Hello and basic question

[jh9067]

Hello!

I'm new here.  I'm an audio engineer who has a fair amount of experience repairing things, but am just beginning to build circuits from scratch.  My first few projects are going well, but I have a basic question -

Both of the pedals I've built have different and varying levels of noise with different power supplies.  Loud hiss with a One-Spot, ground noise with a Voodoo labs, and quiet and clean with a random 9v 1.7a switching adapter.

I'm aware that power supplies have varying voltages and noise, but my pedals are much more susceptible than any of the standard pedals I have laying around - Can someone point me in the right direction?

The circuits are variations of the Tillman booster curcuit.

Thanks so much!

J

[pinkjimiphoton]

you could try adding a simple power supply filter, say a 100u cap and an 1n400X (with the cathode to + of the cap) in parallel with the +/- of your power jack.
a smallish (maybe 100-1000p) cap from + to ground too can help tame some rfi.

most of the time people don't add this kinda stuff to schematics, and you may not need it with batteries, but it's a fairly standard practice to include some power supply filtering. dunno if i answered your question well, hope it helps.
remember, the center pin on a power jack is normally -, not positive!

don't ask me why i warn about that. 

good luck, and welcome to the forum!

[Kesh]

for low current pedals a small resistor in the power line prior to the cap to ground won't hurt, and will improve filtering

[jh9067]

Thanks!  I will try both of those suggestions.  I currently have no filter, that's probably the issue.  What value of resistor would one start with?

Thanks,
J

[Kesh]

depends on the current draw and how much voltage drop you think you can manage. i imagine you know the formula.

if you can take the LED power prior to the resistor/cap you will get more filtering for your money as the LED often consumes half the current.

[jh9067]

Thanks!

In console design, channel strips will often each have their own voltage regulator circuits even though the power supply is regulated.
Is this ever done in pedal design?  Like I said, I am new to design so I am unfamiliar with what the drawbacks might be (besides space constraints!).

J

[Jdansti]

Sometimes additional regulators are used, but mainly when you need more than one voltage on a board or if you need a lower voltage than a you would normally find in a wall wart or battery. As you probably know, regulators need at least a 1.5v higher supply voltage than the desired output voltage, so for a 9v regulator, you'd need at least a 10.5v external power supply.

The One Spot should be pretty quiet, but you might be picking up some stray RF. Make sure your soldering station is off when you test pedals as some will produce noise (from experience ). Also, make sure that the metallic enclosure is grounded to the board to allow it to act as a shield.

Here's the Tillman I've built and I haven't had any noise issues.

http://www.till.com/articles/GuitarPreamp/

[Kesh]

The PSU is usually out of the pedal makers hands, and as 9V is kind of universal, you wouldn't really want to require of the user the 11V+ needed to regulate down to 9V. It's more like, you do what you can, but the user can decide whether they use a noisy wall wart or a specialist pedal PSU. For me, doing what I can means a resistor of a few dozen ohms and a 1000uF 16V cap.

That said, I've built pedals with digital components that need regulated 5V, as well as the unregulated 9V and 4.5V for the op-amps. And some high end commercial pedals require 15V or what have you. And I'm just making a little tube thing that started with 30V but now has a 70V PSU. If you're making for yourself you can afford to be more exotic.

So no clear answer I'm afraid.

[jh9067]

Thanks!  The Tillman circuit is very close to what I'm building, I'm using his values plus a few other sets of values and tweaking for results.  There's a great article here http://www.hawestv.com/amp_projects/fet_preamp/fetpreamp2.htm that uses the same circuit with an MPF102 and different resistor values to adjust.  I've built that one, going to try the stock tillman next.  After researching the above suggestions, it seems like using a diode/resistor filter will cost me about two volts (please explain if I'm wrong!), so I'm wondering if a 7806 will achieve better filtering by actually regulating, even though it may cost me one more volt.  If I optimized the circuit for this voltage, (in the article I quoted it suggests the MPF102 would be best at 5 volts at the source I think?) am I building a more stable preamp with a regulator? 

Thanks!

j

[amptramp]

The Tillman uses a JFET and a drain resistor at the output which means any ripple on the rails goes to the output with almost no attenuation.  The output impedance of the JFET is much higher than the drain resistor.  This requires battery operation or operation from a regulated supply.  Extensive filtering may work in place of regulation.

There are other considerations.  If you have an amplifier and this pedal as the only items in the circuit with a connection to the AC line, try changing the switched connection for the line input capacitor on the amplifier (if it has one) to the other side.  You may have a situation where capacitive coupling from primary to secondary in the power supply causes hum and noise.  The good switching supply will have much less line-to-output capacitance because there are fewer windings and this may be why it works.  You haven't mentioned how it works with a standard wall wart supply with a power transformer, diodes and a filter cap and no switching circuitry.  I have a massive hum problem on a complicated build I am doing right now, so I will be interested in what you find.

[PRR]

What Ron said. The supply must be real-darn-clean because 90+% of ripple will shoot through the 1.5K to the output, almost none diverted through the FET.

It's made for battery. Battery don't hum/buzz.

Battery is your friend. Especially in bebugging, and in this case the battery life will be quite acceptable for gigging.

I'm not sure why the One-Spot would hisss here. Sure it has some small supersonic output ripple, but there's nothing in Tilman to beat a small ripple into the audio band.

I don't think 5V is optimum.

If you can't get batteries, try this.

[jh9067]

Thanks for all the tips and info!
Heading to the garage to build another preamp with this new info.  Same values as Hawes circuit above, with the 470 and 100uf as Paul suggested.  I also got diodes as listed above, should I be using one of those in parallel with the cap?  Also, are there any drawbacks to using a larger cap here?  If the 100 filters much better than the 10, (some circuits have a 47 listed) how does one make this choice?

Thanks so much!  I've learned quite a bit in the past few days, it's very exciting to move from just copying schematics to understanding how the components work together.

J

[jh9067]

Paul - it worked like a charm!

Should I tweak R2 and R3 for this lower voltage?
Is there any drawback to the larger cap?

Thanks!

J

[PRR]

> Is there any drawback to the larger cap?

It costs more.

When you build a million, you figure just how much filtering you need to avoid unhappy customers, and use just that much.

When you hack, you use an educated guess, round-up to be sure, round-down if it costs money or won't fit the box.

This is a small circuit and 470 ohms against 100uFd is probably over-kill (unless your supply is crap). It's cheap overkill, fine for bench or box. If you need to squeeze it into a plug (IIRC that's what Tillman did?) then you may need to experiment with smaller values.