Discussion in 'Electronic Basics' started by mg, Jun 20, 2005.

2. ### Guest

If I understand the criterion correctly, everything between 1)-6) must
have passed the test.; i.e a known load for 5.5 hours did not discharge
below 15.0V. All others failed. Consider result 7), at best the loaded
battery dropped to 15.0V in 5.5 hours then lost an additional 0.1V with
no load on it in 8 hours. Clearly this is not a battery for long term
standby application.

I agree that he wants to minimize uncetrainty. The test I outlined is
not very uncertain regarding ability to deliver a particular charge in
a particular time and having a low self-discharge rate.

3. ### ehsjrGuest

Ok, readings 1) through 6) range from 15.5 to 15.0

So he measures at ~8 hours and his battery reads 15.5.
How does that answer his post:
"We would like to add one additional feature to our discharger. Some
sort of voltmeter that we could hold the last reading. This would
allow us to double check that we stop the discharge process at 15v."

Does the 15.5 volt reading mean that: a) the discharger was not
finished, or b) the discharger was finished, but shut off at
other than 15 volts or c) the discharger shut off at 15 volts,
but battery voltage without load rose to 15.5?

The same questions apply for the other voltages posted.
He ends up where he started - uncertain as to whether the
discharger is turning off at 15 volts.

Ed

4. ### ehsjrGuest

Change R4 to 560 ohms. This will ensure that the
discharger can still turn on, even if the battery
that is to be discharged is almost fully discharged
(say 15.1 volts) when it is put in the discharger.
With the 1K for R4, it is highly probable that it
will turn on at 15.1 - with the 560 ohm for R4, it
is definite. The change of value does not affect
the 15 volt shut off voltage.
Ed

5. ### Guest

You are correct if a voltage comparator is used to terminate the test.

I suggested that the comparator isn't necessary. Just see how much the
voltage falls after some fixed time under load (which is essentially
equivalent to what the originally proposed test does.)

Voltage rebound creates uncertainty if the open circuit voltage
measurement takes place too soon after the load is turned off. This can
be mitigated by waiting "long enough" or better yet by reinstating the

Dave

6. ### ehsjrGuest

No it is not, by a long shot. See below.
A NiCd should not be discharged below 1 volt per cell.
Terminating the discharge based on time rather than
voltage is a damn fool idea that could damage the batteries,
by discharging them well below that level.

It is nothing like terminating the discharge at a specific
voltage level (~1 V per cell). An 18V (nominal) pack subjected
to a discharge of 5.5 hours could end up ruined by that. Its
terminal voltage could be anywhere between 0 and 18 volts.
The same pack, subjected instead to a discharge that terminates
at 15 volts, will end up close to 15 volts.

Ed

7. ### ehsjrGuest

Sue,

This idea had great promise, but I got real disappointing
results. I ordered a 10 uf polypropylene and received it
yesterday. (Aerovox ARPM 10uF 175 VAC 9833)
http://www.aerovox.com/pdf/DC_Axial_Cat-11a.pdf
It is cleary NOS (August 1998 never used)
The maximum insulation resistance is 500,000 meg per the
spec sheet. That seems to contradict the curve on the sheet
on page 5. It looks like the actual is 1,000,000 meg.

I charged it to 30.2 volts for 5 minutes, then disconnected
the test leads. I came back 2 hours later and connected
the DMM - 15.xx volts! It was so low (ie losing 1/2 the
voltage in 2 hours) I didn't even bother to record the decimals.

Then I thought a super cap and a 5:1 voltage divider might
work for the OP. So I charged one up to 4 volts (without a
divider) and measured it 3 hours later - 3 volts. I tried it
again, charging it to 5 volts, and measured 3 hours later -
4 volts.

In all cases, I charged for 5 minutes, then completely
disconnected. All voltage readings after the elapsed time were
made in under 5 seconds. I even tried charging and leaving
the DMM connected to the super cap (.1 farad, 5 V) - the DMM
does not discharge the cap rapidly enough to screw up the

Whatinthehell is going on? The polypropylene idea seems to
make sense. So does the super cap. Actual results do not
bear this out. Both caps I tried came from the same supplier,
and they might be seconds, rejects, call them what you will.
I am beginning to suspect that. I hope someone else will
try this out. It will be a while before I need to order
more parts - when I do it will be from Digikey, where I
won't suspect the parts as I do with these. In the meantime,
it sure would be nice to hear the results others get.

Ed

8. ### John FieldsGuest

---
I don't wanna sound insulting or anything, but I think you're wasting
your time as far as zeroing in on what the OP wants is concerned.

ISTM that what he's really concerned with is the elapsed time between
when the discharger starts and when the battery gets to 15V, the
reason being that if the battery discharges to 15V in _less than_ some
specified time with a given load on the battery, then the battery's
capacity is starting to wane and the battery needs to be replaced.

To that end, then, a clock which starts accumulating time when the
battery starts being discharged and stops when the battery voltage
falls to 15V seems to fill the bill in that all that's required is to
read the clock to determine the time from start to finish and,
therefore, to determine the condition of the battery.

9. ### Roy LewallenGuest

It's very tough to get an extremely high resistance. Even with a perfect
capacitor, you end up with leakage around the outside. I'm not an expert
at this, but a friend who is tells me that what you need to do is wash
the surface very well with deionized water. You can use alcohol or
something first if it's grungy, but the DI water rinse is essential.
Then don't touch it!

I assume that you could use distilled water from the grocery store in
place of the DI water, but I'm sure some of the chemists who frequent
this group could tell you for sure.

Roy Lewallen

10. ### ehsjrGuest

Doesn't sound insulting at all. I believe your understanding
(below) is correct. At least I understand it the same way.
But I think he has already implemented the clock idea, so
recording the elapsed time isn't a problem. I sent him
a circuit design back in January which included an AC socket
switched on by a relay as long as it was energized. I think
he used the relay/AC socket idea, but the TL431 & relay driver
circuit were not suitable for his level of familiarity with
electronics.

I think the problem is that the OP doesn't trust his discharger
to shut off at 15 volts each time. He can't wire up much more
than a relay circuit, so I was hopeful that Sue's idea would
work.

As far as wasting my time with regard to designing a solution
for him - at this point, I have to agree with you. I'm out of
ideas he would be able to implement. )-:

Ed

11. ### ehsjrGuest

Roy,

Thanks. That probably is the killer. The humidity
is very high, and no effort was made to clean the
parts, let alone wash them with de-ionized water.
They have my fingerprints all over them.

Ed