Ground Isolation for Daisy-Chains

[Baran Ismen]

Can't see the image 

Thanks as always, Antonis!

I've seen such regulation modules on a cheap price around, some are buck converters, some are EMI/Noise filters.

I'll also give a try on a spare LM2596 based buck converter tonight, I suppose it'd work without any problem as 9V -> 9V, maybe its filtering would help, who knows? And it costs around 1$ here..

[Ben N]

I'll also give a try on a spare LM2596 based buck converter tonight, I suppose it'd work without any problem as 9V -> 9V, maybe its filtering would help, who knows? And it costs around 1$ here..

I don't think you'll get 9V out for a 9V input - those things are maybe 90% efficient, so there is always some loss.

[amptramp]

This may be a flyback converter with complete isolation from input to output.  If it is unregulated, it is easy for it to be isolated.  If it is regulated, there may be some connection from input to output for feedback of the voltage.  It doesn't look like there are enough parts for isolated feedback but there may be a second output winding that is isolated.  Flyback designs can achieve regulation of all outputs from one output, so a second winding could be used just for regulation.

[R.G.]

It doesn't look like there are enough parts for isolated feedback

I'd guess that the four-pin module to the left of the transformer is a four-pin optoisolator.

It's looking like all forms of ground isolation work, so the OP is very near to proving that the issue with noise is ground current whine from a bad-actor pedal.

[FiveseveN]

OK but you guys know that's a picture of a random converter that PRR posted and not the JOYO ZGP that Baran purchased, right? It's a different size and form factor.

[R.G.]

oops...

No, I thought it was the one.  Per Emily Littella (you're mostly all too young to remember )  "Oh. Never mind."  8-)

[Baran Ismen]

There we go..

There seems a lot going on, which i guess would be beyond my capabilities to reverse engineer it. There are 3 or 4 more caps or resistors below input power Jack, inductor and the big cap. But it seems it converts the voltage to ac then down to dc again and theres also an optocoupler for isolation i suppose..

[Matthew Sanford]

I had a thought deemable as loony, and I know we don't put resistors on ground lines due to current creation issues, but would 0.1 ohm resistors between effect ground to the jack tell other grounds to stay out? Or in a circuit to join analog and digital grounds? Or would it just cause problems?

[PRR]

would 0.1 ohm resistors between effect ground to the jack tell other grounds to stay out?

Same way a 0.1 inch tall wall keeps the monsters under my bed from getting out. Not very well.

[PRR]

I don't think you'll get 9V out for a 9V input - those things are maybe 90% efficient, so there is always some loss.

Efficiency is on Power. Voltage and Current can be exchanged (if sufficiently clever).

So it could be 9.0V 0.5A in and 9.0V 0.45A out. There's no free lunch, but if you slice the baloney right you can hit a specific sandwich thickness despite some waste.

And yes "my" image was a generic small isolated converter, not the actual gut shot. To show the likely number and relative size of parts. Thanks for the pictures of the real thing.

[Baran Ismen]

This circuit is quite similar to an isolated LED Driver circuit I see. Just thinking out loud..

Anyway, this unit is 15$ here. I wonder if I can make it DIY (or buy a ready module) under this price, because I may need to buy/make multiple qtys. Would it worth?

[FiveseveN]

an isolated LED Driver

The brains of the operation (DF6109A) is listed as a "High-Performance Cold Cathode Fluorescent Lamp Controller" for LCD backlights and such, so we're in the same ballpark What I'm not seeing is a rectifier on the secondary side, unless that's what the 4-pin device is and not an optoisolator.

[Baran Ismen]

an isolated LED Driver

The brains of the operation (DF6109A) is listed as a "High-Performance Cold Cathode Fluorescent Lamp Controller" for LCD backlights and such, so we're in the same ballpark What I'm not seeing is a rectifier on the secondary side, unless that's what the 4-pin device is and not an optoisolator.

It's an opto, the EL817, if you mean that. What dazzles me is that, its output current is so low, around 50mA, I wonder how it's being boosted there up to 700mA.

[FiveseveN]

It's for feedback, like on PRR's converter. For example:



In JOYO's case two MOSFETs would indicate a push-pull configuration to me but I hardly have any experience with this stuff .
One more thing: is there anything under the big electrolytic cap?

[Baran Ismen]

It's for feedback, like on PRR's converter. For example:



In JOYO's case two MOSFETs would indicate a push-pull configuration to me but I hardly have any experience with this stuff .
One more thing: is there anything under the big electrolytic cap?

I saw 2 or 3 resistors (or caps), but can't read them, need to lift the big cap for that.

[Baran Ismen]

an isolated LED Driver

The brains of the operation (DF6109A) is listed as a "High-Performance Cold Cathode Fluorescent Lamp Controller" for LCD backlights and such, so we're in the same ballpark What I'm not seeing is a rectifier on the secondary side, unless that's what the 4-pin device is and not an optoisolator.

Would this be the reason?

Features
1. IC wide operating voltage range : 4.5V – 18V

2. Built-in power management circuit

3. Very low IC standby current (<2µA)

4. DC-AC full bridge ZVS technology with high conversion efficiency

5. Excellent wide voltage working characteristics

[duck_arse]

an isolated LED Driver

The brains of the operation (DF6109A) is listed as a "High-Performance Cold Cathode Fluorescent Lamp Controller" for LCD backlights and such, so we're in the same ballpark What I'm not seeing is a rectifier on the secondary side, unless that's what the 4-pin device is and not an optoisolator.

SS34 looks like rectifiers to me.

[Baran Ismen]

I've got another Joyo ZGP for the reverb pedal as well, and my total noise is gone around %95 percent, which is more than enough for me. Also experienced my first live stage last Friday with the pedal set, and it was quite good.

Also yesterday, I found an old, linear-type Casio keyboard adapter (black & heavy one), which says it's rated for 9V / 1.2A, but I did a measurement and saw that its output is actually 13V solid. It's original, I've checked the guts, a big a.s transformer, 4 rectifier diodes, 10V/4700UF cap, and a 2.5a fuse, and that's all.
I wonder if it's possible to fix it to 9V..

[idy]

You may not need to fix it. If you measure the voltage on a an unregulated power supply when it is not powering anything you are measuring it "unloaded." It will measure high. Try hooking it up to a pedal or something. It will probably sag to 9v.Some of those wall warts actually vibrate and hum audibly in the air when they are not loaded.

[amptramp]

The voltage on an unloaded transformer / rectifier wall wart will be SQRT(2) times the loaded output so you could get high voltage when it is unloaded.  If I were to continue using it at low loads, I would replace the electrolytic cap with a 16 volt unit.  A 10 volt cap running at 13.5 volts is living on borrowed time.

Unitrode put out a number of applications manuals for their power supply IC's, so if you want to get into switching power supply design, you can look there.