Dying battery simulator

[yeeshkul]

I was thinking about making DBS. People seem to just add a pot between the battery and power pad in the circuit, although such a resistance can add itself to collector resistors and work as volume increase, which is what i don't want. Can this be sorted by a cap across the pot, so the signal will bypass this additional collector resistance?

[brett]

a LARGE cap will be necessary, as R is small (my almost dead 9V batteries seem to have 200 ohms internal reszistance).  Try 100uF.  That'll give voltage drop without increasing gain.

[yeeshkul]

here we go - an external DBS 

[Ice-9]

I'm not sure if i'm on the right track here,
with a fully charged battery it can supply x amount of current at the given voltage but as the battery is dying and the circuit trys to draw that x amount of current then the voltage falls.

If this is correct then the voltage divider part of a circuit can possibly be desinged to simulate the dying battery, eg if your VDivider is made up of 10k resitors to provide a good current for Vref , try using 1M resistors, this would limit the current that it can supply thus starving the circuit of current.

values are just examples and you would need to do some calculations to work out how much current your circuit requires and how much to starve it by.

just my 2p worth. Maybe someone with a bit more knowledge could could confirm / deny this idea.

[R.G.]

If you're going to simulate a dying battery, do it right.

A battery is a voltage source which varies from 9.5 or so when it's fresh to 7V or so when it's all tapped out, in series with a resistance, which varies from a fraction of an ohm or so when it's fresh up to a few K when it's all tapped out. Both conditions need to be diddled to get a realistic simulation. Using just a resistor simulates... just a resistor. The voltage varies with load in the pedal, and the resistor needed to drop the voltage may not be reasonable for the impedance of the battery when the voltage is right. And the resistor for the same voltage will be different from pedal to pedal depending on how much current the pedal draws.

There's been a dying battery simulator circuit at GEO forever. Well, OK, since 1998 at least. It's here: http://geofex.com/Article_Folders/pedalbdpwr/pedalbd.htm . Granted, it takes more than one resistor and a cap, but it actually does the job correctly.

As a humorous aside, that circuit is patented, and not by me. The patent application was six months AFTER the file was put up on geofex. 

[yeeshkul]

6 years later but anyway - is there a way of simulating a dying 9V battery without necessity to use 12V power source?

Musique.co has this article : http://www.muzique.com/lab/patent2.htm

There is a simple simulation circuit at the bottom of the page that uses diodes in series to the 9V power source (7809) and it comes with the following text:
"The diodes will have an internal voltage drop that increases with the current draw from the effect circuit, i.e. the sag effect."
What makes diodes in this position different from a resistor?

[yeeshkul]

I understand the voltage drop on diodes is not linear (depending on the current draw), i just wonder how that makes it closer to the dying battery simulation.

[R.G.]

I understand the voltage drop on diodes is not linear (depending on the current draw), i just wonder how that makes it closer to the dying battery simulation.

I don't think it makes it closer to the DBS, particularly. The change in forward voltage isn't terribly close to what a battery does, nor a great simulation of the drop in voltage.

A battery is an electrochemical cell. Actually, two half-cells. Each terminal reacts with the electrolyte to pump electrons out of the negative side, sucking them in through the positive side. As a battery discharges, the electrolyte and terminals are chemically changed into "leftovers", stuff which cannot further supply energy in the battery. A perfect battery would provide a constant voltage from completely fresh right up to completely exhausted, then the voltage would drop to zero. However, in real batteries, the "leftovers" interact with the still-working half-cells to (1) increase the resistance between the battery terminals and (2) lower the total voltage of the two half-cells. There is both a voltage drop and an increase in internal resistance.

A carbon-zinc, old-school 9V battery is about 9.2V when perfectly fresh, and this voltage tapers down as current is drawn out of the battery; the internal resistance increases too. A fresh CZ battery might push 1A into a short, but one that has discharged down to 8V would not be able to supply that much current because its internal resistance is bigger. The makers of batteries decided that a 9V CZ is discharged when its open circuit voltage reaches 7V. At that point the internal resistance is bigger, but they don't specify that very well.

Dropping a voltage or so off a 9V regulated supply gets the voltage down in the "dying battery range", OK, but doesn't do much about the internal resistance. As a gross generalization, a nominal silicon diode has a forward resistance in the range of 10-15 ohms. Batteries typically have an internal resistance that's different from that, but exactly what the value is depends on the battery chemistry and state of discharge. So both lowering the "open circuit voltage" of the DBS and adding a resistance is a much more accurate simulation.

Then there's the issue of whether the circuit cares if its battery is dying. Circuits vary in how much current they pull. The less current a circuit pulls, the more a "dropped voltage only" DBS is accurate for it. The more current it pulls, the more difference the internal resistance of the DBS makes.

[Mark Hammer]

Not many people use portable radios operated with a 9v battery anymore, so they tend not to have the more informative experiences that a continuous audio signal can provide.  I had my great learning about carbon-zinc batteries when I was laid up in bed with mononucleosis in 1966, and had to rely on the near-dead batteries in my radio to be able to hear my favourite tunes on the top 40  playlist.  I'd leave my radio off  for 90 minutes or so, to be certain I had enough juice regained in them to hear the tunes I knew would be played at relatively predictable times (and at the time, I was enamored of the guitar intro to The Hollies' "Look Through Any Window:)

Carbon-zinc will spontaneously "recharge" if you leave them for a bit.  That is, the chemical reaction will reverse with disuse, and they will regain a little bit of life.  And I emphasize the word "little".  If I kept the volume down, I could hear a whole tune, but if I cranked it up, i'd get a very audible compression, until finally the batteries crapped out.  This "Hold on, give me a sec to catch my breath" phenomenon would only occur when the batteries had dropped in voltage a bit. 

The "dying battery effect" is NOT, as some would have it, merely a lower, but stable, voltage.  You won't get it from an alkaline, or lithium-ion, or from a wallwart with a bunch of diodes in series, or anything similar, because they are providing a power source that can meet the current demands of the pedal.  What is fundamental to the impact a dying battery has on a fuzz (and nobody seeks dying battery sounds for delays, tremolos, EQ, wahs, etc.), is that a brief respite allows the battery to catch its breath but a peak transient saps the battery of current.  That is, the impact of a dying battery on a fuzz is a dynamic phenomenon, that corresponds to picking strength, and input signal amplitude.  That's what imparts a certain sponginess or compression, in addition to a bit of gateyness here and there.

I'm confident something can simulate that, but I'm equally confident that much of what gets suggested as a dying battery simulator probably won't, simply because they do not incorporate any means for the power source to "choke", cough, wheeze, or whatever you want to call it.  That is, they provide a steady and stable current delivery capability.

[yeeshkul]

Guys thank you very much for taking time to explain that.  It is all clear now.

[Brisance]

Mark Hammer: Now you got me into enthusiastic design mode for a new product!

[Mark Hammer]

Like I say, I'm confident somebody will be able to simulate it.  The first step is for people to have a clearer notion of what it is they are attempting to simulate.  My own very biased view is that we have been trying to simulate only one part (and the simplest part, too) of what takes place with a well-used 9V carbon-zinc battery when it is asked to power a high-gain circuit amplifying a guitar signal that has lots of peak transients.

Best of luck in your experimentation!  If you are successful, I'm certain you will make many happy. 

[slacker]

Not exactly what you're discussing but year's back The Tone God did a  power supply sag thing,  sagged in response to playing dynamics.  It's probably still around here somewhere in the FX-X competition archives.

[Brisance]

At the moment I am thinking of something similiar to a compressor but instead of being controlled by audio level it's controlled by current and which reduces "gain" of a setting of a constant current source. Attack and release(especially) will be much, much slower though.

[slacker]

That sounds pretty cool. I was thinking maybe just a current limited power supply might do interesting things, would only work with circuits where current draw varies in response to playing though.
Here's The Tone God's thing http://tech.thetonegod.com/punisher/punisher.html

[yeeshkul]

... would only work with circuits where current draw varies in response to playing though.
[/url]

Is one of these Si Fuzz Face? Hmmm, it probably is ...

[Mark Hammer]

That sounds pretty cool. I was thinking maybe just a current limited power supply might do interesting things, would only work with circuits where current draw varies in response to playing though.
Here's The Tone God's thing http://tech.thetonegod.com/punisher/punisher.html

The concept of the design looks about right, but something deep inside me says the time constants (attack/recovery time) may be too long.  It's an empirical question, though, I suppose.

[Brisance]

For a simple approximation though:

Although the cap(s) have to be fairly large, and the resistor needs to be tuned accordingly. Possibly add another about 4 times the pot resistance between ground and the pot as well, provided large enough, the pot regulates voltage and "regen" time.

[yeeshkul]

Holger do i get it right that in your scheme you are simulating the dying battery by the large cap and you are limiting the current that charges the cap?

[Brisance]

Exactly. also the pot regulates the max voltage, the cap will be charged to. Porbably needs lots of tweaking and large caps though.