meaning of -48V (minus 48?) output, Earth ground vs Return issues

Oct 24, 2010
2,405
Hunter 30 Everett, WA
Thanks, that is exactly my concern. At high currents I bet 48V could cause a health issue in the wrong circumstances, especially at such high current rates.
I could enclose the case in neoprene, put huge safety-orange Shock Danger stickers?

That is, if there is any downside to my isolating the rectifiers from the outer chassis.

Is there?
Yes even 12 volts can cause a very nasty shock. For example leaving your sweaty body across the large chrome bumper and grabbing a socket wrench and putting it against the battery terminal. Oh yes this is the voice of experience. Now and you multiply that voltage by 4 to 48 volts you just multiplied the current that would run through your body by 16. Very dangerous indeed, these can be lethal shocks. The big concern is to isolate all chassis surfaces. Remember the chassis of all your marine electronics will always be at the negative ground potential. If you can touch both chassis without isolation you get zapped.

I still haven't heard about conformal coating on circuit boards (this is a saltwater longevity issue) stranded wire (vibration issue) or nickel plated wire (another corrosion issue).

Ken
 
Jan 7, 2017
92
beneteau 36 new london
Is it all built to marine standards?
This will not be "installed" as such, but mounted in an airtight Pelican style case, only opened while charging maybe an hour or two.

But mfg states designed for remote telecom station usage.

All wiring infrastructure will be to high standards, marine / milspec.
 
Jul 23, 2009
926
Beneteau 31 Oceanis Grand Lake, Oklahoma
I'm a power tech for a telcom. I work with AC & DC power almost every day, diesel generators and large batteries too. I'm very interested in solar as well. I just don't see a good reason for feeding a boat with rectifiers when normal shore power is available. Sure it is possible to do it safely, I just can't justify the cost and complexity.

I handle the 48 volt buss bars all the time, the only time I ever get shocked is when I'm sweaty and I'm leaning on something grounded, BTW almost everything is grounded in a telephone office.

Why do you what to do this?
 
Jan 7, 2017
92
beneteau 36 new london
Not sure if the chassis helps with dissipation for the PSU. If so, a coating or cover would impede that.
Each internal module is designed with powerful active fan cooling, designed to be inserted into OEM rack server cases etc.

But def an issue to be aware of.
, tx
 
Jan 7, 2017
92
beneteau 36 new london
Last edited:
Jan 7, 2017
92
beneteau 36 new london
The question of using telecom-standard " -48Vdc " output (note the negative,) seems to be in itself, no longer an issue. As long as the input and output sides are isolated, no connection between them, in other words output is floating wrt input, then

I can simply connect the (negative voltage) energized conductor output line of the PSU to the negative return/common "ground" input line of the load device,

and connect the "ground" return/common output line of the PSU to the positive input line of the load device.

Thanks to Ken Cross, uncledom, JohnShannon for helping me get that far.

So now it seems the critical next step is the question:

By default each modular rectifiers shares a common reference with the enclosure's outer chassis.

Obviously I can't leave it exposed!

So the choice is, either A. further enclosing or insulating that chassis, or

B. isolating it from the modules within, which the vendor says he'll walk me through, and can be verified with a DMM.

At this point B seems the safer route. Yes? So far I know of no downsides.
 
Jan 7, 2017
92
beneteau 36 new london
Before cellular I had a mobile phone company. We had telco (-48V) interfaced with radio transmitters (+12 and 24V). I can tell you we had to take great pains to keep the 2 systems isolated. Also the telco rack frames were referenced to the radio system grounds. The telco "grounds" were different from the actual site grounds. Short answer is this will be a really complex undertaking.
I'm a power tech for a telcom.

I handle the 48 volt buss bars all the time, the only time I ever get shocked is when I'm sweaty and I'm leaning on something grounded, BTW almost everything is grounded in a telephone office.
Wow, thanks much to both of you for your replies!

Unfortunately, for my situation true Earth Grounding will rarely be available, which in S&B building may be an unorthodox challenge, but my understanding common on boats. The whole galvanic corrosion (not! "electrolysis" I've learned) issue is related but maybe not directly?
 
Jan 7, 2017
92
beneteau 36 new london
I would assume that the equipment case is connected to the supply ground.
I don't know if the connection between the chassis and rectifiers is to their AC input or DC output side "ground". Seems from others' statements the latter is more likely?

The question at hand is whether or not to isolate that chassis completely I think, so if Yes, the above becomes moot.

> In other words, the power output is at -48V referenced to the case.

More precisely and generally, the energized DC output lines referenced to its negative return line, only.

> I believe that would cause problems if you connected the case to boat ground for safety reasons

Yes, I believe either way (A or B) that will be prevented.

> Alternatively, if you connected the power output to boat ground

I believe that would be the case

> and used the supply ground as positive 48V

and this would also be true

> the case would also be at 48V relative to the boat ground.

Not with option B, where the case gets isolated.

The other option A, is to fully insulate it, put big warning decals all around it.
 
Last edited:
Oct 24, 2010
2,405
Hunter 30 Everett, WA
This will not be "installed" as such, but mounted in an airtight Pelican style case, only opened while charging maybe an hour or two.

But mfg states designed for remote telecom station usage.

All wiring infrastructure will be to high standards, marine / milspec.
It sounds like this is thought out so far. Just keep your diligence for safety, reliability and maintainability.

Ken
 
Jan 7, 2017
92
beneteau 36 new london
Thanks Ken. Obviously that A vs B choice is still open in my mind, so if at some point you feel like weighing in, I'd greatly appreciate it.
 

SG

.
Feb 11, 2017
1,670
J/Boat J/160 Annapolis
I'm confused about how this system is supposed to work (on a boat)?

What sized battery bank of what kins of batteries, etc?
 
Sep 15, 2013
708
Catalina 270 Baltimore
OK. I'll bite again. The only way I can see this happening is with this scenario.
The 48 volt supply is driven by AC dock power. The dock (earth) ground is attached to the AC side of the PS for safety reasons.
There can be no reference to earth ground on the 48VDC output side. You will need a schematic and a statement from the manufacturer to confirm that.
Reference the negative output wire to DC ground (yellow) and the positive output wire to positive (maybe red but I am sure ABYC will take issue). Now you will have 48 volts DC negative ground presented inside your boat.
My question now is "what in the world are you going to do with 48VDC on your boat?" It can't be reduced via a transformer. A DC to DC converter that would handle any kind of current reliably would be really expensive and would add an unacceptable level of complexity in a marine environment. Even if you are planning to use one of those cheap golf cart devices I still don't see the sense in it. This advice is worth as much as you paid for it. Best of luck.
 
Feb 6, 1998
11,756
Canadian Sailcraft 36T Casco Bay, ME
Telecom equipment is normally a "positive ground" system where the + battery terminal is at Earth potential.. It is a left over from many, many moons ago when a negative grounded system was leading to corrosion issues in high moisture environments, so they flipped it. It is still in wide use today. A boat is no place to try and mix a positive and a negative grounded system.
 

SG

.
Feb 11, 2017
1,670
J/Boat J/160 Annapolis
JohnCT61
I do appreciate any such feedback there, but would like to keep this thread specific to the current open question regarding the PSU case: A insulate vs B isolate .

The open questions in this thread -- which remain, if answerable:
  • In what circumstance would someone do this?
  • Why?
  • How would you practically do it? (Compared, by inference with what one would otherwise do.)
You're the only one that can shed light on that.
 
Jan 7, 2017
92
beneteau 36 new london
The use case is I think well laid out in the other more general thread, so constructive discussion most welcome to continue there.

The question of how to deal with No Earth Grounding remains, and must be dealt with every day, on what nearly every boat running from genset?

But perhaps I'll just need to learn more elsewhere. . .
 
Jan 7, 2017
92
beneteau 36 new london
FACP systems have sent DC control/operating voltages in the midst of other AC and RF since before we were born.
I see that means Fire Alarm Control Panel systems, and I've read a little on GFCI, RCD and RCCBRCCB devices.

Not sure how to apply those two general topics with my specific use case, but definitely want to climb that learning curve.

Thanks for the feedback.