 18 VAC as pictured here . At €68 I suspect the transfo to be a weak link,again. That's only me being suspicious, of course... Most people here probably know how paramount current production is, in analog audio.

Usually, I wouldn't try to second guess them audio gear manufacturers, in the matter of power supplies. Fact is, when I opened the Axon 12VAC wall wart (1.2A, mind you), I saw nothing but a couple of coils around an iron core... no specific isolation nor protection or regulation. This is when I thought that I could replace with a stronger, cooler, RACK FRIENDLY transfo. I just didn't look for fancier design; good, cause I know nothing about them !

My question was more like : should I dump the Art's (let them live outside the rack as DI's with wall warts) or could a dual tranny do for both 9V and 18V. I had my answer Got it, duly noted.

Now I have only this doubt remaining (hair on the neck): was I correct using a single dual out tranny (2X 9Vac X 4A), 4 wires outs as two separate 9V transformers, or should I have stayed with two separate ones, like the original wall warts ? Is it because of the ground meeting back in the audio chain ? After all, all wall warts in the rack do share the mains including ground, was my idea... Now I suspect I sound silly again. Oh well...

When the 18V Alesis comes in, I will inspect the supplied transformer (low hopes at €68) and probably replace it with the overspec'ed 72 VA already in the rack. That is until I investigate these toroidal, cool, quiet, low footprint, low emissions trannies... How seductive are these... or is it me overstating clean and cool energy in an audio chain ?

Done, the Art's will return in a bag, along with their wall warts... feels good saying again, not that' I'm happy with the conclusion ...

That's good work ! Only remains this doubt ... Seemed to work nice, cool and silent... How stupid was I? Can you educate me on the science here ? I confess I did try and do some reading: most of it is way over my head. That kind of reading seems to always require some previous reading to be meaningful. I guess that's why they invented schooling. Sometimes I wish I got there.

Oh well...

 

[rojer64]

Well, I just read this (after posting conclusions and thanks) and now I'm kinda puzzled :

You should be able to get both 9 and 18 VAC off the transformer a the same time. Just stack the windings in series for the 18 volt 4 amp output and use the center tap at 9 VAC for the 9 volt power. Just make sure you use the same end of the windings sets for your common referance point.

As far as most audio gear goes That I am aware of they usually all share a common earth or common ground on all of the preamp level audio I/O connections any way.

I say give it a test run with a few set of 2 amp fuses in your two supply leads and see if it works. If everything works and neither fuse pops and you don't have any excessive hum introduced you are good to go.

If not then it back to individual power packs for each device or at least individual power supplies for the 9 VAC and the 18 VAC power.

I never inserted any kind of fuse anywhere but I feel like starting to now...
Pff, I'll need to go get some breakfast and get back to you (live on gigging musician time here)
Seriously ?

 

[KeepItSimpleStupid]

OK.
1. Isolation occurs because of the iron core.
2. Full wave Rectification of a 9 VAC voltage yields about 12 VDC. It's basically VAC * 1.414 or the sqrt(2). It's very coom that the first thing an instrument does is rectify the input. It would not work, if the instrument created a +-6 V supply internally just through rectification.
3. The torroidal xformers mount with a rubber and metal washer.
4. I had a custom one designed for me once: 4x35 V @ 3A secondary.
5. Audio would tend to rectify, get 12V and then take the newly created "reference" to be 1/2 the supply voltage (aprox. 6 VDC). Then Connecting all of these together from a single power supply may or may not work. One might be at 6; another at 6.5 and another at 6.1 V
6. Each components DC supply would be slightly different due to component tolerances if run from a single AC supply.
7. Custom is possible. e.g. specified as ? VA, and say 6 x 9 VAC @ ? Amps ; so you would get 6 windings at 9 VAC. What you get is smaller/lighter and more expensive.
8. PTC (Positive Temperature Coefficient) Thermister/fuse protection would not be a bad idea.
9 one AC input fuse would be OK too.
10. One ZNR across the AC line would work for surge protection.
11. Putting the wall warts inside of a rack case is also an option
12. Usually you put a separate panel in the case to act as a chassis
13. In one design I used: I took the back off the rack case and replaced it with two rods. I was able to use a lathe to create the holes in the ends and tap them. The case then was mounted on a slide and magnets held the lid. The rods strain relieved the wires coming out. Tied with zip ties.
12. LMB Heager makes some very nice and the cheapest I found rack cases.

 

[rojer64]

>1. Isolation occurs because of the iron core.

That is useful... I take it my two ART 9v pre's would NOT be isolated from one another, which sounds bad. One is an electric guitar line (single coil pickups) which triggers all kinds of noisiness...

>2. Full wave Rectification of a 9 VAC voltage yields about 12 VDC. It's basically VAC * 1.414 or the sqrt(2). It's very coom that the first thing an instrument does is rectify the input. It would not work, if the instrument created a +-6 V supply internally just through rectification.

Em..... Not that english is not my language, I would probably even get less of it in french. Seems informative and useful though. i'll make sure and try to make sense of it and see how I can use it in that context.

>3. The torroidal xformers mount with a rubber and metal washer.

Good. That 1U pc case was already abused beyond insanity and will surely not mind a couple additional drillings...

>4. I had a custom one designed for me once: 4x35 V @ 3A secondary.

Ouch! you do not deal in two channels pre's now, do you ?

>5. Audio would tend to rectify, get 12V and then take the newly created "reference" to be 1/2 the supply voltage (aprox. 6 VDC). Then Connecting all of these together from a single power supply may or may not work. One might be at 6; another at 6.5 and another at 6.1 V

Back at >2... sorry about that. I will have to scrapbook this one and read on until I see the light...

>6. Each components DC supply would be slightly different due to component tolerances if run from a single AC supply.

Is that good or bad? I mean, those pre's are probably though little things and the wall wart is usually the one that goes first, heavy as it is. Hence I can relate to the tolerance thing. Now I don't like that one pre would not get the same current supply as the next. I like my sound steady and predictable. It's more than subtle a thing to speak about... Well I still like cheap gear, not that I have a choice. But I'd rather move and pay a lower rent than lose those subtleties. Am I threatening my sound with current inconsistencies?

>7. Custom is possible. e.g. specified as ? VA, and say 6 x 9 VAC @ ? Amps ; so you would get 6 windings at 9 VAC. What you get is smaller/lighter and more expensive.

I am definitely on the market for smaller, lighter and more expensive. I'll add cooler/quieter on my own authority .... Well, we all have our own notion about expensive now, don't we ? ...

>8. PTC (Positive Temperature Coefficient) Thermister/fuse protection would not be a bad idea.
>9 one AC input fuse would be OK too.
>10. One ZNR across the AC line would work for surge protection.

I need to get fluent in that PTC stuff and implement the idea. AC input fuse, dead stupid as it sounds, my rack is still lacking. Not proud about that. Good for me visiting here...
I use surge protection in the form of a multiple outlet. not sure how good these cheap things are...

Note that people building and using racks similar to mine (bigger, more professional, in the sound engineer scene) all have Furman like devices and even UPS. The scene I'm talking about is SAC . I'm way behind them on that area. But they are sound engineers with trucks (well SAC removed the "S" few years ago) and seriously paid gigs, while I'm more of your basic troubadour who just can't cope with Mackie stuff or worse any more. Not in 2012. I definitely won't ever use a UPS (don't care if power gets cut, I'll just break and have a well deserved beer. AC regulation is appealing though and I wish I had room for one of those; say 1U. In the end, it's more my ignorance preventing me from making the jump... "How good will those be to me any way?" kind of ignorance...

Oops, feel like I'm preying for more here

>11. Putting the wall warts inside of a rack case is also an option

Right. Only they are heavy, a pain to secure, get hot, often underspec'ed, and need an actual socket to be connect to (which I just don't allow in my racks). Plus I do want total control on their location, down, to the inch, due to the insane amount of emissions they're outing, along with their fellow out cable... Sorry how this comes out : even I can feel the french accent there...

>12. Usually you put a separate panel in the case to act as a chassis.

This sounds professional. I wish I could visualize an illustration. I'd still want to spread EM emissions where they hurt less, in a super-confined environement. I suppose the pro version of my rack would not have to be so tiny and constricted. But the fact is that I like to have one speaker in one hand, one bag on one shoulder, the rack on the other and the guitar in the back. Remaining hand doubles as mic stand handler and bartender greeter...
I'm not ready to join my gigs on a bike but that's close. I can definitely use a cab. Handy when they remove my driver license, which I am entitled to, driving mostly after bars just closed...
This is where I make the connection between rack design and driver license. Phew...

>13. In one design I used: I took the back off the rack case and replaced it with two rods. I was able to use a lathe to create the holes in the ends and tap them. The case then was mounted on a slide and magnets held the lid. The rods strain relieved the wires coming out. Tied with zip ties.

Sounds nice. My designs rely heavily on those tricks. Actually, the current one was factory-designed with those ideas... zips and all. I solved the wires problem by basically suppressing them. Only thing out of this rack is mains. Apart from midi connection, a small snake connection on top is all their is. CAble wight relief seems way to go, I'll keep that in mind for further designs. I have that idea of a bigger rack to offer high end mixing with my own SAC system. Those guys are actually experiencing the future, albeit in a a weird, rebelious, off-standards community, and I sure would like to be part of it. Totally disgressing here...

>14. LMB Heager makes some very nice and the cheapest I found rack cases.
Good to know. I'llgo check on them.


That's serious food for thoughts Thanks! You know how they say simple is beautiful ? I like the idea of stupid simple, a lot. How comes it takes so much planning and dedication to achieve that pure, simple, light and efficient design that looks stupidly so ? Note to self : "when you have kids, don't ever let them jump school... Sometimes I wish I wasn't me. Well that might be overstating it a bit

Good day all. I'll get back to the thread after some work. So far, this has been a fruitful one, amazing inputs! I'll probably have to edit some of these posts, get the syntax right and dump some of the foreign language stuff.Don't be offended if you notice, I'll like my posts readable, if not pertinent

Roger.

 

[KeepItSimpleStupid]

Isolation and ground loops are hard to get a grasp of. In simple terms, if each widget is isolated from power ground, then tieing the analog commons does not result in "loop". That "loop" results in hum.

Professional equipment generally uses XLR balanced connections, so the problem that consumer equipment has goes away. If your equipment uses XLR type connections, then it's likely that using power from the same supply for multiple units will have no effect.

If this was a consumer piece, I'd connect both pieces of equipment to the same power source with no inputs and look for a DC potential between the two units at the shield end of where the cables go. No potential would generally indicate that it would be OK to do so.

Another way would be to fuse the AC supplies individually. If they interact, the fuse will "hopefully" blow or you would get hum.

Now, if you do implement a 1RU rack solution, I would consider shielding the 9 VAC power cords, but I doubt you can. The coax power jacks don;t really like it. There are some screw terminal to 5.5/2.1 and 5.5/2.5mm plugs/jacks available and their are adapters to go between that series. To effectively shield, the shield should be connected at ONE END ONLY to ground. The END CHOSEN is normally the source.

**broken link removed**
These are available in 5.5/2/5 too - Be careful

https://www.powerstream.com/ac-0520w.htm for adapters from 5.5/2.1 to various connectors.

There are barrel connectors that can lock. Sometimes this can be good or bad.

Unfortunately, there isn't a good solution. There is one that's been developed which effectively would make a power supply communicating. It would tell the source what it's requirements are.

That transformer cost me $130 USD Qty=1 in the early 80's for my version of this: https://users.ece.gatech.edu/mleach/lowtim/ The Leach Amp Built from scratch. The PC board done the old fashioned way without computers using artist tape.
The un-rolled off frequency response is from like 0 Hz to 500 Khz. Unheard of at the time. Mine is rolled off, so it's from 0.5 Hz to 40 Khz, which is respectable. I have a 4bx signal processor and I've seen some amps choke on high rise time music and I mean noticeably choke.

 

[rojer64]

Thanks again That some more serious reading for me to do unitl I can get my mind around all these...

There are no unbalanced connections, except for the guitar jacks with hi- or line. That might explain the currently clean audio path despite a shared psu on 2 pres.

I will try and fuse the dual transformer for 2 X9v + 1 X 18v. [EDIT: Wrong]will report here and post a pic.

You've been very helpful and I took more than I was expecting. Thank you very much alll of you.

Good night,

Roger.

 

[Pommie]

You can't wire it to have 2 x 9VAC plus 1 x 18VAC as that will short the transformer via the rectifiers in the amps. You have to do 1 x 9VAC + 1 x 18VAC. The 2 9V preamps have to use the same winding.

Mike.

 

[rojer64]

Oops. Good one !
I was warned already but soon forgot

1 X9 + 1 X18 it is

 

[ChrisP58]

It is fairly common in small audio equipment to generate both positive and negative 12 volts DC from a single 9V AC winding. They tie one side of the the winding to gnd, and the other side goes to the the anode of one diode, and cathode of a second diode. Add a cap to each of those and you've got +/- 12 volts. In many cases, they aren't even regulated, trusting that the PSRR of the op-amps won't pass any of the power supply ripple.

 

[ccurtis]

You can't wire it to have 2 x 9VAC plus 1 x 18VAC as that will short the transformer via the rectifiers in the amps. You have to do 1 x 9VAC + 1 x 18VAC. The 2 9V preamps have to use the same winding.

Mike.

That's still iffy business, I think. The rectifiers in the amps may be grounded to the shields/enclosures/whatnot and if the grounds from the different amps get connected somehow, through the I/O cables, for example, the results might not be pretty. If going this route, I would definitely take a good, hard look at how the rectifiers are wired and, if necessary, make sure the ground/enclosure/cable shields of each amp are electrically insulated from all the other amps/cables connected to that xformer. Am I missing something here?