# Math Checking

Status
Not open for further replies.

#### Dragon Tamer

##### Member
i have been working on a MAX712 battery charger, and i was a little unsure of some of my calculations. my problem lies with the value of Rsense, i plan to have a charge time of 4.4 hours, charging 8 series AA cells at 2.5A hours each. the 2 results i got are 0.4Ω and 2.27Ω of resistance. one is just the inverted of the other. are my calculations correct?

#### MrAl

##### Well-Known Member
Hi,

What is the charge current supposed to be and what is the sense voltage supposed to be?

#### Dragon Tamer

##### Member
i'm assuming that the charge current is 2500mA like the batteries. the Rsense voltage is just supposed to be higher then ground. that's all i was able to pick up on from the data sheet.

#### MrAl

##### Well-Known Member
Hi again,

It looks like the sense voltage is 250mv, but the max current is 2 amps, not 2.5 amps.

In any case, a resistor value of 0.250 ohms should provide a current of 1 amp, and you
can check this first if you like. To get 2 amps you would have to use 0.125 ohms.

#### Dragon Tamer

##### Member
thank you, i will give it a try

#### colin55

##### Well-Known Member
What is the mAh capacity of the AA cells?
How fast do you want to charge them?

#### Dragon Tamer

##### Member
the AA cells have a curent rating of 2500mAh. i calculated that a charge time of 4.4 hours will keep the charge curent under the maximum power disipation levels of the parts i used. what i am a little woried about is the fact that when i measure the voltage output to what would be the load, the voltage jumps to 15V when the batteries are removed. this is not of as much of a concern as when you remember that in the final project the power is going to be removed from the MAX712 while the batteries stay in place (cells are never intended to be removed).

#### audioguru

##### Well-Known Member
2500mAh cells are charged at 568mA for 4.4 hours, not 1A and not 2A.
But the cells get hot when charged at such a high current. The heat is wasted energy so additional charging time or a higher charging current must be used so that the cells become fully charged.

#### MrAl

##### Well-Known Member
Hi again,

If you want to keep the current at around 500ma then you would use a 0.5 ohm resistor.
Charge time for a 2500maH NiMH cell charging at 500ma would be 5.5 hours.
If you can push up to 1 amp then you'll halve that time.

#### Dragon Tamer

##### Member
i think i might be able to pus one amp. now i have 2 more questions.

1.) what kind of tolerances can i expect?

2.) for a test circuit that has been giving me some trouble, i am using 4 AAA cells, 2 are rated for 900mAh, and 2 are rated for 850mAh. i think they're charging, but not by the power of the MAX712. i have set it up with an LED indicator on pin 8, and it won't come on.

and it still concerns me a little that the output of the PNP is 15V without the batterys installed.

#### MrAl

##### Well-Known Member
Hi again,

The PNP could have some leakage current flowing in the collector, so to check that you can not use a volt meter alone on the output, but you can try connecting a resistor of say 1k across the output and see that the output voltage goes down perhaps.

What kind of tolerances you are talking about? The voltage of the NiMH cells can not be specified exactly. The output current
is set by the sense resistor which may be 10 percent, and the Vsense is 10 percent, so max deviation in current set should
be plus or minus 20 percent maximum but will probably be much better than that.
The dv/dt is about 2.5mv, which should be just fine.

Last edited:

#### Dragon Tamer

##### Member
The tolerances that I am talking about, are the tolerances of the batteries. How many mA can they have as a tolerance. And thank you for the advice with the PNP I had to stack 3906s till I got the desired power disipation. I found 5 bad transistors out of 8. Thanks again

#### MrAl

##### Well-Known Member
Hi again,

NiMH cells can take quite a bit of current during charge, although they heat up during the process.
For example, a 2500mAh cell can take 10 amps for a very fast charge, at least one fan is required to keep the
cells cool during this time. More typical max current settings would be 1.25 amps for a 2 hour charge,
and 2.5 amps for a one hour charge.
The charge time actually has to be slightly more than the exact calculation...for example a 2500mAh cell
charging with 2.5 amps theoretically should actually take about 1.1 hours to complete, not 1 hour...just something
to keep in mind.
If the cells appear to get too hot then use a small fan to cool them and that should do it.

You may want to look for a better PNP transistor that can handle the required current.

If you were instead talking about the tolerance of the cell as purchased as to the ACTUAL mAh rating vs the
published rating, well that varies by quite a wide margin. The more important thing i have found in the past
is not so much the rating itself, but what has come to be known as the "Self Discharge" of the cell.

The self discharge is a big pain in the neck, because we charge the cell up fully one day and the next week
we find it only has as little as 80 percent of the full charge left already. The week after that it can be as little
as 50 percent of full charge or even less simply because the cell discharges all by itself even without any load
whatsoever connected to it.

The best (and probably the only) way to get around this is to purchase cells that are made to have low self discharge.
The biggest name brand that has this feature is called "Eneloop" and they come in AA and AAA sizes. There are
other brands too however, like Kodak and i think even Rayovac makes these types too now.
The rating itself is a bit lower, like around 2000mAh, but they work so much better it's almost unbelievable.
They really do keep their charge for a long time too, like regular alkalines, and i did a long term test myself to
verify this (i dont take any published ratings on cells for granted anymore after my experiences in the past with
various types and sizes).
These cells are especially good in devices that draw little power, as they dont die down anywhere near as fast as
the 'regular' NiMH cells.

BTW the low self discharge cells have something to that effect written right on the package so you know that they
are low self discharge. Some of them say, "Low self discharge", and others say something like, "holds charge longer",
or something to that effect.

The worst cells for self discharge were apparently the AA Energizer 2500mAh cells. I've heard many complaints about
these cells from many people over the past few years or so. I dont know if they improved them by now or not,
but back when myself and many others tried them we found that they discharged so fast they they were almost
not useful for anything that did not have to eat up all the power the very same day they were charged.
I wrote to the Energizer company and they sent me a coupon for 10 dollars, which i used to buy 4 AAA size NiMH.

Even better yet are the Li-ion cells, which need a different charger but hold their charge for a long time and always
seem to work well, and also have a more predictable discharge curve so you know when the battery is getting low.
Of course their terminal voltage is a bit higher, so the device they are to be used in has to be retrofit a little to get
going. Some devices are very easy to modify, while some are not so easy at all. LED flashlights that use two AA cells
are very easy to modify for example.

Last edited:

#### Dragon Tamer

##### Member
okay, makes sense. as the tome increases, the current decreases, and the current always remains inversly proportional.

i will try to find a better PNP, but until then, i'll have to make do with i have at my disposal.

#### MrAl

##### Well-Known Member
Hi,

Oh sure, that was just a suggestion to get possibly better operation eventually.
Do you order parts online? There are several good places to get these kinds of parts.

Last edited:

#### Dragon Tamer

##### Member
instead of using a PNP transistor (because for me they seem too sensitive to over voltage) I used an IRL3103 P-channel MOSFET. it seems to be working, even though the output is still insanely high. but now the MAX712 knows when there is charging taking place and it will indicate appropreately. you were talking about Li-ion cells, the only thing that i've heard about them is that they can not be left unattend while they are charging. not that that's any kind of real problem. the thing is though i am making this circuit to go into a portable amplifier that i'm making, and i just need to find a smaller transformer to fit in the prodject box. could i charge 12V off of 15V even though the equation said to use a minimum voltage of 16.7V?

#### audioguru

##### Well-Known Member
Eight Ni-MH cells are 11.2V to 13.0V when fully charged. The MAX712 charger IC needs 1.5V more so the minimum input for a linear charger at 1/4C is 14.5V.

#### Dragon Tamer

##### Member
okay, that's good to hear.

#### colin55

##### Well-Known Member
using a PNP transistor (because for me they seem too sensitive to over voltage)

This is totally FALSE. PNP are just as good as NPN in nearly all instances. In the early days of transistors, PNP were not quite as good a NPN, but any major differences have nearly all disappeared.
For general use, you can use PNP or NPN with total interchangeability, providing the circuit is modified to suit.

#### colin55

##### Well-Known Member
If you are going to charge cells at a rate that is less than a "10 hour rate" you are going to get heating of the cell and if you are venturing into "fast-charging" you need a "smart charger" with heat sensing and peak voltage detection to detect when the cells are fiully charged and to prevent overcharging.
Otherwise you should resort to overnight charging, where you won't have any problems at all.

Status
Not open for further replies.

Replies
1
Views
981
Replies
4
Views
1K
Replies
12
Views
5K
Replies
8
Views
1K
Replies
2
Views
880