6 ms·

Remember a fully charged lithium battery with nominal 48v can be close to 60v just like 12v in your car is actually closer to 14v.

wongarsu 3y ago
However there is a lot of leeway on the "48V is the highest safe voltage" statement too. 48V has a special place in regulations because of its use in telco, but 60V DC is still very safe.
- hinkley 3y ago
  Doubling again to 96 is not safe, however. I’m not sure why they settled on powers of two. Something perhaps to do with noise filtration, and fewer new tricks to learn?
  - myself248 3y ago
    It's convenient to build things in multiples and powers of 2 and 3. A nice two-tier or four-tier battery rack full of 2-volt cells just works out nicely.
    - hinkley 3y ago
      Smaller transformers with whole numbers, right? But why 4 and not 3 or 6?
      - eternityforest 3y ago
        We seem to stick with 7-smooth numbers, and really extra prefer highly composite numbers (https://en.m.wikipedia.org/wiki/Highly_composite_number https://en.m.wikipedia.org/wiki/Highly_composite_number) for things we chose before the age of calculators. 3 and 6 were the standard voltages for consumer stuff back in the days of people needing to buy multiple packs of AA cells every year
        jjeaff 3y ago
        Are we not all still buying multiple packs of AA batteries every year? Or maybe it's just those with kids.
        eternityforest 3y ago
        I think it might mostly just be those with kids! And non techies. They always have a few disposable battery flashlights hanging around it seems like.
        myself248 3y ago
        Oh I hope nobody's doing that, rechargeables have been so good for so long. I'm just now retiring some that I bought in the 2007-2011 timeframe which have started to fail, but my 2015 purchases are still going strong with hundreds and hundreds of charge cycles on 'em. I haven't used alkalines since the 80s.
        jjeaff 3y ago
        I have not had great luck with rechargeables. They have a lower voltage than regular AA, so I find that most sensitive electronics just start out thinking the battery is already low and then stop running before the battery is fully discharged. My Z-wave keypad lock is one of those. AA batteries last about 2 months, rechargeables start giving a low battery warning after a week and hardly last 2-4.
        eternityforest 3y ago
        Have you tried lithium rechargeables with the built in converters?
        myself248 3y ago
        I return garbage products like that. Not supporting rechargeables in 1980 was sort of forgivable, but they've had 40 years to get their act together.
  - eternityforest 3y ago
    Aside from modern barrel jack supplies, none of it is actually powers of two, it's like, 13.8v lead acid or a little higher for lithium. I think we just stay close to the numbers we do, because we just really like multiples of 12 and numbers with lots of divisors. 24 hours in a day, 360 degrees, etc. Most numbers people really like a lot seem to be on the 7-smooth numbers(https://oeis.org/A002473 https://oeis.org/A002473) list and related/overlapping sets like highly composite numbers. Which I think is super cool, because it means you can choose one and there's a good chance someone else chose it too for something similar, unless they were using renard numbers or something instead.
  - nippoo 3y ago
    A lot of 48V infrastructure comes from the marine world - where 4 x 12V lead-acid batteries in series is common for this reason. So a lot of components already exist and can be made to work in the automotive sector with simple modifications. (In the marine world, it's much simpler to combine and split batteries in powers of 2 for cell balancing - you can take 4 x 12V batteries and charge them in parallel, discharge them in series)
TaylorAlexander 3y ago
Sure but they can feed the system through a regulator if they'd like. Do we have any reason to believe they are tapping directly on to a pack for this part of the system?
- SigmundA 3y ago
  Normally the battery is the regulator it serves to buffer inrush loads and absorb spikes.
  - TaylorAlexander 3y ago
    > Normally the battery is the regulator For a 12v system, of course. But on a modern 48v system who knows what they decided. Having a switching regulator with built in current limits might be helpful.
    - SigmundA 3y ago
      Its actually difficult to for a "regulator" to do what battery does, the battery can output significant surge as needed and its resistance drops as voltage rises absorbing large spikes without issue. In the Cybertruck there would need to be a DC-DC buck that could handle the 48v inrushes properly from the 800v traction battery and spikes would have no where to go as the DC-DC would not be bi-directional or even if it was could not react fast enough.
myself248 3y ago
For battery-powered systems the nominal voltage is used. Telephone "48 volts" is 55.2 volts in practice, only falling near 48 if there's a power failure and the office generators don't autostart in a timely fashion. That's never caused any regulatory problems for Ma Bell, despite OSHA saying 50v is the cutoff. And personally having spent roughly a decade of my career crawling all over such systems, 55.2 doesn't bother me one bit. Span-powered T1 at 130VDC, on the other hand.... that'll poke ya. That gets little plastic covers over all the terminals, but they have been known to fall off. So there is a meaningful threshold, and 55.2 is solidly below it. Which suggests to me that there's a good bit of leeway built into the standards, perhaps specifically so they don't have to wheedle about whether a battery system should be measured at its nominal voltage, its float voltage, its absorption voltage, its peak/equalization voltage, its....
- Tempest1981 3y ago
  > That's never caused any regulatory problems for Ma Bell, despite OSHA saying 50v is the cutoff I seem to recall getting a buzz when touching phone wires - while the line was ringing. I think I measured around 100 VAC. Apparently that's "ok", safety-wise.
  - myself248 3y ago
    Ringing is 90VAC, 20Hz, current limited. And it's only on for 2 seconds and then off for 4, so it's not like you're gonna grab on and be unable to let go. So, yeah, as a momentary thing it's not a hazard unless it tickles you off a ladder or whatever. Punchdowns and terminal blocks are normally mounted where they can be worked on from ground level so that's unlikely to be an issue.
- tzs 3y ago
  > That's never caused any regulatory problems for Ma Bell, despite OSHA saying 50v is the cutoff. I'd be surprised if the wired phone system is not allowed to do a lot of things that would not be allowed under current codes and standards. By 1920 about 35% of US households had a phone. Upgrading all of that to keep up with evolving safety standards would probably have been way too expensive, and so I would expect that there was a lot of grandfathering.
reportingsjr 3y ago
60V DC is the upper limit for circuits that untrained personnel can work on and be exposed to as listed in IEC 62368-1. This is the standard that most consumer goods and IT stuff use to validate and verify electrical safety. 60V isn't going to be hazardous to a person in basically any situation.

wongarsu 3y ago

However there is a lot of leeway on the "48V is the highest safe voltage" statement too. 48V has a special place in regulations because of its use in telco, but 60V DC is still very safe.

hinkley 3y ago
Doubling again to 96 is not safe, however. I’m not sure why they settled on powers of two. Something perhaps to do with noise filtration, and fewer new tricks to learn?
- myself248 3y ago
  It's convenient to build things in multiples and powers of 2 and 3. A nice two-tier or four-tier battery rack full of 2-volt cells just works out nicely.
  - hinkley 3y ago
    Smaller transformers with whole numbers, right? But why 4 and not 3 or 6?
    - eternityforest 3y ago
      We seem to stick with 7-smooth numbers, and really extra prefer highly composite numbers (https://en.m.wikipedia.org/wiki/Highly_composite_number https://en.m.wikipedia.org/wiki/Highly_composite_number) for things we chose before the age of calculators. 3 and 6 were the standard voltages for consumer stuff back in the days of people needing to buy multiple packs of AA cells every year
      - jjeaff 3y ago
        Are we not all still buying multiple packs of AA batteries every year? Or maybe it's just those with kids.
        eternityforest 3y ago
        I think it might mostly just be those with kids! And non techies. They always have a few disposable battery flashlights hanging around it seems like.
        myself248 3y ago
        Oh I hope nobody's doing that, rechargeables have been so good for so long. I'm just now retiring some that I bought in the 2007-2011 timeframe which have started to fail, but my 2015 purchases are still going strong with hundreds and hundreds of charge cycles on 'em. I haven't used alkalines since the 80s.
        jjeaff 3y ago
        I have not had great luck with rechargeables. They have a lower voltage than regular AA, so I find that most sensitive electronics just start out thinking the battery is already low and then stop running before the battery is fully discharged. My Z-wave keypad lock is one of those. AA batteries last about 2 months, rechargeables start giving a low battery warning after a week and hardly last 2-4.
        eternityforest 3y ago
        Have you tried lithium rechargeables with the built in converters?
        myself248 3y ago
        I return garbage products like that. Not supporting rechargeables in 1980 was sort of forgivable, but they've had 40 years to get their act together.
- eternityforest 3y ago
  Aside from modern barrel jack supplies, none of it is actually powers of two, it's like, 13.8v lead acid or a little higher for lithium. I think we just stay close to the numbers we do, because we just really like multiples of 12 and numbers with lots of divisors. 24 hours in a day, 360 degrees, etc. Most numbers people really like a lot seem to be on the 7-smooth numbers(https://oeis.org/A002473 https://oeis.org/A002473) list and related/overlapping sets like highly composite numbers. Which I think is super cool, because it means you can choose one and there's a good chance someone else chose it too for something similar, unless they were using renard numbers or something instead.
- nippoo 3y ago
  A lot of 48V infrastructure comes from the marine world - where 4 x 12V lead-acid batteries in series is common for this reason. So a lot of components already exist and can be made to work in the automotive sector with simple modifications. (In the marine world, it's much simpler to combine and split batteries in powers of 2 for cell balancing - you can take 4 x 12V batteries and charge them in parallel, discharge them in series)

TaylorAlexander 3y ago

Sure but they can feed the system through a regulator if they'd like. Do we have any reason to believe they are tapping directly on to a pack for this part of the system?

SigmundA 3y ago
Normally the battery is the regulator it serves to buffer inrush loads and absorb spikes.
- TaylorAlexander 3y ago
  > Normally the battery is the regulator For a 12v system, of course. But on a modern 48v system who knows what they decided. Having a switching regulator with built in current limits might be helpful.
  - SigmundA 3y ago
    Its actually difficult to for a "regulator" to do what battery does, the battery can output significant surge as needed and its resistance drops as voltage rises absorbing large spikes without issue. In the Cybertruck there would need to be a DC-DC buck that could handle the 48v inrushes properly from the 800v traction battery and spikes would have no where to go as the DC-DC would not be bi-directional or even if it was could not react fast enough.

myself248 3y ago

For battery-powered systems the nominal voltage is used. Telephone "48 volts" is 55.2 volts in practice, only falling near 48 if there's a power failure and the office generators don't autostart in a timely fashion. That's never caused any regulatory problems for Ma Bell, despite OSHA saying 50v is the cutoff. And personally having spent roughly a decade of my career crawling all over such systems, 55.2 doesn't bother me one bit. Span-powered T1 at 130VDC, on the other hand.... that'll poke ya. That gets little plastic covers over all the terminals, but they have been known to fall off. So there is a meaningful threshold, and 55.2 is solidly below it. Which suggests to me that there's a good bit of leeway built into the standards, perhaps specifically so they don't have to wheedle about whether a battery system should be measured at its nominal voltage, its float voltage, its absorption voltage, its peak/equalization voltage, its....

Tempest1981 3y ago
> That's never caused any regulatory problems for Ma Bell, despite OSHA saying 50v is the cutoff I seem to recall getting a buzz when touching phone wires - while the line was ringing. I think I measured around 100 VAC. Apparently that's "ok", safety-wise.
- myself248 3y ago
  Ringing is 90VAC, 20Hz, current limited. And it's only on for 2 seconds and then off for 4, so it's not like you're gonna grab on and be unable to let go. So, yeah, as a momentary thing it's not a hazard unless it tickles you off a ladder or whatever. Punchdowns and terminal blocks are normally mounted where they can be worked on from ground level so that's unlikely to be an issue.
tzs 3y ago
> That's never caused any regulatory problems for Ma Bell, despite OSHA saying 50v is the cutoff. I'd be surprised if the wired phone system is not allowed to do a lot of things that would not be allowed under current codes and standards. By 1920 about 35% of US households had a phone. Upgrading all of that to keep up with evolving safety standards would probably have been way too expensive, and so I would expect that there was a lot of grandfathering.

reportingsjr 3y ago

60V DC is the upper limit for circuits that untrained personnel can work on and be exposed to as listed in IEC 62368-1. This is the standard that most consumer goods and IT stuff use to validate and verify electrical safety. 60V isn't going to be hazardous to a person in basically any situation.