Current Flow Through Grid-Tied Outback Radian

[BackwoodsEE]

How do you treat the aluminum bus bar so it does not instantly oxidize and become a resistive connection ?

Did you include any provision for fusing the caps, or just rely on their leads ?

I'm not sure about how to treat it. Open to suggestions. I was just assuming I could Scotch-Brite the surface around the capacitor screws and use some decent-sized washers, then coat it all with something once I got everything screwed down tight. Another option would be perfboard with copper foil tape, maybe two layers of it to reduce resistance.

Fusing is done by the 80A breakers, one breaker for three parallel caps. If a cap blows, the breaker should limit fireworks. I plan on keeping everything inside a metal box. At any rate, there's plenty of headroom between the 80V cap rating and my batteries, even at cold-weather absorption voltage.

[BackwoodsEE]

A fun exercise; I think you will find, the real world is never quite as good as the model. Sometimes
the connections to a bus bar have more resistance than the bar. Make sure the filter input leads
have no common wire with the filter output leads. Can't do much about the cap resistance, except
choose a good cap. Bruce Roe

There's no output distinct from any "input." Just capacitors in parallel with the battery to provide a lower impedance path for the 120 Hz ripple current.

[Sunking]

I'm not sure about how to treat it.

I can help you out with that. Use NO-OX-ID A Special made by Samchem. This is the best anti-oxidant ever made and every utility, Telco, Water Plants, and even the San Francisco Golden Gate Bridge used it between all joints, Bolts, and Cables use. When applied correctly will last at least a Century. In electrical it is applied to the wire skinner before a compression terminal is put on a cable. Lightly buff contact surfaces with green Scotch bright and apply a thin layer or No OX ID. Coat buss bars terminal connectors with a very light coat. A little bit goes a long way. It is inexpensive and a small can or tube will last a lifetime. Be sure to coat Battery Term Post. When used properly the connection will last a lifetime and your children and grandchildren will die before it fails. Just search the internet for it. Lots of specialty electrical supply houses carry it. Strictly a professional product.

On large DC plants sparky's use a hot plate with a can to melt the grease and dip wire skinners, compression terminals, and paint it on buss bars. Just make damn sure wire wipe it off with a rag so you end up with a very thin light coat. Once coated air will never penetrate. Not even salt air and hard driving wind with rain will not penetrate of wash it off. It looks like bearing grease and about the same viscosity.

Electric Model Train enthusiast use it to coat the tracks for trouble free operation. It is good chit.

[Mike90250]

I'd pass on aluminum bus bars and use tinned copper or even bare copper. I worked with aircraft and spacecraft, with lots of problematic issues of aluminum non-conductivity, so thats why I asked. Without some sort of protective coating, the oxide film, which forms instantly after abrasive scrubbing, is thin enough in less then a couple minutes, that the coatings work. But more than a couple minutes, you have to re-clean before coating.
Aluminum Oxide is REALLY TOUGH, it's used in sandpaper, and is completely non-conductive. This was revealed with all the house fires in the 60's & 70's, before they learned to use some sort of no-ox goop on the fresh connections.


Aluminum conductive coating
https://en.wikipedia.org/wiki/MIL-DTL-5541 Type 3 for electrical conductivity

And discussion here : https://www.finishing.com/387/55.shtml
excerpt: Class 3 chromate is conductive, but requires some contact pressure. It should NOT be wiped off.

At room temperature, high-purity Al forms a 1-2 nm film of amorphous alumina in air ~instantaneously after cleaning. This film thickens to about 10 nm in a few weeks, then more slowly to ~100 nm over 5 years at ~55% relative humidity. The oxidation rate increases with humidity and is higher for alloys. In electrical connections, resistance can increase enough over time to cause heating, further oxidation and sometimes fires.

The black residue is hydrated alumina (aluminum oxyhydrate), with color due to oxidized alloying elements.

For documentation of chromate conductivity beyond that in MIL-C-5541E, see Table V of Mil-F-14072 [link by ed. to spec at TechStreet]D FINISHES FOR GROUND BASED ELECTRONIC EQUIPMENT. Electrical contact (for RF only) for Class 1A coatings may require toothed lockwashers. Class 3 electrical contacts (power & RF) do not. A footnote adds

"Touch up may be required around areas where toothed lockwashers are used. Use additional chromate conversion coating (chemical film) applied via brush to seal outside surfaces of mechanical fasteners and all abraded areas."

To summarize: cleaned aluminum has an initially very low electrical resistance which significantly increases over time, becoming insulative. Class 3 coatings maintain a low resistance electrical contact.

[BackwoodsEE]

Sunking I sure appreciate the detailed info but all that complexity plus Mike's points convinced me to go with copper. Four 12" pieces of 0.675" x 0.125" rectangular bar stock will cost me $17 (or about 1/3 of a capacitor). I think I'll use two separate bus bars for the negative of each set of three caps and connect them together with a 4" stub of 4 AWG.

Maybe I can just mount a little shelf to the wall of my equipment shed, tape the caps together, set them on it terminal-end up, and screw the bus bars on the terminals on top. I can secure them to the wooden wall with a couple of perforated metal straps. No box required.

Eighty amps is a lot of current, but I guess I'm not really worried about any of the caps exploding; I've only ever seen that happen when one had reverse voltage applied to it, and it's not all that dramatic. Besides, these big caps have vent holes to let the electrolyte spew out rather than blowing the can apart. Letting a screwdriver or wrench fall onto the busbars will be best avoided.

[Mike90250]

Even with copper (it never stays shiny) you should use some anti-ox goop so that the connections don't go bad.

Also, 4ga interconnects for apx 80A is marginal, and you may benefit if you look at flat braid (lower inductance).

If I was doing this, I''d machine the bus bars to fit directly into the inverter terminals and hang it below the inverter, with the caps mounted to the bus bar and supported by something other than their electrical screws.

And any crimped connections, done with hydraulic crimpers for best bond.

[BackwoodsEE]

Even with copper (it never stays shiny) you should use some anti-ox goop so that the connections don't go bad.

Also, 4ga interconnects for apx 80A is marginal, and you may benefit if you look at flat braid (lower inductance).

If I was doing this, I''d machine the bus bars to fit directly into the inverter terminals and hang it below the inverter, with the caps mounted to the bus bar and supported by something other than their electrical screws.

And any crimped connections, done with hydraulic crimpers for best bond.

Thanks for the input. This forum is amazing! I really appreciate the expertise from so many wise old hands here.

Yeah, 4 AWG is a bit smaller than I'd like, but the studs on the circuit breakers are 1/4" with a fairly tight squeeze. Outback provides 6 AWG wires for connections to those breakers (originally intended for PV ground-fault), so I'm at least doing better than they did.

The problem with fitting directly to inverter terminals is that the Outback GSLC is already hanging below the Outback Radian. This is the closest I can get to the inverter without messing up the rat's nest already in the GSLC.

I do have a hydraulic crimper and will be ordering crimp ring terminals from Digi-Key. I'll get me some of that anti-ox.

[Sunking]

Makes absolutely no difference if you use aluminum or copper. Use No-Ox-Id on all skinners and contact surfaces, especially around batteries. In fact in my Industry all Compression Terminals are LEAD PLATED and use No-Ox-Id. If you were to work in most Telcos and Cell Sites if you did not use No-Ox-Id, you would be escorted off the property and banned from ever working for that company again. Only smaller hack Telcos like T-Mobile and Sprint where you can get away with poor workmanship. Water ans Sewer plants are really strict. Like I said No-Ox_is is cheap and a small tub will last a lifetime.

At 80-amps 4 AWG is just fine for short distances. Really the issue with conductors larger than 6 AWG is terminating conductors. DIY just do not have the proper tooling and training to terminate connectors properly, nor can they afford the tooling. Even the Compression Connectors are expensive. People just do not understand the importance of proper termination. Not only does it affect performance, but critical to life safety. Do battery work for a company like ATT or Verizon and they use DLRO meters to measure contact resistance and termination resistances . 25 micro-ohms or less takes a lot of care and preparation. I have built a many battery plants that use as many as 3 750 MCM in parallel to a battery term post to Overhead buss as some of those plants are rated at 10,000 amps 48 volt DC

[BackwoodsEE]

Makes absolutely no difference if you use aluminum or copper. Use No-Ox-Id on all skinners and contact surfaces, especially around batteries.

Let me make sure I understand you right: When I strip the wire, I should wipe a film of this stuff onto the bare copper and also onto the inside of the crimp terminal, and then stick the crimp terminal on and compress it with my hydraulic crimper? Can I use a lint-free cloth (like the ones the store gives you with your new eyeglasses to wipe them with) to do that?

For the copper busbars, I assume I wipe the No-Ox-Id on the area of the copper around the hole I'm going to drill for the bolt, and also on the underside of the hex-head stainless bolt, and also both sides of the ring terminal?

I foolishly didn't use this stuff with assembling my battery bank. So, for that, just remove each M10 bolt, coat the ring terminals and the top of the battery button terminal, re-torque the bolt, and coat the exterior of the whole bolt+terminal for good measure?

I have built a many battery plants that use as many as 3 750 MCM in parallel to a battery term post to Overhead buss as some of those plants are rated at 10,000 amps 48 volt DC

Wow, that is a lot of I^2 R to avoid! Impressive. You can see why the power companies prefer to switch the V and the I around in those figures! Insulation and insulators are cheaper than copper.

[Sunking]

Let me make sure I understand you right: When I strip the wire, I should wipe a film of this stuff onto the bare copper and also onto the inside of the crimp terminal, and then stick the crimp terminal on and compress it with my hydraulic crimper? Can I use a lint-free cloth (like the ones the store gives you with your new eyeglasses to wipe them with) to do that?

Yes you are on track. So important caveats. When you cut the wire to length, you want a good flush cut clean cut on the conductors. Three good ways to do this. For lack of a better words and what sparkies call them is to use Lobster Claw type of wire cutters, they shear and do not crush. Unfortunately those are only good up to 6 AWG. Similar is a ratcheting type good up to 4/0 but difficult to get flush. Lastly my favorite and preferred method is Band Saw.

Bes sure when you skin the wire it completely fills the barrel. You know that hole in the end of the barrels? It is called the inspection hole so inspectors to make he can see the conductors completely filling the connector. Now Terminals made for Battery Term post do not have the Inspection Port as that would allow battery gasses to enter, and they are lead plated.

Just a light coat and I have sen a lot of things used to wipe the grease down. Today a good microfiber towel you would find in the auto detailing section. Large jobs sparkies just hot dip the wire skinners and sling off the excess.

[QUOTE=BackwoodsEE;n368692] For the copper busbars, I assume I wipe the No-Ox-Id on the area of the copper around the hole I'm going to drill for the bolt, and also on the underside of the hex-head stainless bolt, and also both sides of the ring terminal? Yep all exposed surfaces and the hardware. Best practice is too use terminals with TWO holes to maximize contact area wiht locking hardware. For large batteries they are 4-hole terminals using 5/8 hardware. Otherwise 1/4-20 for 6 AWG and larger. Even the intercell bus connectors get treated if you use buss bars to connect inter cells.

Wow, that is a lot of I^2 R to avoid! Impressive. You can see why the power companies prefer to switch the V and the I around in those figures! Insulation and insulators are cheaper than copper.

Large plants have large demands and we get really strict with voltage drop. Telco equipment must meet Bellcore Standards which includes voltage operating ranges. -48 VDC equipment is required to work from -62 down to -40 volts. Its the 40 volt at full rated current is the biggie. A FLA or SLA battery is discharged at 1.75 vpc so on a 48 volt plant is 42 volts. That only leaves you 2 volts to play with for total voltage loss. That includes everything. Batteries to Charge/Discharge Buss (.25 volt), Battery Frame Distribution to Power Distribution Frame in the equipment room which can be as much as 100 feet away from the battery room in a large office (.25 volt). Lastly from Power Distribution frame to utilization equipment which can be as far as 50 feet (1.5 volts). 2-volts on a 48 volt plant is roughly 4%. Even a small cell site plant is rated 600 and 1200 amps. Every micro-ohm counts.

Yeah that is why utilities run as much as 1,000,000 volts on transmission. Secondary distribution behind your house uses 13.2 KV and 4160 stepped down to 240, 480, and 600 volts to the user. Max current on a Transmission line can run as high as 1000 amps, typical is 400 to 600.

[bcroe]

There's no output distinct from any "input." Just capacitors in parallel with the battery to provide a lower
impedance path for the 120 Hz ripple current.

I would suggest you use this physical arrangement to keep ripple out of the battery if
practical, for best results. Other stuff can connect in anywhere. Bruce Roe



RipFilt.jpg