Diode reverse protection and current limiting.

[Ignoramus31174]

15V is way too high. I don't know if that is dangerous, but it probably
damaged the battery. The smart charger I bought uses a bi-color LED to
indicate the charging.

yes, indeed it is all very upsetting. It overcharged it because its
software locked up.

Well, that is a serious battery.

yes, they are quite heavy.

I would contact the company at the
link I provided and see if you can agree on an acceptance criteria for
whatever charger you determine fits your need.
OK.


I have charged a very narrow range of 12V batteries, 50AH to 70AH. Thus
far, no problem. These are not deep discharge, but SLA.

mine are marine deep cycle batteries. i have three: one for my onan
diesel generator, maintained by a 1.5 A battery charger/maintainer,
and two for general purpose fooling around and the boat.

My issue with home brew battery charging is the fecal matter can really
hit the fan if you screw up. Nothing like burning down the house to
save $50 on a commercial charger. When I was designing chargers, I'd
run the charger in an environmental oven, just in case something bad
happens.

you mean the battery would explode, or what. My plan is to put a timer
on the power supply, like this

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220023368642

i

 

[Chris Jones]

I was going by the speciication on the auction document..


Seems doable.

I agree that it should be possible. Because they are Darlingtons, they will
probably need to drop a minimum of more than 1 Volt (maybe even 2V) whereas
a MOSFET can work with much less voltage dropped across it, so
theoretically could give more efficiency, especially in a switching
circuit. On the other hand, if you already have these darlington
transistors, and if you can get rid of all that heat, then they should work
in a linear regulator. You might want to consider "fold-back current
limiting", which means that you apply a current limit which is say 50A at
14.5V, but the current limit might drop to only 10A when the output voltage
is near 0V. The advantage of this setup is that the output transistors
will be less likely to melt if the output ever gets short circuited.
(There is more voltage dropped across the pass transistors when the output
is short-circuited, compared to normal operation. Since power dissipation
is the product of voltage dropped across the transistor multiplied by the
current, it is useful to drop the current limit when the transistor has
lots of voltage across it.) I think there is something about this in the
book "The Art of Electronics" which I recommend highly.

If you make a switching regulator, then you would need an inductor, a
current sense resistor, and two capacitors that could handle the required
current. (50 Amps ripple in the capacitors will make them hot if they are
not designed for it, and 50 Amps DC will saturate the core of an inductor
if it is not designed for it. Old computer power supplies, especially
bigger ones than you would find in a PC e.g. old minicomputers, would be a
source for these parts.) Apart from that you would need some MOSFETs,
these are pretty small and cheap for 50 Amps, and probably some diodes and
control circuitry. It should not need much heatsinking because it would be
efficient.

Whichever system you use, it might be a good idea to put a relay / contactor
and fuse or circuit breaker between the battery and the charger. You could
connect the contactor to a comparator circuit which would disconnect the
charger from the battery in the event that something in the charger fails
resulting in wrong voltage or current. This could also prevent discharging
when the charger is switched off.

Chris

 

[[email protected]]

yes, indeed it is all very upsetting. It overcharged it because its
software locked up.


yes, they are quite heavy.


mine are marine deep cycle batteries. i have three: one for my onan
diesel generator, maintained by a 1.5 A battery charger/maintainer,
and two for general purpose fooling around and the boat.


you mean the battery would explode, or what. My plan is to put a timer
on the power supply, like this

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220023368642

i

Well, a timer is a start, but for lead acid, a constant float is good.
Just off the top of my head you need these protections:
1) undervoltage. My charger refuses to charge anything under 4V. I
don't know where they got that limit
2) overvoltage
3) current limiting
4) reverse current

Your diode takes care of reverse current. I suppose your power supply
will take care of overvoltage. However, the smart charger I have is
multi-stage. I believe it is current feed up to a point, then goes to
voltage float.

Gates Energy published a book on charging lead acid. I can't find Gates
Energy on the web unfortunately.

 

[Ignoramus11477]

I agree that it should be possible. Because they are Darlingtons, they will
probably need to drop a minimum of more than 1 Volt (maybe even 2V) whereas
a MOSFET can work with much less voltage dropped across it, so
theoretically could give more efficiency, especially in a switching
circuit. On the other hand, if you already have these darlington
transistors, and if you can get rid of all that heat, then they should work
in a linear regulator.

I think that getting rid of 200 watts of heat on that heatsink is a no
brainer.

You might want to consider "fold-back current limiting", which means
that you apply a current limit which is say 50A at 14.5V, but the
current limit might drop to only 10A when the output voltage is near
0V. The advantage of this setup is that the output transistors will
be less likely to melt if the output ever gets short
circuited. (There is more voltage dropped across the pass
transistors when the output is short-circuited, compared to normal
operation. Since power dissipation is the product of voltage
dropped across the transistor multiplied by the current, it is
useful to drop the current limit when the transistor has lots of
voltage across it.) I think there is something about this in the
book "The Art of Electronics" which I recommend highly.

Chris, I have AoE, so if you can refer me to some particular page or
chapter, that's all I need to get started. I am looking for something
simple that I could use to drive motors, charge batteries,
electroplate etc.

Whichever system you use, it might be a good idea to put a relay / contactor
and fuse or circuit breaker between the battery and the charger. You could
connect the contactor to a comparator circuit which would disconnect the
charger from the battery in the event that something in the charger fails
resulting in wrong voltage or current. This could also prevent discharging
when the charger is switched off.

Agreed on this, I will start with a fused circuit.

i

 

[Ignoramus11477]

Well, a timer is a start, but for lead acid, a constant float is good.

What is a constant float?

Just off the top of my head you need these protections:
1) undervoltage. My charger refuses to charge anything under 4V. I
don't know where they got that limit

I would rather avoid that.

2) overvoltage
yes

3) current limiting
yes

4) reverse current

Your diode takes care of reverse current.
Yep.

I suppose your power supply will take care of overvoltage.

Yes, it is CV by nature.

However, the smart charger I have is multi-stage. I believe it is
current feed up to a point, then goes to voltage float.

Gates Energy published a book on charging lead acid. I can't find
Gates Energy on the web unfortunately.

I would be interested in anything on this matter. I think that battery
capacity is a necessary input to a proper charging algorithm, and it
is missing on all those "smart chargers".

i

 

[neon]

I give you a reasonable idea not to put a diode in series what for? current flow from + to - it never reverse unless you change the polarity and for the life of me i don't understand the 100 amps charge not on a car battery for sure unless totaly dead and in that case throw it away. a car battery being charged at a current of 100 amps it will become hot and eventualy fail faster why do you want to do that. A car generator will provide that as a safety of components not meant to charge at 100 amps rate. anyhow it is regulated by means of current not voltage like maybe max 20 amps if discharged. And finaly battery will loose charge with time and age. And the best thing that you can do for a battery is cycle on-off. they have memory keep it fully charged for a long time they tend to go to sleep. the voltage is there but not the power. And finaly they care less about the voltage they need current to charge the voltage has to be greater to fully charge. there has to be a load when off to drain the power you have allready diodes in the P.S. the possibility is that the F/B regulator is active when off there you may put a diode on that circuit. or at least find out where the current goes and why remenber +to-.

 

[Chris Jones]

Ignoramus11477 wrote:

[snip]

Chris, I have AoE, so if you can refer me to some particular page or
chapter, that's all I need to get started. I am looking for something
simple that I could use to drive motors, charge batteries,
electroplate etc.

P 316 in 2nd ed.

[snip]

 

[Ignoramus13308]

Ignoramus11477 wrote:

[snip]

Chris, I have AoE, so if you can refer me to some particular page or
chapter, that's all I need to get started. I am looking for something
simple that I could use to drive motors, charge batteries,
electroplate etc.

P 316 in 2nd ed.

[snip]

Chris, thanks, I looked at it and it is very straightforward. Thank
you.

i

 

[[email protected]]

What is a constant float?

A float charge is basically put 13.8V on the battery and let it draw
what it wants.

I would rather avoid that.


Yes, it is CV by nature.


I would be interested in anything on this matter. I think that battery
capacity is a necessary input to a proper charging algorithm, and it
is missing on all those "smart chargers".

The smart chargers that do a current dump followed by float are matched
to the size of the battery. That is one of the reason the chargers are
given a range.

 

[jasen]

Thanks to many people regarding your earlier advise on modifying my
TIG welder. The sequel to that saga is that the welder works and I use
it quite often.

Now I have a smaller modification to ponder.

After several unpleasant experiences with battery chargers (which did
not work) I want to try something different. My typical use of a
battery charger is to use it to replenish some battery that does a lot
of engine cranking.

I have a power supply that I am thinking should become my battery
charger. It is a PP-1104B/G. If you google for PP-1104C/G you will
find a description of a very similar power supply. It has a dual
range: 13-19 volts at up to 100 amps, and 26-39 volts up to 50
amps. It is a constant voltage power supply with no current limiting
other than built in breakers. I added a few not very relevalt things
to it (like handles for easier handling), but it has a few electrical
quirks that I would like corrected.

The first is that when it is turned off, it drains the battery (ie the
current flows "backwards" as opposed to current flow when charging)
and quickly consumes a lot of amps. It appears that adding a diode
with a small heatsink should take care of the issue.

you do like industrial sized machines!

For battery chargers current limiting is often done by using SCRs
in the rectifier.

 

[jasen]

Chris, you know me. I am basically looking to see what would it entail
to make the cheapest/easiest solution to making a more versatile power
supply out of this PP-1104B/G. If the power converter is lossy, I
think that in the end it is not a huge consideration.

Is it true that a linear transistor based regulator is basically an
extremely simple solution?

apart from the heatsinking, yes.

Bye.
Jasen