Stop when vIn below 3v

Thread Starter

moeburn

Joined Aug 16, 2013
31
Hello folks!

I have a simple AA battery USB charger. It can take a vIn from 2.4v-4.5v, and output a vOut of 5v, perfect for charging smartphones and such using NiMH batteries.

But it can only output 1 amp if vIn > 3.5v. This means using 3 NiMH batteries. But you should never discharge a NiMH battery below 1v, or you will damage the battery. With 3 batteries, this means the circuit should stop when vIn reaches 3v. But the circuit keeps on chuggin to well below 2.4v, overdischarging my batteries.

Now I'm no electronics wiz. I'm just a project kiddie. I bought the finished circuit on ebay for $5, soldered a battery and a switch to it, and stuffed it in an Altoids tin. I asked on another message board if there was a simple solution to my problem, and they started talking about comparators and hysteresis and SMT components and getting a PCB prefab'd by a professional PCB manufacturer... all of which are a bit above my skill level (and more effort than I was hoping to put in).

Is there any simple way of stopping my circuit when vIn reaches 3v? Because at the moment, the simplest way I have is checking the batteries with a voltmeter every 30 minutes or so, and then manually flipping the switch when they get too low.

Thanks for any help!
 

bance

Joined Aug 11, 2012
315
RC guys have this problem quite a lot, because they often use lithium batteries that have no protection circuit..... You could use a comparator circuit, it's really not that difficult.....

Or try an ebay search for battery checker, you can get them quite cheaply, however they won't shut your circuit down, only allow you to monitor what's going on.

HTH Steve.
 

ronv

Joined Nov 12, 2008
3,770
I think that one above stays on all the time. Here is a cute circuit that uses a TL431 as a comparator. The PFET needs to be a logic level fet because of the low voltage. It has a little hysteresis so it doesn't turn off and then turn right back on because the load is gone. Turns off at about 3 volts and back on at about 3.2 volts.
 

Attachments

Thread Starter

moeburn

Joined Aug 16, 2013
31
Oh, and the other requirement was that it fits in the 1"³-1.5"³ remaining space that I have in my Altoids tin.

Is it possible to make one of these comparator circuits, without using any difficult-to-solder SMT components, or ordering a prefab PCB?

Thanks for all your help guys, I really appreciate it!
 

ronv

Joined Nov 12, 2008
3,770
It should fit and yes we could probably find thru hole parts to be put on a small breadboard. Can you read the schematic and solder?
 

Thread Starter

moeburn

Joined Aug 16, 2013
31
It should fit and yes we could probably find thru hole parts to be put on a small breadboard. Can you read the schematic and solder?
Great! I looked at your schematic you posted, it looks like exactly what I needed! Although the graph in the picture seems to imply that it turns off at ~3.6v, not 3v, is the graph accurate? EDIT: Sorry, I realised that the green line is actually current, not voltage. The amount of current determines the rate at which I can charge a phone, will it be significantly affected by putting this circuit between?

I already have a ton of resistors on hand, so all I would need to find would be the TL431ED and FDR840P.

I have no trouble soldering thru-hole sized parts, and I can even translate a (relatively simple, like the one you posted) schematic into a breadboard layout. For a reference as to my skill level, I think the most difficult projects I have successfully completed were a ChuMoy amplifier and a TS808 guitar effect pedal. But I just bought premade PCBs for those projects, bought the parts, and soldered them in their appropriate holes. What I have difficulty with is designing a compact layout based on a schematic.
 

ronv

Joined Nov 12, 2008
3,770
OK. Let me look for a good fet. It should not bother your phone. We are going for 1 amp maximum - correct?
 

Thread Starter

moeburn

Joined Aug 16, 2013
31
OK. Let me look for a good fet. It should not bother your phone. We are going for 1 amp maximum - correct?
Yeah, 1A max. The DC-DC converter circuit claims that it can do 1.2A if the voltage is >4.0v, but I've never seen it do more than 1A, even with 3 fully charged NiMHs (3x1.4v = 4.2v).

I didn't mean bothering the phone, i meant would I get the same (or close to the same) amount of current, or would the charging speed be slightly less because the current would be limited somehow?
 

tindel

Joined Sep 16, 2012
936
The circuit Mr. Chips posted will work... kinda - the problem is that it doesn't provide hysteresis. It will turn on somewhere between 2.8 and 3.5V and shut-off at the same point.

Adding hysteresis (positive feedback) will force the boost converter to turn off at say 3V and won't allow it to turn back on until 3.7V or something similar.

The thing you will run into (and isn't simulated in ronv's simulation) is that the battery impedance will kill you if you are not careful. Battery impedance of NiMH can be as high as 0.5ohm with 3 in series. So if you're pulling 1A out of the batteries... you turn off at 3.0V then your battery will then show 3.5V, and if you don't have adequate hysteresis (more than half a volt) then you will oscillate.

I had started to do a similar circuit to ronv's, but ultimately I like his better because it consumes less parts. I'm not quite convinced, however, that it will work because of the battery impedance issue I described earlier.

I have attached my thoughts, but I'm not quite happy with mine either (it comes out of lockout too slow).
 

Attachments

tindel

Joined Sep 16, 2012
936
Oh, two questions:

1. Is the output current you're supplying to the USB 1A or is the battery supplying 1A to the converter? Big difference!
2. Does your boost converter have a shutdown pin - you can possibly use it to your advantage.
 

tindel

Joined Sep 16, 2012
936
Here's a second attempt - note that mine ended up very similiar to ronv's schematic. I'm not crazy about this scheme because 3.0V battery-1.2V reference is pretty tight on guaranteeing the FET will turn on completely. I'd really prefer another stage, similar to what I depicted in my last post, to drive the transistor hard into saturation.

Adjust R5 for hysteresis
Adjust R3/R1 to set voltage trip.

Note that if you make the R5 too small, the circuit will latch.
 

Attachments

Thread Starter

moeburn

Joined Aug 16, 2013
31
Oh, two questions:

1. Is the output current you're supplying to the USB 1A or is the battery supplying 1A to the converter? Big difference!
2. Does your boost converter have a shutdown pin - you can possibly use it to your advantage.
Okay, ronv said that his schematic has 0.2v of hysteresis: "Turns off at about 3 volts and back on at about 3.2 volts."

I have experienced the problem of a lack of hysteresis with the circuit the way it is now. Before I even knew about overdischarging batteries, I was letting them discharge completely, and once the circuit reached the minimum threshold voltage, it would turn on and off about a thousand times per second, which was a good indicator to remind me to turn the charger off, because it would make my phone scream and beep and vibrate like the world was ending. But yeah, some hysteresis would be nice, but not entirely important.

To answer your questions:
1. I measured the current between the USB socket of the charger, and the USB socket of the phone, to be about 1A at start, slowly decreasing to level off at about 0.75A as the battery's voltage decreased to 1.2v ea.
2. It might, I don't know. I didn't make this one. I bought this one:

DC-DC Converter Step Up Boost Module 2-5V to 5V 1200mA 1.2A

I also made a MintyBoost by hand, and I know that it has a shutdown pin, but I couldn't figure out the chip ID number on this circuit that I bought, because they used an SMT IC and it's too hard to read the numbers on it to look up the datasheet, or they may very well be on the underside of the chip.
 

ronv

Joined Nov 12, 2008
3,770
Tindle is right. It might chatter at .2 volts. You cab change R6 to 36K and that will make it .8 volts. From the internal resistance standpoint Nicd would be better. It has about 1/2 the internal resistance of the Nmhd.
 

Thread Starter

moeburn

Joined Aug 16, 2013
31
Tindle is right. It might chatter at .2 volts. You cab change R6 to 36K and that will make it .8 volts. From the internal resistance standpoint Nicd would be better. It has about 1/2 the internal resistance of the Nmhd.
Again, I really don't care about hysteresis as much as I care about a voltage minimum. NiCads are out of the question; three 2300mAh NiMHs, discharged to 0.7v ea (since I never tested discharging them properly to only 1v ea), only gives me about 3-4wH of power (based on a 1850mAh 3.8v Liion battery being charged by about 50%).

While it would be a nice bonus, there could be no hysteresis at all for all I care, as long as the circuit oscillates between on and off at 3v, and not 2.4v.
 

tindel

Joined Sep 16, 2012
936
If this thing is really putting out 1A, then you are pulling 3-4A out of your batteries... multiply that by about 0.350ohm and you'll have somewhere around 1V of drop out of you battery alone! You need some good hysteresis.

If you don't care about hysteresis in this circuit I suggest 3 things (not joking).
1) keep a electrical grade fire extinguisher near by.
2) Discharge your batteries under a vent hood of some sort.
3) Put it in a shielded box so you don't get a visit from our pals at the FCC.
 

ronv

Joined Nov 12, 2008
3,770
I don't think it's 3 or 4 amps, but I would have expected the cell battery to get to about 75-80% on the 3 AA's.
 

tindel

Joined Sep 16, 2012
936
Ah yes, you're probably right. I've never done the calculations, but it seems that the current out of the battery is a function of the load, inductor, switch frequency, and input voltage. So yes, I'd expect the battery current to go down as voltage goes up (assuming the inductor, load, and switch frequency stay constant.)

I do know when my converter is putting out 5V with a 2.4V input and 0.5A load my batteries are drawing 2.5A.
 
Top