LTC4060 leaking batteries!

Hi,

I have a project using E2H NIMH and a LTC4060 to charge them (attached picture of circuit)

I have several E2H battery packs and all have started to leak, have I got something wrong, I have checked and checked but I think now I can't see wood for trees!

5v input with a max 5A current supply

Thx in advance

If you have source and drain of your MOSFET reversed, it will probably damage your batteries. Otherwise, I don't see anything obvious. Have you monitored the voltage on the batteries while they are connected to the charger?

Scarr said:

I have several E2H battery packs and all have started to leak, have I got something wrong,

Click to expand...

If they are leaking then you are over charging them, I would suggest you monitor both the voltage and the current to see what's happening (monitoring temperature might be helpful as well).

The "el-cheapo" batteries at the Dollar Store are also leaking and they are just sitting there without being charged.
Are your batteries "no-name" or name-brand??

What do you mean by leak?

Do you mean that they discharge quickly even though no load is connected or that the chemicals are literally leaking out of the battery??

Leak

By leak I mean chemical. However I might have found the problem as the main circuit is normally driven by the battery (2.4v) when I need to update the FW on the Atmega88 the STK500 supplies +5v this gets fed directly to the battery, whilst it's only for a short while if I program it several times or leave it plugged in to the programmer I suspect its damaging the battery.

Anyone comment on if applying +5v for a short period would cause this?

Scarr said:

By leak I mean chemical. However I might have found the problem as the main circuit is normally driven by the battery (2.4v) when I need to update the FW on the Atmega88 the STK500 supplies +5v this gets fed directly to the battery, whilst it's only for a short while if I program it several times or leave it plugged in to the programmer I suspect its damaging the battery.

Anyone comment on if applying +5v for a short period would cause this?

Click to expand...

Almost certainly. It's going to force 5 amps (actually, probably more) through the battery.
I don't understand how you are using the charger. I interpreted the circuit as being a "UPS". The +5V keeps the battery charged through the LTC4060. The load sees +5V through the Schottky diode (well, probably 4.7V). The MOSFET keeps the two circuits isolated. If the +5V goes away, the MOSFET turns on, providing backup power from the battery.
How are you using it?

It's usage

Hi Roff,

Not being very good at electronics I borrowed this circuit, it's supposed to be a battery charging circuit like one for a cordless phones a % of time on charge and a % running on E2H battery pack.

I did make a mistake in what I said about +5v being fed directly, from what I can see of the Mosfet (https://www.electro-tech-online.com/custompdfs/2008/02/FDN338P.pdf) has some sort of diode across D & S, I am going to check this to see what effect it has when a programming voltage is applied to VBATT.

Will update post with results

Yes when I stick the STK500 on it to program the AVR I can measure 4.93v on battery connector! might place a diode in here or just rememeber not to program with battery conected

Anyway thanks all. Usually I find after comming on here it's a really stupid problem caused by me and I could have fixed it myself, however I think it's the fact you have to explain things, it makes you think clearly after getting totally confused.

The diode in the MOSFET should not cause a problem if you have installed the MOSFET correctly, and have used the correct part number.
Is your circuit identical to the schematic you posted?

Circuit checked

Hi Roff,

Yes it all appears correct everything connected the correct way round, if D3 was omited with no voltage on the charge points and +5v was applied to VBAT then would the Mosfet allow +5v to the battery connector?

If this is the case then thats the problem as D3 is missing (doh) but I am unsure why omitting D3 does not cause a problem when charge is applied as charge being applied as you say should isolate the charge circuit and the only way for the main circuit to get power was via D3 in this condition.

Thx for all your help so far!

If this is causing you any hassle don't continue helping as I have a work around (not connecting battery when programming)

If you are applying voltage to Vbat, with no voltage applied to Vchg, then the MOSFET will come on and ruin your batteries, because Vchg is what keeps the MOSFET turned off. That circuit was not meant to have an external voltage applied directly to Vbat. Vbat is the output from the batteries OR Vchg to a load connected between Vbat and GND. D3 allows Vchg to appear at Vbat through it (with a ~0.3V drop), powering the load when it is present.
I'm gonna get redundant here - as I said before, the battery is like a backup source, powering the load on Vbat when Vchg is absent. When Vchg is present, it turns off the MOSFET and powers the load through D3. It seems to me that all you need to do is install D3, and leave +5V connected to Vchg.
And you made another noob mistake - posting a schematic which wasn't really what you have.

To Roff

Hi Roff,

When I said that D3 was not there I didn't mean from a point of view of it's no pads on board, what I meant it's that it was simply not placed on the PCB the pads are there but no component!

I will place D3 on board and also will make sure batt is not connected when programming.

thanks for all your help

P.S. it was hard to see D3 was missing as it's under another component for my project.

It's still not clear to me as to how the MOSFET was turning on. When you applied +5V to VBAT, were you simultaneously removing it from VCHG?