ideal power section?

[Rob Strand]

Assuming 1N400x (1A) and 9V supply, anything over 9 Ohms will protect the diode +and+ the rest of the board. If you persist, any pedal-size resistor at nearly 9 Watts will burn up real quick. 270 Ohms 1/4 Watt will sit there too hot to touch for months.

That's where a PTC helps but you need to make sure it will trip.   If you want filtering and not have the resistor fry you probably (then) need to move the resistor after the diode.   The methods using MOSFET switches are a little more graceful under attack.

[vigilante397]

I got sick of people sending pedals back after plugging in the wrong polarity, so I started doing a schottky bridge rectifier on the input of every pedal, so polarity just doesn't matter. Costs me an extra couple cents per build, but nobody fries things anymore.

[iainpunk]

I got sick of people sending pedals back after plugging in the wrong polarity, so I started doing a schottky bridge rectifier on the input of every pedal, so polarity just doesn't matter. Costs me an extra couple cents per build, but nobody fries things anymore.

also lowers chance of ground loops, the problem that could occur is that the ground currents find it easier to go back via another pedal instead of the diode, causing power current through signal cables. no problem with analog low power pedals, but tube pedals and digital pedals might not behave quite nice.

cheers

[vigilante397]

I got sick of people sending pedals back after plugging in the wrong polarity, so I started doing a schottky bridge rectifier on the input of every pedal, so polarity just doesn't matter. Costs me an extra couple cents per build, but nobody fries things anymore.

also lowers chance of ground loops, the problem that could occur is that the ground currents find it easier to go back via another pedal instead of the diode, causing power current through signal cables. no problem with analog low power pedals, but tube pedals and digital pedals might not behave quite nice.

cheers

In theory yup, it's a possibility, but did a fair bit of testing and haven't had any issues, nor have I had any reported. Another thing to keep in mind if you're selling pedals, test the hell out of EVERYTHING before sending to customers.

[Eb7+9]

I have NEVER seen a pedal circuit that requires an actual current limiting resistor.

Bob Moog's "oberheim/maestro" sample-hold filter ...

[vigilante397]

Bob Moog's "oberheim/maestro" sample-hold filter ...

I'm not seeing anything in the FSH-1 that would try to fry itself without a current limiting resistor. I see the original has them on both + and - rails, but I don't see them as being there to prevent the thing from burning up.

[Rob Strand]

'm not seeing anything in the FSH-1 that would try to fry itself without a current limiting resistor. I see the original has them on both + and - rails, but I don't see them as being there to prevent the thing from burning up.

The control stuff is on 9V and the audio is on 8V after the RC filter.    To me the intent here is filtering

[Eb7+9]

I'm not seeing anything in the FSH-1 that would try to prevent ...

Bob used the same resistor value twice to protect his 3080's - then again, in his LabSeries soft clipper design ...

you can fry OTA's by forcing more than 2mA into their bias port ... he obviously intended to limit bias current to around half a mA in the s/h and just under one mA in the clipper ...

[vigilante397]

Bob used the same resistor value twice to protect his 3080's - then again, in his LabSeries soft clipper design ...

you can fry OTA's by forcing more than 2mA into their bias port ... he obviously intended to limit bias current to around half a mA in the s/h and just under one mA in the clipper ...

Power consumption is determined by the load, not by the source. A 9V wall wart is not a fixed current supply, you aren't "forcing" current into your pedal. You're making current available, which the circuit is drawing.

[ElectricDruid]

Bob used the same resistor value twice to protect his 3080's - then again, in his LabSeries soft clipper design ...

you can fry OTA's by forcing more than 2mA into their bias port ... he obviously intended to limit bias current to around half a mA in the s/h and just under one mA in the clipper ...

Power consumption is determined by the load, not by the source. A 9V wall wart is not a fixed current supply, you aren't "forcing" current into your pedal. You're making current available, which the circuit is drawing.

Yes. And the Iabc/"bias port"/transconductance input/whatever it's called/thingummywhasit on an OTA isn't a supply pin either, so it's not really relevant here, except tangentially.

Tangentially speaking, the Iabc input on an OTA is either one or two diode-drops above the negative supply voltage (depending which OTA we're talking about exactly) and can only stand 2mA of current. That means that if you whack it with 9V, you need a decent resistor in series with that input or the chip will fry. Several KOhms. Nothing like the few hundred ohms you'd usually see as a maximum for power supply filtering.

[marcelomd]

Talking about power sections. I've seen a lot charge pumps doubling the 9v.

Apart from the Klon, is anyone using symmetrical power supplies?

And boost converters? They take +- the same real state as a charge pump and switch a lot faster, so less change of audio range noise.

[garcho]

is anyone using symmetrical power supplies

Seriously? That's audio industry standard albeit guitar pedals break most of those.

[vigilante397]

Talking about power sections. I've seen a lot charge pumps doubling the 9v.

Apart from the Klon, is anyone using symmetrical power supplies?

And boost converters? They take +- the same real state as a charge pump and switch a lot faster, so less change of audio range noise.

I've done a couple pedals that were ±15V rails, there are a handful of overdrives out there that do it for the headroom. I use SMSP boosts in just about every pedal I build these days (because toobz).

[marcelomd]

is anyone using symmetrical power supplies

Seriously? That's audio industry standard albeit guitar pedals break most of those.

I was thinking specifically about pedals. My chain of thought went like this:

1- How nice it is to have extra headroom for equalization;
2- I can get double voltages with a charge pump. You get 18V from +9V;
3- The power amp module I'm using in my bass amp has +- rails for the preamp;
4- Charge pumps can also invert voltages. You get +-9V from +9V;
5- Most audio stuff uses symmetrical power. If you are going to design something new, why go +18V instead of +-9V?
Bonus- Charge pumps are +- expensive, noisy, and the capacitors, bulky. Why not use simple buck/boost switchers?

The only pedal I know that uses symmetrical power (not from a transformer) is the Klon, with a quirky +18/-9 or something.

For new tube stuff, SMPS is the sensible choice IMHO.

[ElectricDruid]

The only pedal I know that uses symmetrical power (not from a transformer) is the Klon, with a quirky +18/-9 or something.

The +18/-9V might make sense since many op-amps were designed for +/-15V (so 30V across them) and often have 36V as the absolute max supply. So +/-18V is out, but +18/-9 is easily derived from a 9V input but uses the op-amp at almost its ideal voltage. Just a thought.

In my ideal world, pedal power supplies all provide +/-15V power to stompboxes. Sigh.