• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Something very strange with current measurement.

Status
Not open for further replies.

bigal_scorpio

Active Member
Hi to all and seasons greetings!

I recently bought some Li-Ion battery protection boards with the idea of making a few extra batteries for our power tools as my son now has a 3d printer and can make the housings for them.

While in the testing process I have noticed a few problems, like the tools will only function if the trigger is pressed slowly and if that is not observed then the battery must be removed and replaced before the circuit will reset. I can understand the reset problem but the boards are rated 20A constant load and 30A cutout limit. The drill I am testing on is a Makita 18v battery drill and however I try I can't get it to draw more than 8A without actually stalling it. Odd!

However the real thing that stumps me is that when not in use the board gets warm. The row of 7 tiny 3 legged devices marked 431, which seem to be NPN transistors get warm to the touch, hot to the lips which are a lot more sensitive and my go to form of temp test. Now the text for the boards says the static draw is 0 to 500uA which if true would surely mean nothing got warm enough to feel? So I decided to test the drain on the batteries.

Now I got really baffled. With the boards plugged onto the battery pack and full cells (the boards actually balanced the cells to within 1mV!) and testing the current through the wire from the battery neg to the BAT- pad on the board with the meter on its 200mA scale it wavers between 0.4 and 0.5, however I kept lowering the scale and eventually measuring on the 200uA range I get 0.4 how can it be both? So I dragged out my good meter and on the 200mA scale it shows 13.33, on the 20mA scale it was 4.35 and on the 2mA scale it is 0.583 What is happening here?

If even one of my measurements if close to true how is the board getting warm and how long would anyone think the cells would last before flat. In all the time I have used PIC microcontrollers with the low usage sleep modes I have never felt anything even approaching warmth from them so is it possible that these boards could be only using microamps and still become warm?

Sorry for the long tome but I am just stumped. Oh I nearly forgot to put the link to my boards on. Doh!http://www.banggood.com/7S-Cells-4_...ery-Protection-Board-p-1063304.html?rmmds=buy

Thanks, Al
 

JLNY

Active Member
It is likely that there is a current spike right when the tool spins up that is higher than the 30A cutout and is tripping the protection circuit.

As for the boards getting warm, that is unclear. the laws of thermodynamics would suggest that it would take more than a few uA of current to make something get noticeably warm in open air. It is possible that the sense resistance of your meters are changing when you switch the ranges, so whatever is causing the current reading may be dependent on the resistance in series with the battery?
 

dr pepper

Well-Known Member
Most Helpful Member
As above for the start up issue, and maybe the onboard protection will only allow you 8 amps.
As for the warm its a long shot but I wonder if the cells are trying to balance themselves, dumping energy in the ones that are high in voltage, if there are differences in the cells then in discharge the voltages may get unbalanced.
 

bigal_scorpio

Active Member
Hi JLNY,

Hmm, if there is a spike it must be very fast so I wonder if I could somehow damp the sensing?

I see what you mean about the ranges but which one is likely to be nearest to true?

Al
 

spec

Well-Known Member
Most Helpful Member
Hi BS,

Here are some things to try:

Charge your battery pack, with the balance circuit connected, and then discharge the battery pack with a resistor a few times and see if the system sorts itself out.

When the batteries are idle make sure that each battery is close to 3.6V and that all batteries are the same voltage.

Give this a whorl: put a 47uF or bigger aluminium electrolytic capacitor across each battery and also across the whole battery pack. Add a 100nF of higher ceramic capacitor across the same points.

spec
 

bigal_scorpio

Active Member
Hi dr pepper,

It does state 20A in the blurb, and I have tried 4 of them so its a prob with them all. As for the discharge theory, the cells were newly charged and not used at all when I discovered this? They were all within .02 mV.

Al
 

bigal_scorpio

Active Member
Hi Spec,

When the batteries are charged the boards manage to balance them superbly, all being within 4.19 to 4.21v

I will try your suggestions and report on the outcome.

Al
 

bigal_scorpio

Active Member
More thoughts.

I just realised that I can't find a current sense resistor for the main connections at all. The only reasonable sized ones are in the interlink circuitry and I assumed these would be charging senses as there is one for each cell. I am also puzzled at the use of 2 x opto isolators on each cell line. What could they be for?
Edit: In fact how does it do any of the supposed stuff with the only IC's being what seems to be a surface mount 4011 logic chip and only the tiny DW01u lithium charge protection chips?

Al
 
Last edited:

JLNY

Active Member
Hi JLNY,

Hmm, if there is a spike it must be very fast so I wonder if I could somehow damp the sensing?

I see what you mean about the ranges but which one is likely to be nearest to true?

Al
Yes, it might only take a pulse on the order of a few microseconds to trip the circuit. Possibly a capacitor across the output of the battery pack to absorb some of the current spike, or maybe a small choke in series might help. You might be able to measure magnitude of the spike on a digital storage oscilloscope (or analog storage if you have one of those) by connecting a very low value resistor (maybe 0.01 Ohms or so) in series with the battery and measuring the voltage across it right at the moment you turn on the drill. This would give you a way of seeing what effect different damping methods are having.

Probably the range with the lowest series resistance would be the closest to being "correct" (generally higher current ranges have a lower sense resistance), but it is probably more important to determine why there is a current in the first place before we worry about measuring the exact value.
 

MrAl

Well-Known Member
Most Helpful Member
Hi,

My drill measured 30 or 35 amps stall current and that was with extra series resistance and a series high current diode.
These drills draw a LOT of current, period. You would be lucky to get one to work well with a 20 amp limit.
The overcurrent detect delay is 75ms so that's not much time really in the realm of DC motors or any motor really.

Try two in parallel :)

You are going to need more than 20 amps anyway, maybe you could devise a current shunt to shunt some of the current around the detect circuit.

But my main question here is where did you get the Li-ion cells? They may have their own protection circuit in there.
To get cells that would work well with drills you'd have to buy the ones that have high current drain capability. I had to do that with a very high powered LED light that i had which is rated something like 70 watts (at 4v that is).

If you have a DC clamp on ammeter you can measure the stall current.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top