Pedal Power Supply Debugging

[MrFish]

Long time reader, first time poster.  My apologies ahead of time if this has been covered before, but I did quite a bit of searching without any luck.

I'm building my second pedal power supply, don't have a schematic, since it's cobbled together from bits of other people's and flavored with notes I've picked up from this board.

Here are some pictures, please excuse the disquieting use of hot glue... (more on that later)




It's a Triad Flat Pack FP 30-200 transformer.  (two secondaries at 15v/.400 amps) 

I've got a 1A fuse before the trans.  After that there's a 1.5A W005 bridge rec, then a 470uf electro and then the LM317t reg (with .1uf and .01uf caps).  I adjust the reg with a 560 ohm resistor and a 10k cermet pot.  (what I had lying around in abundance)

There's two regulators on the first secondary, one feeding out 9v and the other 18v.  (I know that I've got a transformer with 15v, but the actual value on my meter was about 21.5v after the bridge rec, so that seemed fine for 18v.)

The second secondary has one regulator doing 12v.

After each regulator (and don't ask me why I did this, I got confused I think, reading pages and pages about power supplies) there's another 100uf cap to ground.

The two regs coming off the first secondary share ground, but each reg has its own caps.  The reg on the second secondary has its own ground.

The 9v reg is powering 4 DC jacks daisy chained from the same ground/power.

The 18v is powering 1 DC jack and the 12v is powering 1 DC jack.  Lastly, there's an LED coming off the output of the 9v reg with a 560 ohm (I believe, might also be a 1k) series resistor.

Whew!  So, everything works, each part of the circuit reads as it should (or at least as I expected it to before soldering it together).  Right down to metering the individual output jacks from INSIDE the enclosure (measuring at the pins).

The problem is that from the OUTSIDE, only the first DC jack in the chained portion, the one directly connected to the regulator, spits out power.  The others, including the 18v (which reads fine from inside and is NOT daisy chained directly to the first four), give nothing.

The 12v works fine, reads a steady 12.29v.

So this is my second power supply and here's the three things I did differently from the last one.  (which was same concept except it was four 9v's on one secondary and three 12v on the other).

1. I added 100uf caps to the output of each regulator.
2. When daisy chaining those four jacks, I started ground and power on the same jack and carried it across the others (the last supply, I started ground on the first one and power on the last and had them cross each other.
3. Due to gross negligence and poor planning, I secured the jacks (5 of them are the square mini variety) and most of the other pieces (perfboard, transformer, iec connector, etc) with hot glue instead of screwing into standoffs.  (it's gross, a mess, and I'll never do it again.)

So I'm stumped.  I don't understand how I can read the correct voltages on every single jack from the inside, but that is somehow not coming through to the outside. 

Any ideas or help would be much appreciated.  If anyone can think of any troubleshooting I'm missing, that would be awesome.  I'm really dreading ripping this glue and the jacks out and re-soldering the mess.

Thanks!

[Mike Burgundy]

Just a quick reaction: *draw a schematic*. Just do. Not only does it help us understand what is going on, but it is a brilliant way to communicate with yourself. If I can't find anything wrong comparing the schematic I used to build from and the circuit,  I trace the circuit into a new schematic - ah, so that's where the wiring error was ;P
It looks to me like the daisy chain "in" and "out" from the DC jacks aren't on the same lug. If so, are these switching jacks? Theose two lugs are probably connected when nothing is inerted, but insert a plug and contact is broken. Might be the problem.

Edit: Had another look, and it looks like a contender. You're feeding the first DC jack to the lug on the left in the top picture. The rest is fed to the upper lug. The side connector (18V?) also has 3 lugs, two of which are used. Insert a plug into each type of jack and *measure* what the connections are, then use those. Ignore the 3rd (switching) lug. So for the ones connected to the side in a row, use the left lug I think, and parallel from that lug to the same lug on the next jack. Measure to be sure beforehand.

[MrFish]

Mike,

Two things. 

1.  You were exactly right, thanks a ton.  Those jacks (first time I've ever used the the small square ones) were wired totally wrong and all over the place.  I completely neglected to look them up like I do/should with any other component.

2.  Roger on the schems, I read you.  As I'm learning as I go, I definitely need to nip the whole "oh, I'm sure I'll remember what connects to what" in the bud.

Thanks again, after some quick resoldering, the supply is chugging along!

[Mike Burgundy]

Glad to hear it, and quick work too!
Be advised that the 15V secondary is designed to be 15V at a certain current. If you pull less current (less load on the xformer), secondary voltage will be higher, hence the high voltage at no load! So getting 18V stabilised from a 15V winding might very well be possible, as long as you keep current draw low enough. You can experiment how high you can go before the voltage drops too much and the 317 isn't happy anymore.

[R.G.]

You may be running into the internal switching of the DC jacks themselves. If you connect to the wrong lug, it's switched off when a plug is inserted.

[PRR]

> nip the whole "oh, I'm sure I'll remember what connects to what" in the bud.

I've run a few wires and switches over the years. If you skim my posts here, I know some circuits.

But when I wired a 4-way stairway light circuit in my garage, I drew it all out and nailed that to the wall.

(For the 3-way on the cellar stair, I found an old book and tucked a scan of the exact plan in each switchbox.)

Why?

• #1 - It IS easy to get confused mid-joint and wire it wrong.
  (Bad wall wiring is serious.)
• #2 - In the year 2014, 2015, 2019, I _know_ I will not remember *exactly* how I did it.
  (Moreso for whoever lives here after me, but that is not an issue in pedalbox work.)

(In 3-way and 4-way light switches, the key switch topology is fixed. But you can put the lamp at the source or at the end. And in my garage, the lamp ended up in the middle. Further complication: in houses, white is always circuit return BUT in switching you end up short on black wires with extra white, and the Code allows the "wrong" color *in cable* IF it is Re-Identified. So there are a lot of loose ends to keep in mind, or re-discover for some later repair/extension.)