how do you predict battery life?

[gaussmarkov]

i'm going to build R.G.'s hum free signal splitter and am guessing that the battery consumption is pretty low so that i would not need to build in or use a voltage jack.  but then it occurred to me that i have no idea how to predict battery life.  so i checked around and i see that it depends on the type of battery and (obviously) on the current drawn by a circuit.

how would you measure current consumption of a circuit for predicting battery lifetime?

does anyone have formulas for various batteries?  i found one for carbon zinc here (http://www.epanorama.net/links/psu_battery.html):

Estimated total battery life = (Ib / Id) x 0.7
Where: Ib = Total Capacity Rating of battery (mAh) and Id = Current Consumption of the device in milliamps (mA)

but they don't seem to give a ballpark formula for alkalines.

any help is appreciated!   

[R.G.]

A battery is a source of milliampere-hours. Old carbon-zinc batteries are around 160ma-hr, and modern alkalines are in the 400-600ma-hr range. That current is measured until the battery gets down to 7V, as I remember.

To measure your drain, you put one terminal of the 9V battery in the clip, but twisted so you have access to the other battery terminal and the other battery clip terminal. You put your DMM between the two loose terminals, set to measure current. Be sure to measure both bypassed and not, because the indicator LED in an effect is often as much current as the whole rest of the circuit.

Once you know the current, you divide that into a guess at your battery's ma-hour rating. If you found a drain of 10ma, then a 400ma-hr battery would last for 40 hours. A 160ma-hr carbon zinc would go about 16 hours.

[B Tremblay]

Put this one in the FAQ!

[Dave_B]

Put this one in the FAQ!

I was just adding it to the wiki.   

[gaussmarkov]

perfect.  thanks, R.G. 

is this a frequently asked question?    i don't know how i missed it with the search function.  perhaps because i was looking for battery life info and not current consumption. 

[Ortiz]

No he's saying it's a good question to be added to the FAQ.

[gaussmarkov]

No he's saying it's a good question to be added to the FAQ.

super!  thx.

[Dave_B]

No he's saying it's a good question to be added to the FAQ.

Ah ok, thanks for that.  When doing searches I never notice any hits in the FAQ section, so I forgot it was even there.  I started using the Wiki after Aron pointed it out.  It's got some nice features (going straight to the relevant post in a thread, for example) and seems like a good way to organize all the FAQs.

I'm unintentionally threadjacking again... 

[Noplasticrobots]

I just tried this and it took my meter about a minute to get to a stable number. I followed your directions RG, so is this standard when measuring current?

[stm]

It seems that you are not measuring current right.  Most DVMs require you to change the socket for the red or (+) test lead to a position named "mA" or "10A" or something like that.  The former is for low current measurements while the latter is to be used only on the 10A current scale.

[stm]

Regarding battery life, it is a fact that alkaline batteries die much sooner when discharged at high rates than at slow rates.  That being said, it is typical to specify the battery capacity in mAh at something like 1/10th the suppossed 1hr current rating, meaning that if a 9V alkaline battery has a 550mAh rated capacity, it'll have so when discharged at approximately 55mA only.  If you discharge that same battery at 550mA you won't get 1hr, but significantly less.  Luckily typical pedal currents are way below 55mA, so we are on the safe side here.

The other issue to take into account is "how do you define when a battery has been discharged?".  Alkaline batteries have 1.5V per cell, nominally.  A 9V battery is composed of six cells in series, totalling 9V nominally.  When brand new, each cell will put out more like 1.6V, and thus this explains why you can read voltages as high as 9.5 to 9.6V on a fresh-out-of-the-box battery.  Now let's go into the ugly part: usually the battery will be considered discharged when the per cell voltage drops to as low as 0.8V depending on the manufacturer and datasheet.  Thus, the 550mAh value said above has meaning only if you know up to which voltage per cell the battery was discharged.  Let's assume the best scenery (for us) that the manufacturer specified this rating for 1V per cell.  Then your battery will be putting out 6V only!  Is this enough for your pedal to be usable?  On most pedals this is not the case, as you will get audible clipping in the buffer stages and other unwanted artifacts due to misbiased transistor/fet/mosfet stages, etc.  I woulnd't think that a pedal might work properly with less than 7.5V perhaps, or 1.25V per cell, thus the previous calculation with the 550mAh figure is not valid, unless this rating is for a residual voltage of 1.25V per cell--which I'm pretty sure it's not!  Battery manufacturers specify their ratings according to their convenience, thus 1V or less per cell is typically used.

In summary, the simplified calculation should be taken with a large grain of salt.  According to my current knowledge, it might give you perhaps a "this is the absolute best that we can hope" type of limit.  Perhaps it is not so difficult to define a correction factor, for instance multiplying the figure with a correction factor like 0.7, as in one of the first formulas presented in this thread for carbon-zinc batteries.

Just some ideas for thought.

[gaussmarkov]

great stuff.  thanks very much!

because you mention battery data sheets, i would like to ask where you are finding these.
last time i looked, i could only find one--and it wasn't much of a data sheet.  manufacturers
don't seem to make data sheets readily available.  or am i just not looking hard enough?

--gm

[Seljer]

they are there apparently

http://data.energizer.com/PDFs/522.pdf
http://data.energizer.com/

can't find anything for duracells except for their bulk batteries for commercial use (I think i remember once seeing some places where you could get these cheap, go find some online battery store  ) http://www.duracell.com/procell/products/default.asp

[stm]

OK, here are the facts:

Based on this technical article from Duracell: http://www.duracell.com/oem/Pdf/others/ATB-5.pdf

Table 1 indicates the following figures for a 9V alkaline battery (model MN1604):

"0.58Ah when discharged by a 620 ohms resistor down to 0.8V per cell"

So, 9V over 620 ohms gives 14.5 mA, which is a pretty good number for our purposes, as it can be considered as a representative stompbox consumption (assuming typical stompboxes range from 5mA to 30mA).

Then, 0.58Ah/0.0145A = 40h (forty hours is not bad!)

But wait!  This is down to 0.8V per cell, or 6x0.8V=4.8V final battery voltage, which we know it is useless for most pedals.

How do we reference the 40 hours value to a 1.2V residual voltage per cell (or 7.2V battery voltage)?
Let's take a look at Figure 2 in the article.  It relates voltage v/s duration for three different load conditions for a AA battery (model MN1500).  Looking at the 24 ohms discharge curve we read 51h for 0.8V and 28h for 1.2V.  Thus the time ratio is 28h/51h = 0.55 times.

Using the above info I can infer that I'll have a predicted duration of 40h x 0.55 = 22h until the battery is discharged to 1.2V per cell or 7.2V.

Mmhhh, but the aware reader could ask some embarrasing questions like:

-Why did you choose the 24 ohms curve and not the others?
-How dare you use the results of Figure 2 which are for an AA battery on a 9V battery?

Good questions!  First, there is no other usable data to do this for a 9V battery (MN1604 model) that relates voltage.  Second, in order to make a fair comparison I chose the 24 ohms discharge curve because it represents 1/45th of the rated Ah discharge for the AA battery, while the data for the 9V battery is based on 1/40th of its rated current, thus they are close enough and I can say I am comparing apples to apples.  Well, at least Granny Smith apples to Richards apples, instead of apples to oranges.  Anyways, if you pick the 62 ohms discharge curve you get nearly the same correction factor.  The 3.9 ohms discharge curve is too cluttered to be able to extract useful data visually, however we know in this condition the battery is being discharged at a much higher rate and thus it is not suitable.

Bottom line:

For "typical" or "average" stompbox usage you could get a better estimate of your battery duration (based on Duracell 9V battery data) as:

0.55 * 580 mAh / pedal_current

where the pedal current is expressed in mA.

[markm]

I don't know about all this math stuff.....
My Energizers just keep on going......and going.....and going....
sorry.