Battery Boxes...What One Company Designed

[this is gett | >Perhaps. However, I just looked at the hookup instructions and it appears | >that the outputs are directly paralleled with the battery bank, i.e., whatever | >over-current protection the devices offer does not actually protect the load | >side. That suggests that no matter the listing the output is going to have to | >be treated like the output of any other battery bank which can deliver very | >high current, requiring, e.g., (expensive) class T fuses. And of course, the | >wiring at that point isn't even class 1. (I used to think that it could be | >class 1 subpar (b) which goes to 600V with no power limitations, but that's | >for signaling/control only...) | | Yes. See the ASCII circuit diagram in my previous post and at the end of this | one. which I think makes this clear. Over-current protection is included.

Yes, you were clear. I was confused by the description on the manufacturer's site. With all the talk of sophisticated over-current detection I thought the supply had to interpose some electronics between the battery string and the output. I guess the protection is just for the wires to the battery. :( In case you care, I'm pretty sure that there do exist supplies that work the way I thought these work.

| Where do I find out about class T fuses ?

A Google search seems to turn up quite a few hits. One thing I noticed is that they aren't as absurdly expensive as I remembered...

| Searching the 2005 draft NEC came up | blank. Why do I need them?

To be honest, I'm not sure you do. This came up some years ago when I added a remote battery string to a UPS. To my annoyance I found that while inexpensive 100A automotive fuses could be used for similar applications in cars, boats, mobile homes, and even internally to said UPS and its packaged expansion batteries, I had to use a class T fuse because I was running the circuit in a building. This might have been a local issue and/or I might have overreacted to some of the requirements. After getting the correct fuse I sort of stopped looking for additional problems because, well, the box I was using wasn't a listed battery enclosure and the UPS wasn't actually listed for connection to anything but their packaged expansion batteries (though at least the string I used fell within the allowable range of capacities) and so on... :)

| Plain old Square D QO circuit breakers are UL'd to | 60vdc.

As I recall, a circuit breaker with the required DC arc interrupting capacity (20kA?) and response time was even more expensive than the fuses. But since a lot of this is listing politics rather than genuine engineering it's certainly possible that a commodity device would have been acceptable and I just missed it.

| >If they have added a genuine class 2 supply requirement | >I would think that many of the existing landscape lighting transformers are | >now out of spec since even the multi-output ones typically exceeded 100VA on | >each output. It may be that there is now a specific (and different) type | >of class 2 listing for such supplies that makes this all work out. There | >is an unfortunate trend in this respect to make listings extremely application- | >specific, thus thwarting non-standard or unanticipated custom assemblies. | | Right. ABIK, it has changed and become more restrictive/ less useful. That was | my point. | | Outdoor wiring low-voltage wiring was separated in the 2002(?). I haven't | pursued what the draft 2005 has to say. but I didn't stumble across anything | that clarified how to get from one t'other or made life easier.

I just checked the Juno Flex 12 system for which I happen to have a catalog. It's one of those exposed-conductor indoor low-voltage track systems. The transformers go to 600VA per circuit, so I think even indoor systems are going to have a problem with a class 2 supply requirement. Unless those were never under 411 to begin with (like your tracks?).

| >Here's the problem I had when I looked at the whole low-voltage DC distribution | >idea a while back. Whether you classify your circuits under 720 or 725's | >class 1 you still have to use the same Chapter 3 materials and methods that | >you would use for line voltage circuits. Plus you have to keep the two | >(or maybe all three) separate. Plus you have to deal with non-standard | >(and thus more expensive) ancillary components like DC-rated switches and | >fuses. Plus you have the inherent loss disadvantages of low voltage. As | >far as I can tell, the only thing you gain under class 1 is the ability to | >use No. 16 and 18 conductors, not that I'd want to. Am I missing something? | >

| | Yes and maybe not :-) | | Yes, from the perspective that you are analyzing the benefits as if the | installation is something to be commercialized -- which it is not.

I was really trying to frame some simpler questions. What is it that you are allowed to do by remaining < 30V that you would not be allowed to do if you used, say, 48V? Similarly for using DC instead of AC.

| Yes, from social reality that in a system with multiple, distributed components | that require electrical power, the conventional approach of providing that | energy with Class 2 wallwarts quickly becomes untenable. Spousal factor -- the | answer to "Can I put three _more_ wall warts over here so that the open-close | and tilt drapery motors and the controller can be powered?" is "No". So what | to do ? One part of the answer is distributed DC power in my case.

Don't the DC/DC converters take up about as much space as the wall warts they replace? Or do you have enough consumer devices that require neither voltage conversion nor isolation nor class 2 current limitation that you come out ahead?

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani
Loading thread data ...

IOW, get back to the Q&A that might have transfer value to other folks. Thank you ;-)

Allow me to assert that, side for legacy requirements for some thermostats and HVAC controls, nearly all home automation/control/measurement electrical power needs are intrinsically DC. (Air conditioning, heating, large motors etc are beyond our scope here.)

The system I've described and shown in schematics consists in building blocks in increments of nominal 12vdc (~13.8 vdc. One can easily add additional voltages in series on _top_ of the 0-12-24vdc voltage, 80/60 amp configuration.

So I could create a 1.25 amp 48vdc supply with two 1.25amp 13.8vdc Powersonic wall-wart chargers and two 7AH sealed lead cells (the most cost-effective size) that I have on hand and move around for various projects. I have been using this particular combination from this supplier for isolated power supplies for environmental monitoring for at least 20 years.

This would provide a system with 0-12-24-36-48vdc at 80/60/1.25/1.25 amps

48vdc just doesn't happen to be needed in my case right now, which (based on the devices I have) _does_ need ~40 amps at nominal 24vdc to power the DC dimmers that I happen to use. Requirements of other folks will vary. Point is that the approach is modular, and one can move things around as needs change, which in a experimental/hobbyist environment, they inevitably do.

Recognize that if the battery bank weren't there, and aside from the specific listing issue, the 80-amp 12vdc "tap" is Class 1-compliant so one could partially "de-build" to reach a particular code goal/requirement should it come to that.

Using a 40 amp instead of 60amp charger/supply would make the 24vdc tap also Class 1 (same provisos).

No, in part because the DC-DC converters can be internal to the devices they power. Compare the hodge-potch of wall-warts, power strips, surge protectors, and local UPS's that conventionally are used to power a pc and associated accoutrements with these:

formatting link
(the 12vdc on the latter is pass-through, not regulated.
formatting link
Which, *inside* an (eg) low-power PC such as VIA EPIA is smaller, more energy efficient, has fewer connections to fail, and has longer battery backup among other things.

A standard/conventional hermaphroditic connector for DC connectors are the Anderson Powerpole series

formatting link
have been adopted by hams as informal quasi standard for Radio Amateur Civil Emergency Service (RACES) among others

Purchased inexpensively here:

formatting link
And recognize -- importantly! -- that the short back-up time on under-desk UP's don't actually solve *anything* if the PC is supposed to stay up permanently which is often the case in HA applications. So the conventional under-desk AC-based tangled jumble fails and the almost invisible, internal DC solution elegantly and dependably meets the needs.

Depending on the specific CPU requirements (part of this approach involves getting real with what we actually need in terms of computational horsepower) sufficient 5vdc and(or) 12vdc power may be available from the same DC-DC converter to supply a router or other gizmo. If not, add another DC-DC converter, perhaps within the same case, but certainly not on the wall or on an AC power strip.

I think that you are asking whether I use devices that can use unregulated nominal 12vdc or 24vdc. Yes. 12vdc devices abound. And this is a factor in my purchasing decisions. It also prompts me to take the covers off devices to see what their internal power supplies _really_ need.

.... Marc Marc_F_Hult

formatting link

Reply to
Marc F Hult

Let me tackle the 'AC' part here.

(We need to stick to comparing pomes (as in apples to pears) so large diesel generators and other solutions beyond the modest ones previously implied are beyond the scope of the solutions to be discussed here.)

I've recently discussed the aesthetic problem of "wall acne" in a thread in comp.home.automation, so won't repeat that in this thread which is cross-posted. Let it suffice that there are advantages to low voltage distribution that are unrelated to the need for uninterrupted power. It is the need for UPS that I address here.

Starting from the design objective of making power available ad infinitum, not just for a few minutes to (eg) allow a computer to shut down. This (necessarily, methinks) results in a _central_ UPS system the power from which must be distributed as either DC or AC.

The DC supplies I've described provide about (60 x 14 + 80 x 14) =~ 2000 watts . Not coincidentally this is about what 15 amp (1800 watt) and 20 amp (2400 watt) AC circuits can provide.

Noting that in the US, the National Electrical Code prohibits supplying a (eg) a conventional wall outlet with less than a 15 amp source, it follows that an AC UPS with the same capacity as the DC system I described must be on a _single_ circuit There can be mo overcurrent protection between the panel and the outlet less than 15 amps.

So in a NEC-compliant 15-20 amp distributed 120vac UPS wiring system, *every* outlet goes dead if the trip point is exceeded.

Summary:

1) After putting all the most important devices that should never be allowed to go dead, a NEC-compliant AC system is vulnerable to someone plugging in a vacuum cleaner during Saturday house cleaning, or to a space heater plugged in during an emergency and shutting everything down. This is a recipe for certain failure in my opinion. With the DC distribution system I've described, over-voltatge protection can be used _ab_ libitum_ (although not necessarily as effectively as desired -- 'nuther topic ;-)

2) Elsewhere in this thread, Dan correctly notes that the DC system described requires running new/different wires. How does having power from a single ~

2000 watt UPS system distributed throughout the house obviate the need to have new/different wires? As best I know, it doesn't.

3) From the efficiency standpoint, DC distribution also has the advantage when efficiency is most needed, namely when running from batteries. In a DC distributed system, the power used by (eg) PC goes through exactly one DC-DC conversion (battery-> DC-DC converter inside PC). In a AC distributed system, there are at least twice as many conversions with attendant decrease in overall efficiency of battery utilization (battery--> DC_AC UPS -->

AC_DC_DC converter in computer).

Marc Marc_F_Hult

formatting link

Reply to
Marc F Hult

AC

.... Marc Marc_F_Hult

formatting link

Reply to
Marc F Hult

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.