[time-nuts] Re: Backup DC power with current battery technology?
Denis Dowling
dpd at opsol.com.au
Tue May 28 09:06:34 UTC 2024
Instead of using Schotty diodes and possibly a relay a more modern
solution would be 2 ideal ideal diode circuits. Ideal diodes are simply
mosfets with an additional controller IC to switch them on as needed.
This removes most of the voltage drop of a diode and are they are very
robust in operation. I have used the LM74704 in a few designs. There are
also some modules available from ebay and other places if you don't want
to build the complete circuit.
Regards,
Denis
On 28/05/2024 11:52 am, John Ackermann N8UR via time-nuts wrote:
> All -- thanks for the responses! They've been very helpful.
>
> Just to summarize a couple of interesting/important things I've learned:
>
> *Â Don't confuse generic "lithium" batteries with the specific LiFePo4
> type -- the other/earlier types have significant safety issues that
> the newer technology greatly moderates.
>
> *Â I looked at inverter/charger options but most are much larger than
> I need (seems like 3KVA is about the smallest) and consequently a
> whole bunch more expensive. If you ordinarily have mains power
> available, a traditional UPS seems to do the same thing in more
> reasonable sizes and for less money.
>
> *Â Bob LaJeunesse suggested paralleling AC power supply with battery
> using Schottky diodes. That allows the battery charger to be whatever
> it needs to be without having to power the load as well. And it
> provides an additional redundancy in that you can disconnect the
> battery/charger for maintenance without affect the downstream gear. I
> like that idea a lot and think I will do something along those lines.
>
> Thanks!
> John
> ----
>
> On 5/27/24 11:38, paul swed via time-nuts wrote:
>> Good morning to the group. Since this thread started I have been reading
>> about the LiFePo4 batteries.
>> They really can be treated much more like a traditional 12V battery
>> in some
>> respects. It seems they come with a battery management
>> system internally. Price is easily 3-4X typical batteries. Though
>> honestly
>> PB batteries are all over the place in cost.
>> Very hard to compare.
>> There are youtube videos of LiFePo4 tear downs and those are really
>> interesting in the wide range of mechanical construction.
>> By looking at the specs you have no idea what you are buying. Well
>> the same
>> goes for PB batteries....
>> John thanks for getting the mind working and relooking at the battery
>> technology.
>> Paul
>> WB8TSL
>>
>> On Mon, May 27, 2024 at 11:07â¯AM Germán Herrera via time-nuts <
>> time-nuts at lists.febo.com> wrote:
>>
>>> Hi John
>>>
>>> Probably a little overkill, but I have good experience with the
>>> Mini-Box
>>> OpenUPS boards (the original, one not the OpenUPS2). It can be setup
>>> with multiple cell chemistries, does balancing, has wide input/output
>>> voltage, 6A (10A peak) output current, etc.. Only issue (beside
>>> cost) is
>>> that it's switching, so maybe the added noise makes this a no-go.
>>>
>>> BR, German
>>>
>>> On 2024-05-26 11:39, John Ackermann N8UR via time-nuts wrote:
>>>> I need to replace the batteries in my 28 VDC power distribution system
>>>> in the clock room. It's currently two series 12V AGM batteries
>>>> with an
>>>> IOTA Engineering float charger.
>>>>
>>>> This system has two purposes: (a) primary power for some OCXOs and
>>>> other gear; and (b) failover power for some AC/DC gear like HP
>>>> standards. Under normal conditions, power draw is 2 or 3 amps. During
>>>> mains outage, it could reach 10 amps. Now that I have a house
>>>> generator, long run-time isn't a big issue so 25 Ah or so of capacity
>>>> is plenty.
>>>>
>>>> I haven't explored LiFePo4 batteries until now, and I'm having trouble
>>>> finding a charger that can provide significant continuous load current
>>>> while maintaining a battery floated across the line. Does anyone have
>>>> recommendations for something like this?
>>>>
>>>> Thanks,
>>>> John
>>>> _______________________________________________
>>>> time-nuts mailing list -- time-nuts at lists.febo.com
>>>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
>>> _______________________________________________
>>> time-nuts mailing list -- time-nuts at lists.febo.com
>>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
>>>
>> _______________________________________________
>> time-nuts mailing list -- time-nuts at lists.febo.com
>> To unsubscribe send an email to time-nuts-leave at lists.febo.com
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com
More information about the Time-nuts_lists.febo.com
mailing list