Moving Bat+ bus to inverter side of main DC breaker, FlexNet DC mounting, etc.

[jnh]

I just unpacked an e-Panel today (MNE250ST-L), and am planning on moving my Outback system into it over the next week or so to clean up a messy installation, bring it into NEC compliance and get all the equipment & batteries out of my crowded home office area.  This looks like a well-designed panel.  I have just a few of questions about it:

1.  I'm a little nervous about that unfused Bat+ busbar  to one side being so close to the grounded enclosure, thinking a slight slip of the screwdriver could have dire consequences if I ever tried to work on the panel without first disconnecting a battery cable.  Is there any good reason not to move its 4AWG supply wire from the bottom to the top lug of the main 250A DC breaker, so that throwing the big breaker will also drop power to all smaller DC loads and charge sources?  I'm OK with not being able to power these while the inverter is shut down.  Would this wiring change violate code in any way?  Each smaller DC device would still be protected by an appropriately sized breaker, of course.

2.  Right now, in lieu of a main breaker I have a 250A Class-T fuse holder in the battery positive cable, plus a separate disconnect switch.  What are the merits of retaining this fuse inside my battery box for additional protection?  Unless there's a code issue with doing so, I guess the only real drawback would be a slight additional voltage drop?  Has anyone tried to quantify resulting losses?

2.  Regarding polarity of DC breakers, I have always understood the LINE (or +) should go to the battery if either pole does, but the main 250A Carling breaker in this panel has been pre-wired the opposite way around.  Is polarity not of concern on this breaker?  What about the smaller MNPV and MNDC type?

4.  I'd like to place an Outback FlexNet-DC in one of the three panel-mount breaker slots at top right.  Has anyone else tried this?  Given an FNDC takes up more room than a normal breaker, does it obstruct those two topside flex-conduit channels from the door too much?  My e-Panel is the original, narrow size.  Would proximity to AC wiring from the bypass breakers degrade measurement precision on FNDC's shunt inputs?

5.  Squeezing the Outback Hub4 inside as well looks impractical (maybe if this was a "stretched" panel...), so I'll probably end up mounting it above, leaving room for a couple of MNSPD's that are still on their way, or else on the opposite side of my charge controller.  What is the recommended way of getting Cat5 cables in and out?  Plastic flex conduit?  Would code allow a standard NM-type clamp to be used?  How about the watertight nylon clamps used on solar panel junction boxes?  I have a lot of those on hand.

Thanks in advance for any help.

[Robin]

Lots of good questions.
1. You can move the 4AWG wire to the top of the 250 amp breaker. We place it at the bottom as you realized so turning off the inverter does not turn off the charger. No code issues in either place. I have wanted to make a plastic boot for that bottom battery cable connection for a long time. We did get the gaurds for the terminal busbars. Next will be that battery connection.
2. Use the ClassT fuse in the battery box. There are no code issues. There is an upcoming requirement to have over current protection in all strings of the battery bank when there are 3 strings or more. We are working on adding breakers in our battery boxes for this purpose.
3. The panel mount breakers are from Carling. They are marked line and load. That is an AC description. There is no polarity for DC. I don't understand it but I have had many conversations with Carling regarding DC polarity and they assure me there is no polarity issue.
That holds true for the big 250 amp as well as for the C series 3/4" wide breakers. We stock from 1 amp to 100 amps in the C series.
The MNEPV din rail breakers do have a polarity. The + on the bottom side of the breaker means it wants to be hooked up to the highest potential. That will be the PV+ in a combiner or to the battery + when connected to the output of the controller. Now you may think that the output of the controller is at a higher potential than the battery plus terminal. Wrong! When a controller fails it shorts battery + to battery -. The battery+ terminal is the real highest potential when the controller fails. That condition is when you need the breaker to trip. I have found that unless you are at the top end of the voltage rating and at very high current, these din rail breakers trip just fine even hooked up backwards. Do it right though. It doesn't cost any more.

4. I don't know if the flex conduits obstruct things when the Flexnet is installed. Try it. It it doesn't work, we will be glad to send you a new side plate or a breaker to fill up the hole. I do know it fits fine on the wider E-PAnels. I am at home at the moment and don't have an E-Panel to go look at.

5. The network cable from OutBack should go down one of the flex conduits into the E-PAnel and out underneath the Mate mounting bracket. We supply a splined snap in grommet for this. You are correct that the Hub fits inside the E-PAnel just fine on the wide E-Panel. I wouldn't put it inside the E-Panel though. I designed in those green tweenkie lights on the Hub just to let you know the Hub was working. Its hard to come up with moving parts on electronics. That was the closest I could do with the OutBack system back then. Can you guess where the idea for the disco lights came from inside the Classic?
You can use plastic conduit for the Hub wiring, but there are no conduit hook ups on the hub. Your call. The green Cat5 wire is low voltage, limited energy and do not have to be run inside of conduit.

[Halfcrazy]

I want to post a HUGE WARNING about putting the bat + buss bar on the switched side of the 250 amp breaker. DO NOT do this. Here is the issue, say you have a wind turbine, Hydro or solar hooked to it and you forget to turn it off first and you flip the 250 off. KABOOM goes the inverter from over voltage. I have personally seen a dear friend blow up two outback inverters and 2MX60's this way.

If you want it switched then by all means add a breaker to it but do not EVER make it so the charging sources can feed the inverter with no batteries.

Ryan

[TomW]

I want to post a HUGE WARNING about putting the bat + buss bar on the switched side of the 250 amp breaker. DO NOT do this. Here is the issue, say you have a wind turbine, Hydro or solar hooked to it and you forget to turn it off first and you flip the 250 off. KABOOM goes the inverter from over voltage. I have personally seen a dear friend blow up two outback inverters and 2MX60's this way.

If you want it switched then by all means add a breaker to it but do not EVER make it so the charging sources can feed the inverter with no batteries.

Ryan

Ryan;

Exactly my concern on first reading this.

Been there, done that just not with an E-Panel.

24 volt inverters don't much like 150 volts from a freewheeling turbine. The turbine is not too happy with no battery load holding it back, either.

Tom

[mtdoc]

I want to post a HUGE WARNING about putting the bat + buss bar on the switched side of the 250 amp breaker. DO NOT do this. Here is the issue, say you have a wind turbine, Hydro or solar hooked to it and you forget to turn it off first and you flip the 250 off. KABOOM goes the inverter from over voltage. I have personally seen a dear friend blow up two outback inverters and 2MX60's this way.

If you want it switched then by all means add a breaker to it but do not EVER make it so the charging sources can feed the inverter with no batteries.

Ryan

Ryan;

Exactly my concern on first reading this.

Been there, done that just not with an E-Panel.

24 volt inverters don't much like 150 volts from a freewheeling turbine. The turbine is not too happy with no battery load holding it back, either.

Tom

Not sure I understand this.  The inverter would see the charging voltage as controlled by the charge controller which is still on (briefly). Why would it see 150 volts or an unloaded turbine?

[Halfcrazy]

If you hook the turbine or controller to the battery plus bus and it is hooked to the inverter side of the battery breaker. The inverter presents no load to the turbine or controller so the voltage will spike and let all the expensive cool smoke out of the inverters.

The best way to set a system up is so the Inputs go to the battery and the inverter comes from the battery. So in basic terms no switch or fuse should be allowed to disconnect the battery from Input and Inverter and yet leave them hooked to each other.

Ryan

[TomW]

Not sure I understand this.  The inverter would see the charging voltage as controlled by the charge controller which is still on (briefly). Why would it see 150 volts or an unloaded turbine?

Maybe not an issue now but future upgrades may connect a source to that bus.

That is the concern as I see it. Maybe not understanding what you are trying to describe?

Tom

[mtdoc]

If you hook the turbine or controller to the battery plus bus and it is hooked to the inverter side of the battery breaker. The inverter presents no load to the turbine or controller so the voltage will spike and let all the expensive cool smoke out of the inverters.

The best way to set a system up is so the Inputs go to the battery and the inverter comes from the battery. So in basic terms no switch or fuse should be allowed to disconnect the battery from Input and Inverter and yet leave them hooked to each other.

Ryan

OK so if I'm understanding correctly,  there is a brief instance before the controller shuts off due to no battery input where is still can output to the inverter and since it has no battery load for that brief instance it can spike the voltage output beyond it's charge parameters.  Is that right?

Sorry if I'm being dense - just trying to learn these subtleties.   

I've also been concerned about the hot battery positive bus bar in my ePanel when wanting to do some quick work and not having the patience to unhook power at the battery terminals.

My ePanel was a earlier version - before the busbar protectors were added.  Is there someplace to purchase these as an add on?  I haven't seen them listed for sale at the usual places.

I've been thinking about using a panel mount breaker between the busbar and the  battery positive coming off of the inverter breaker terminal (at its present location). Maybe with the busbar protectors that would be unnecessary?

Thanks.

[Halfcrazy]

Well some controllers are better than others

There are some other charge controllers out there that regulate slowly and will spike the battery side for a second or two before catching themselves. If they catch themselves.

The other issue is Wind or Hydro that is direct connect to the battery bus.

Ryan

[mtdoc]

The other issue is Wind or Hydro that is direct connect to the battery bus.

Ryan

OK got it. Diversion controller instead of a using a Classic. Now why would anyone want to do that!!

What about getting the bus protectors? Can they be bought separately?

[jnh]

Thanks for all the feedback.  My system's actually been wired the way I proposed since the beginning (main disconnect between batteries and everything else), and on at least one occasion I did accidentally disconnect my battery bank while leaving the inverter, DC loads and solar charge controller tied together.  There were two sets of batteries connected at that time with a Perko marine-style A/B/Both/Off switch in-line, which was inadvertently set to Off in mid-day.  The system continued to work for a minute or so in "Sell" mode, until DC supply and consumption drifted out of balance and the PV controller shut down with an error specific to this situation-- I think something like "Load Rejection" or "Unloaded."  I didn't realize what was happening right away or I would have immediately stopped that, but at least no magic smoke leaked out.  I was surprised it kept going so long with no battery.

The controller is an Outback FM-60 (camouflaged in an MX-60 chassis; Outback swapped in their latest Flexmax boards while servicing the MX to help a problem I was having).  Arrays supplying it are wired for low voltage, only 36V nominal, and depending on temperature usually settle around 45Vmp, 60Voc.  The highest Voc logged over several years has been 66V, and the FM's lifetime "Max Bat" voltage is 33.3V... possibly from the battery-drop incident.  I can't remember if that happened before or after the old MX was rebuilt.  I guess the worst-case failure mode would be for the FM-60 to cut PV voltage straight through to the inverter while the battery breaker was off-- could the GVFX3524 withstand 66V from an unloaded array for a few seconds without permanent damage?

I don't have any wind or hydro generation to worry about for voltage spikes, and (sadly) never will at this location because the site just isn't suitable.  It's marginal even for solar, with so many tall trees around.  I have to move the arrays around several times per year to keep them in a sunny spot, and still at least 1/3rd of my potential daily energy harvest to partial shading.

Besides PV, the only other 24V power sources connected are a couple of AC->DC battery chargers that I use to charge from small generators (or, rarely fom the grid when I want to run in online-UPS mode for a while).  One of these is an Iota, the other Meanwell, and under testing I've found neither will exceed 29V or so even when suddenly unloaded, or powering a small DC load directly with no battery in between.  So, they should be safe as well.

I've considered building one of those improvised alternator+lawnmower-engine DC gensets for faster bulk-phase charging during a long outage, and if I ever do will certainly make sure that gets lugged to the unswitched side of the main inverter breaker, since the voltage regulation is likely to be poor.

On the blinkenlights Robin put in the Outback Hub, for this and other reasons I had mixed feelings about burying the thing inside a panel even if it would fit.  At one time my Mate was getting into a "stuck" state once in a while where it stopped communicating with the FX/MX or logging data via serial after being left in certain submenus for too long, and the frozen pattern of Hub LEDs was a good way to tell when this had happened (going back to its summary screen always cleared it).  The Mate's been upgraded since, and that bug probably long since fixed, but it's still nice to have a simple indication the comm links are still working.  Another advantage of keeping it outside is easily-accessible jacks for moving the Mate back and forth from its usual indoor mount out to the the E-Panel area during troubleshooting... since Outback's "2nd Mate" feature never got implemented.

It looks like the Flexnet will fit in an E-Panel's frontmost panel breaker slot, leaving the rear two free in case a DC-GFP ever has to be added... for now those thankfully aren't required in my area for ground-mounted PV, and seem to be of little or no safety value on a 36V array.  Not much space is left behind the rightmost AC-side flex conduit, but thin control cables are coming down through that (Cat5, BTS, AUX relay out) should be easy to route around the obstruction.  AC power wires on the other conduit will touch the Flexnet's attached Cat5 cable, unless I use an adhesive tie-wrap or similar on the inner door surface to pull them aside.  It's a tight squeeze, but seems workable.

[Halfcrazy]

Yes a grid tie inverter would probably survive as its goal is to hold the battery voltage down. I was thinking more along the lines of an offgrid inverter. If that had been an offgrid system you very well may have been looking for some new boards for the inverter.

Ryan