my pedal sounds better with a dying battery???

[johnny5]

I built a modified GGG Red Fuzz and have always liked it. Recently I have been thinking it just sounds great for some reason. I figure I better check the battery and was reading 8.1 volts. I thought that interesting and compared the tone with the battery to the tone using a one-spot.

I like it so much more with the battery. With the one spot, it alot middy-er and less dynamic. With the battery, it behaves much more like an amp (less perfect) and just seems to have more personality.

SO, my question; Would you suppose this effect is a function of the lower voltage or is there a sag effect with a dying battery where the voltage drops a bit with higher demands (think tube rectifier)?

If it's just the voltage, I am sure I can scab in a voltage divider and get that 8.1 volt to the circuit.

Thanks for looking

[compuwade]

IMO it's most likely just due to the lower voltages. I wouldn't think it would draw enough current to cause a voltage drop on the battery at higher demands. From my experience most fuzz pedals sound better at lower voltages. I've had some that sounded great at 5-6 volts. You can add a curcuit like this: http://www.beavisaudio.com/Projects/DBS/ to simulate a dying battery to see how it sounds to you.

Later

[R.G.]

What is usually happening is that the bias wanders down as the battery voltage drops, and at some point you hit a "magic" combination for the specific set of components you have. Remember reading all that about selecting transistors for and biasing them properly for Fuzz Faces and the various JFET and MOSFET effects?
This is doing it the opposite way - putting in a set of parts, then tuning the power supply voltage to get a better sound to your ear.

The devil makes me say:

- but what if it sounded *even better* if one stage inside the pedal ran at 8.5V, another at 7.2V, an yet another at 9.5V? 
and that is possible.
Also
- What's up with all those guys wanting to run effects at 12V, 18V, and 24V? Are they crazy? Or have they found other goodness peaks?

[petemoore]

  I tried many voltages and biases, and many fine tones using up or down ramping voltages over the stages [two].
  Voltage made tweekable from low to medium [starting at around 1.5v up to like 7v the LM317 handled it] on each of a couple stages, trimpot for bias.
  Mu Amp...selfbiasing.
  Mostly Jfet's, some bipolar work, booster to distort boost stage type stuff...pretty cool results, I just never actually boxed any of it...trimpots or something started being iffy etc.
  Compressions, sags, ducking HF's on bassy attacks, total signal ducks, various soft distortions...pretty cool stuff.
  I have some voltage control on my Spyder and Spyderlike power supplies, when it matters it certainly does make a difference.
  Logic dictates: clean boost type response is best achieved with enough V+ to start with, when you want a bit of 'stretch tone' like the parts are havin' some troubles and complaining in a way that makes desirable tones...work it to be more or just play it.
  Diddling with V+ isn't really 'done' that much, and the parameters that move under/over/around it are 'personal', adds another element and all it's baggage of complexities comes with it..
 

[Paul Marossy]

If you like how it sounds at 8.1 volts, I wonder if adding in an adjustable voltage regulator like an LM317 and setting it to 8.1 volts operating off of your power supply would work (it would need to supply about 11 volts for that to work, however). Or does the limited current from the battery also come into play here?

[Mark Hammer]

Every time I think I'm out, they keep pulling me back in! 

What is too often overlooked is that when a battery "dies", several things are happening, and drop in voltage is only one of them.  One of the other things that is also happening is a change in the battery's ability to deliver current, particularly in response to sudden demands. 

In my experience, this is a very dynamic thing, and not the same as simply adjusting the current.  In other words, if, as a battery, you make me amplify a signal 500x, and you give me a big transient peak in response to picking hard, I may have enough momentary current capability to handle part of the peak, but mere milliseconds after that I pack it in and under-supply.  After all, there is only so much current capacity in a puny little 9v battery.  Regulators and such can mimic the voltage changes, but something more complex is needed to mimic the dynamic aspects of battery dying gasps and what it means for current supply.

The insistence by some that the "dying battery" effect is more pronounced in one battery technology (carbon-zinc) than another (alkaline), only reinforces my own bias that changes in current delivery capacity are critical to the phenomenon.  That doesn't mean it is pointless to explore things like supply voltage using nicely-regulated sources that have no current-supply issues, since those may very well deliver pleasing sounds.  But my gut tells me that precise emulation of dying battery needs more than that.

[johnny5]

Thanks for the responses. I do not think I want to build a whole new power supply but I may try the Beaves Audio trick for fun. In fact, I'll add a pot (not connected), find what R makes 8.1 V and use that fixed resistor.

I will be very pleased if it's that easy, as Mr. Hammer suggests, but it's definitely worth a go.

[MikeH]

In my experience, this is a very dynamic thing, and not the same as simply adjusting the current.  In other words, if, as a battery, you make me amplify a signal 500x, and you give me a big transient peak in response to picking hard, I may have enough momentary current capability to handle part of the peak, but mere milliseconds after that I pack it in and under-supply.  After all, there is only so much current capacity in a puny little 9v battery.  Regulators and such can mimic the voltage changes, but something more complex is needed to mimic the dynamic aspects of battery dying gasps and what it means for current supply.

So- are we talking about "sag" here?

I think I'm going to do an experiment here.  I have a big muff at home that I built a "dying battery" simulator built into it.  I'll get a couple different half-dead batteries and compare the half dead battery to the simulator set at the same voltage.  I suspect that I will find what you stated above.  When I had a BM on my board I always used to like the way it sounded with a weak battery.  In fact when another pedal stopped working properly I usually took the battery from that one and put it in the BM.  But I don't think the "dying battery" sim on my BM now quite captures what it was that I liked about the actual dying battery.

[Mark Hammer]

So- are we talking about "sag" here?

Yep.  Much like compression itself, "sag" is a dynamic phenomenon, not a phenomenon borne of simply changing the limits to voltage swing.  It is the sound of a circuit (and supply) "giving up", rather than the sound of a circuit boldly striding to a given limit and saying "Nope.  Not one step more.".

[Cliff Schecht]

The dying battery looks like an increasing resistance in series with the "+" lead. You can look at all of the devices the battery is powering as a load connected to the battery. Since the load is an active device (or devices) whose impedance is dynamically changing with the input signal, the output voltage of the battery changes with the signal dynamics. When you're playing softly, the active devices are asking less from the dying battery and the battery can output a higher voltage (meaning more headroom and less distortion). As you hit the strings harder, the active devices draw more current and the output voltage droops and you loose output volume while gaining distortion. This has the natural effect of compression, which tends to make everything sound great. And it does sound nice, doesn't it?

[Mark Hammer]

The dying battery looks like an increasing resistance in series with the "+" lead. You can look at all of the devices the battery is powering as a load connected to the battery. Since the load is an active device (or devices) whose impedance is dynamically changing with the input signal, the output voltage of the battery changes with the signal dynamics. When you're playing softly, the active devices are asking less from the dying battery and the battery can output a higher voltage (meaning more headroom and less distortion). As you hit the strings harder, the active devices draw more current and the output voltage droops and you loose output volume while gaining distortion. This has the natural effect of compression, which tends to make everything sound great. And it does sound nice, doesn't it?

That's a really nice lucid depiction of the phenomenon.  So, thanks.   NOW the challenge is to figure out how to mimic it.  From your explanation, merely lowering the voltage, or merely limiting the current in a fixed manner, is not going to do it.

So what will?

[Paul Marossy]

The dying battery looks like an increasing resistance in series with the "+" lead. You can look at all of the devices the battery is powering as a load connected to the battery. Since the load is an active device (or devices) whose impedance is dynamically changing with the input signal, the output voltage of the battery changes with the signal dynamics. When you're playing softly, the active devices are asking less from the dying battery and the battery can output a higher voltage (meaning more headroom and less distortion). As you hit the strings harder, the active devices draw more current and the output voltage droops and you loose output volume while gaining distortion. This has the natural effect of compression, which tends to make everything sound great. And it does sound nice, doesn't it?

Sounds kind of like a tube amp in some respects...

[Michael Weidenauer]

In my experience different brands of batteries (all alkali) behave very different when aging. i.e. the expensive ones normally used to power rabbits stay above 9V for a long time and then rapidly loose current, while others tend to loose voltage quite fast but deliver enough current for a long time (lower series resistance?).
So my question is how can I measure the series resistance of a battery?

[Mark Hammer]

Most "expert" (i.e., high-profile player opinion) views of dying battery benefits will involve a carbon-zinc battery.  Carbon-zincs have a different junction between cells than do alkalines.  They are generally flat slugs with broader surface area.

[Cliff Schecht]

In my experience different brands of batteries (all alkali) behave very different when aging. i.e. the expensive ones normally used to power rabbits stay above 9V for a long time and then rapidly loose current, while others tend to loose voltage quite fast but deliver enough current for a long time (lower series resistance?).
So my question is how can I measure the series resistance of a battery?

Measure the output voltage vs. current draw of the battery with your multimeter and divide V by I.

Paul, you are correct in that it's like a tube amp. I haven't compared the sag of a tube rectifier vs. that of a dying battery but I bet they both act quite alike. People seem to agree that a silicon based distortion pedal with a dying battery sounds like a tube power amp being pushed into distortion and asking more current from the power supply (which with a tube rectifier can't be delivered immediately).

I've been wanting to design a SMPS that actually looks at a guitar players dynamics and uses them to control the output voltage/current. Too bad I have more projects than free time...

[R.G.]

I've been wanting to design a SMPS that actually looks at a guitar players dynamics and uses them to control the output voltage/current. Too bad I have more projects than free time...

I believe that concept has been patented; at least, modulation of the power supply (of any kind) based on the envelope of an input. I'm pretty sure I read that once.

... probably Randall Smith, if history is any guide. 

[Brymus]

OK time for another one of my idiotic ideas...
But what if you had a sensor (like a mosfet biased to switch on at a certain V)that when the voltage swing increased on harder picking it would switch on a/(some) mosfets with resistors connected to ground to tax the supply harder,which once taxed hard enough would drop the supply voltage and the resistors would then be switched off and the voltage would be allowed to rise again.
It could be configured so that as the voltage swung further more resistance is added to tax the supply harder in that millisecond right after,ect
Perhaps this method of taxing the supply voltage would mimic a dying battery better than adding a large resistor in series with the V+ ?
Even though I'm not advanced enough to implement the circuit I think the concept is valid.

[Cliff Schecht]

That is a very legitimate concept. I used a similar trick in a design contest fairly recently and won second place. I would have placed first but I let my friend who was competing with me design the output voltage monitoring section and this part of the circuit ended up failing. The contest was to design a battery charger that ran off of a solar cell and I essentially ended up designing an SMPS inside of a linear regulator. A P-Channel MOSFET at the voltage input to the SMPS was used to slowly turn off the switcher as the battery voltage rose above a certain threshold. It even did a "trickle" mode that started and stopped the SMPS as the voltage of the battery floated near its fully charged voltage. The only time this trick isn't desirable is when power consumption is an issue because you are literally throwing away energy in the form of heat.

I've been wanting to design a SMPS that actually looks at a guitar players dynamics and uses them to control the output voltage/current. Too bad I have more projects than free time...

I believe that concept has been patented; at least, modulation of the power supply (of any kind) based on the envelope of an input. I'm pretty sure I read that once.

... probably Randall Smith, if history is any guide. 

No surprise there. I'm pretty sure I've seen dynamic sag control implemented even on these forums. As to how well it worked though, I have no clue.

[rnfr]

didn't the Tone God already mimic something close this with the Punisher?  i think he has versions for both voltage and current.  it's something i have always wanted to build but never got around to it.  i've also wanted to design a layout for it for a while as i haven't found one out there anywhere.

[FiveseveN]

About variable power supplies: look into Class H and Class G amplifiers. Of course we'd be using the principle "in reverse".