Integrating DC Power into PCB design

[timkenyon]

Hello,

Newb builder designer here.  I've been working on adapting some existing schematics into more personal projects, building some either ridonculously expensive vintage pedals for my own self.

I've been noodling around with EasyEDA as a schematic/PCB designer, but I'm at the stage I need some feedback on the project, and I'm hopeful that some of y'all might look this over and let me know if I'm on the right lines or not.

I'm looking to incorporate as much of the offboard wiring into the same PCB as the effects circuit as possible (the LED and DC power especially), but I'm coming up on numerous ways of working it for a range of different pedals.  Here's my process so far using Tonepad's Red Ranger schematic (which itself came from HomeWrecker and JD Sleep)

Anyway, thanks, in advance.

[iainpunk]

first of all, welcome to the forum, enjoy your stay!
well, there are quite a bunch of things

the first bypass switch is wrong way round
that D2 diode won't light up since 1) 47k is quite a bit ti high resistance, and 2) there won't be any current anyhow because it goe from 9v to 9v instead of 9v to ground
C7 100uf will leak ALL signal to ground or atleast make everything sound extremely muffled.
the differential gain stage has been wired wrong, the output to the PNP has to be on the other side of the adjacent resistor.
the PNP is probably the wrong way round (unless you deliberately chose to put it in reverse beta)
there is a short circuit connection from the output to ground, silencing the output.

this is some kind of booster, right? very interesting choice to go with a differential gain stage, i like that a lot tho!
cheers, Iain

[timkenyon]

Hi Iain,

Thanks for the response.  Just a few questions:

1)  Do you mean in the first bypass switch I need to flip the switch horizontally so that pin 2 is connect to the jack?
2)  C7 is drawn incorrectly - that's on me.  C7 should be connected to ground, not to R12 and the rest of the circuit.
3)  I'm not sure what you mean by "differential gain stage" - though I have drawn the PNP transistor upside down - that's on me again.
4)  Could you tell me where the short to ground is on the output?

This is the Dan Armstrong Red Ranger boost.

Tim

[idy]

Welcome!

1) First bypass switch: the jack* needs to go to pin 2, 3 to circuit, 1 is correct.

4) C6 goes to ground. It could be replaced by a pulldown resistor (likely 1M) but now it is a short (at audio frequencies) from signal to ground.

The differential mode thing is the two transistors with emitters tied together, you can google this. Fancy.

*I also think the input jack is odd... you have the power ground attached to the signal path, and you have a normally closed switch which attaches the power ground to the tip when things are not plugged in ( and there is no other path for power ground to reach circuit ground). When something is plugged in, the tip goes nowhere and power is disconnected.... (the part that looks like a square root sign is the tip, the arrow it "rests" on is the normally closed connection.

Maybe you mean normal battery connection: stereo jack, tip is signal, sleeve is ground, ring is power ground (battery ground). Or without battery you could have tip as signal, NC to ground, sleeve to ground. Now when unplugged the input is grounded and silent.

[Gargaman]

Maybe you only FILT C6 and leave C7 out.
https://www.diystompboxes.com/smfforum/index.php?topic=127081.0
Good recent discussion about that part of the circuit, in case you haven't been there already.

[timkenyon]

Hey,

Thanks for the reply.

My next question was whether C7 was a filter capacitor that I've drawn next to the DC supply as it's present on most schematics I'm reading these days.

Thanks for the link to the power discussion.  It's certainly the route I'm trying to take - to build a "1 supply to rule them all" approach.

Below is an update with the changes Iain suggested (post 2 - as far as I could, anyway) along with removing C7.

[Rob Strand]

My next question was whether C7 was a filter capacitor that I've drawn next to the DC supply as it's present on most schematics I'm reading these days.

Thanks for the link to the power discussion.  It's certainly the route I'm trying to take - to build a "1 supply to rule them all" approach.

Below is an update with the changes Iain suggested (post 2 - as far as I could, anyway) along with removing C7.

I think the problem is your original schematic wasn't drawn correctly around C7.    (You also have two C6's and the C6 at the output has a short to round where it needs a resistor to ground.)

The idea behind the original C7, when drawn correctly, is it filters noise on the supply and prevents it from going straight into the input of the amplifier.   

Check out this simpler schematic.   See the 2x20k at the bottom with the 10uF cap and the 1M resistor.
http://www.experimentalistsanonymous.com/diy/Schematics/Distortion%20Boost%20and%20Overdrive/DOD%20Overdrive%20250.gif

You keep your R1 + R2 divider and C7 (which are like the 2x 20k and 10uF) but you need to add a resistor between the divider and the base of transistor Q2A.    It has to connect to the base.  The normal value for that resistor is 470k or 1M.      When R1 and R2 are 220k you don't need a large value like 100uF, something like 4.7uF would do.

[timkenyon]

Alright folx,

Thank you for the ongoing feedback, I've updated the schematic as per the feedback from idy: the input jack has been altered, the bypass switch flipped.  I've put the circuit input on pin1 of the switch and the output to pin 7 which should correspond to the correct terminals on a 3PDT switch.  I have added the draw down resistor after C6 which I had missed from the original schematic.

Rob:  If I already have a 100uF electrolytic capacitor (Labelled FiltC6) would I need C7 wired to ground as well?  Surely the FiltC6 would also filter noise to ground, or am I mistaken.  I'm also reluctant to add a resistor in front of the transistor base as this isn't part of the Red Ranger circuit.  Could you tell me why I would need to do this?

Tim

[JerS]

The negative terminal is connected to the input jack tip. This will not work.

I think you are trying to get the power to switch on when you insert a plug into the pedal. To do this you need a stereo jack, and wire the negative power terminal to the "ring" lug on the stereo jack. This effectively completes the ground circuit when a plug is inserted. That is not what you have shown here.

[ElectricDruid]

If we're having power filtering, can we please have a resistor in place so we get a proper lowpass filter, and not something that depends on how long and how rubbish your power cable is?!

I refer my learned friend to the afore-mentioned "ideal power section" thread.

[timkenyon]

@JerS - There's no battery clip in this circuit, so I'm not trying to tie the power to a TRS stereo plug - it's a mono jack only.  I think I may need to add another switch point and run the power in through that rather than what I've drawn here.

@ElectricDruid - what size resistor and where should it go?

[ElectricDruid]

@ElectricDruid - what size resistor and where should it go?

It should go in series between the +ve power input and the filter caps C6 and C5. It makes an RC lowpass filter which removes noise from your power. There's a trade-off between larger resistor values (which leads to a bigger voltage drop) and smaller resistor values (which need larger capacitors to get the same filtering effect). Typical values are probably 10R to 100R. As the current draw of the pedal goes up, smaller resistors and bigger caps makes more sense. For pedals that don't draw much juice, you can use a larger R and save space using a smaller capacitor.

[timkenyon]

I'm assuming it's 10ohm to 100ohm depending on the capacitor size.  Was it series or parallel?

[antonis]

In EXHIBIT A, teke 100μF cap AFTER 10R resistor..
(and before 1k, of course..)

EXHIBIT B configuration makes no sense..  

[timkenyon]

Thank you, Antonis.  I'm still new to all this so I appreciate your patience.

[timkenyon]

OK, y'all,

So here's where it's at.

[antonis]

Change 100μF electro cap polarity IMMEDIATELY..!!! 
(flip it 180^ο..)

Same for 1N4002 Diode polarity..!!!



P.S.
Plz ignore "IMMEDIATELY..!!!  "


edit: On a second thought, if you let diode as it is, reverse placed electro cap should have no problem to withstand about 650mV reverse polarity voltage..
(of course, any 10R resistor of power rating lower than 7W will give the relieving solution..)

[timkenyon]

Done and done.  I flipped that part of the circuit when it got moved and it didn't flip the symbols.

All fixed.  Y'all have been great, thank you.

[ElectricDruid]

I'm assuming it's 10ohm to 100ohm depending on the capacitor size.

Yes, exactly. Bigger cap means you can get away with smaller resistor and vice versa. It's the (R x C) product that matters.

The new version is looking much better. Nice work.

[Rob Strand]

Rob:  If I already have a 100uF electrolytic capacitor (Labelled FiltC6) would I need C7 wired to ground as well?  Surely the FiltC6 would also filter noise to ground, or am I mistaken.  I'm also reluctant to add a resistor in front of the transistor base as this isn't part of the Red Ranger circuit.  Could you tell me why I would need to do this?

OK got it.   It's OK to follow the original circuit (I was assuming it was a new design).  It will work.

As far as being problem free from hum the method I mentioned provides better rejection of hum (and noise).   With a given amount of ripple on the power supply there are many mechanisms which allow that hum to get into the audio signal.   One is from the parts that connect to the supply rail and the other is biasing.   The thing about the basic divider used on the Red Ranger is the hum gets injected from the power supply, through divider and then to the input.  If a circuit has a reasonable amount of gain that hum gets amplifier up.  The input is the worst place you want hum getting in.   Anyway that scheme I mentioned prevent hum getting in through the input.    Given it is an existing design the 1M resistor I mentioned before should be 110k (so use 100k).  The reason for that is the input impedance looks like the two 220k's in parallel so any mods should retain that impedance.    For the mod case the 220k's should be reduced to 22k and the cap made bigger like the Dod example.

Anyway that's the motivation for the changes.


FYI, the base of transistor Q2 connects to the wrong side of the R3 (3k).

Another snippet is the original Red Ranger was never powered from an external PSU so it didn't have to worry about supply hum.