Watson Expedition

Travis Watson

Travis Watson

Member
Joined
Aug 6, 2017
Messages
60
I'm still a few weeks out from starting my build, but I figured I'd go ahead and start up a thread since I took the first $ plunge today and bought my donor batteries, and I'm about to submit purchase for the trailer, as soon as we can agree on windows. This project is now officiallygreen-lit.

I am building a custom RV using an aluminum car hauler as a base unit. The goal is to be a fully off-grid self-sustained energy setup. Current plan is:

  • Nine Panasonic N330 solar panels for 2,970 watts of solar wired in 3p3s to keep the wire runs efficient (see below for a rough initial layout mockup). 2s would have been ok (ish) too, but would have prevented me from having the 9th panel. 1s would require wire too large with losses too high imo, unless I mounted the charge controller in the ceiling, which I don't intend to do
  • Battery pack configuration of 300p13s for a total of 3,900 cells. At 2.2Ah per cell, this looks like 31,746Wh nominal. These are new cells, purchased from alarmhookup on eBay:
    View attachment 8
  • The 13s configuration is due to the DC air conditioner requirements. It has a max voltage of 54v, which would be 4.15v per pack in a 13s configuration. If I were to do 14s, the air conditioner would not run at the higher charge states
  • The least important draws shut off first: 44v for the air conditioner, 40v for the inverter, 35v for the 12v DC line for lights and so on. At 3v per cell that would be 39v, so the inverter and 12V DC should almost never shut off unless I hit low voltage protection
The solar panels will almost totally cover the roof, so I won't be able to have any skylights, roof vents, or roof mounted air conditioning. The mini-split will be mounted either on the tongue or on the bumper:

image_vhimaj.jpg
 

Attachments

  • ebay-LG-batteries.jpg
    ebay-LG-batteries.jpg
    15.4 KB · Views: 475
This is quite a nice setup with minimum number of panels used... similar as what you have in mind. Just look at the generated current from the solar controller...

 
DanLim said:
This is quite a nice setup with minimum number of panels used... similar as what you have in mind. Just look at the generated current from the solar controller...

Click to expand...

Yeah that's interesting. I haven't picked a charge controller yet, but I was hoping my choice of a 48v battery bank would simplify my setup. It seems most everything is rated in amps, and 48v will run 1/4 the amps of a 12v setup. My gut was a Victron, but I've not made up my mind yet. I'm a bit worried that company's products are overpriced.
 
TravisWatson said:
DanLim said:
This is quite a nice setup with minimum number of panels used... similar as what you have in mind. Just look at the generated current from the solar controller...

Click to expand...

Yeah that's interesting. I haven't picked a charge controller yet, but I was hoping my choice of a 48v battery bank would simplify my setup. It seems most everything is rated in amps, and 48v will run 1/4 the amps of a 12v setup. My gut was a Victron, but I've not made up my mind yet. I'm a bit worried that company's products are overpriced.
Click to expand...

Sounds good.

Re the victron, I have their 12v 800va inverter, whilst I think their products are fine I do consider them overpriced.
Have you looked at the PCM60X, I have one of these and seems solid for the money
 
The panels I have will be 2,970 watts total, which at 48v would be 61.9A, a bit over that PCM60X. Plus I'm going to shoot for a 3s3p config, which would be 210 open circuit volts coming from the panels (over the 150 rating on the PCM60X).

These high limits is why I was eyeing the SmartSolar MPPT 250/85. I think the Midnite Solar Classic could fit the bill too, maybe an Outback? Not sure... this is the area I know the least about so I still have a good chunk of research to do.

Thanks for the tip on the PCM60X!
 
TravisWatson said:
The panels I have will be 2,970 watts total, which at 48v would be 61.9A, a bit over that PCM60X. Plus I'm going to shoot for a 3s3p config, which would be 210 open circuit volts coming from the panels (over the 150 rating on the PCM60X).

These high limits is why I was eyeing the SmartSolar MPPT 250/85. I think the Midnite Solar Classic could fit the bill too, maybe an Outback? Not sure... this is the area I know the least about so I still have a good chunk of research to do.

Thanks for the tip on the PCM60X!
Click to expand...

Yes will too. Back to the drawing board
 
Based on reply on other forum post, parallel-first wiring will work well, but diodes are important. I'm not sure the best solution for this, but I have found this diode with MC4 connectors:

http://www.win-solarllc.com/index.p...151&vmcchk=1&option=com_virtuemart&Itemid=113

This seems like a veryexpensive solution. With 1 per panel, that's almost $300 just for diodes!

I'll keep searching for alternatives, but knowing there's at least this option is good.
 
You can always make your own blocks. You just need to get large diodes and connect them in parallel (say, 2 or 3 of them) to share the loads. If you get ones big enough, then only 1 would need to be used.
Here's an example of a set: https://goo.gl/LdjLqN (eBay)
And not sure where you are, but here's one from the USA: https://goo.gl/sdb7xh (eBay)
 
Korishan said:
You can always make your own blocks. You just need to get large diodes and connect them in parallel (say, 2 or 3 of them) to share the loads. If you get ones big enough, then only 1 would need to be used.
Here's an example of a set: https://goo.gl/LdjLqN (eBay)
And not sure where you are, but here's one from the USA: https://goo.gl/sdb7xh (eBay)
Click to expand...

Gotcha, yeah I think the $$ on the one I linked was for the ease factor, MC4 connectors and all ready. Just need to whip up some connector ends and housings and save a ton of $$.

If connecting this in line with the solar panels, what should the amperage rating be? The Panasonic panel I'm looking at would have a current of around 6A, so the 10A rating could handle a single panel's output. With a 3p3s configuration, would that mean there's a potential of two times the single panel amperage for feedback? Seems like I'd need a 12A+ diode rating perhaps?

This looks promising: http://www.ebay.com/itm/20A-MC4-in-...260849?hash=item33d4c8ca31:g:vvwAAOSwsg9ZkZjM

I'm dense -- the current rating is the forward current rating. Still should probably be 3x the single panel rating, or 18A (for a situation where 2 out of 3 panels are shaded, if all 3 of a parallel string are shaded, there's no saving it unless I add in bypass diodes). So 20A would work well.

At this point... it's getting complicated perhaps trying to keep it all on a single charge controller. It might still be worth it, but maaayyybe not. I should investigate options for multiple charge controllers.
 
That inline one looks nice. Definitely a lot cheaper than that other one you posted.

Personally, if I was going to go that route with diodes on the panels, I'd go full out on it. So, there'd be bypass diodes on the panels for the shaded cells, and there'd be bypass diodes between the panels. That way you could get the most out of it as possible.

However, the one thing you really need to worry about when it comes to diodes is the forward voltage drop. Each diode in series will drop the voltage a certain amount. So, if you build these, you need to check the datasheet on the diodes so you can get the least amount of drop-out.
That's when you 'really' start to see if it's really worth the time, money, effort of doing it in the first place, or just put up with a little shading for short period and let the unit bounce back. Although, you will still need a certain amount of diodes regardless. It's a balancing act when you get to this level.
 
I agree if you start working with diodes you need to matrix it full out. Otherwise no use. There are already diodes in the panels as such. Its just a matter of the matrix part.



If you just have minor shading it might not be an issue or just make sure that you design the series paralell in such way that you try to collect all shaded panels to same serie.
 
I had some time to sit down and open all the boxes and get the packs analyzed. Outside of some minor damage (that I don't care about since I'm busting the packs apart), all 200 packs (each with 20 cells) are in good shape! Only 8 packs are sitting below 3.5v per cell. One pack is at 30.2v, I think this was maybe a discharge cutoff test of the BMS that didn't get recharged?

I've attached histogram of voltage measurements of all 200 packs.

Also have a video for this update:


image_mwrqhs.jpg
 
Pretty good haul! :)
You could just strip all the cells of their wrappers and have a true tesla sytle pack; all silver ;)
It's good that the BMS actually cutoff the pack.
 
Yes! Once I realized why the oddness was occurring, it actually comforted me to see that pack voltage. The oddness was actually due to my FINGERS! In the video you can see it, my fingers slipped in and touched the probes, which caused just enough of a voltage sag to partiallytrigger the BMS protection (there's apparently a very short ramp, and my fingers touching the leads put me somewhere in the ramp, the light bulb was low enough resistance to push me all the way off).

Also, bare metal cells make me super nervous! I hope you're joking there! :p
 
hahah, not entirely joking. Just put some protective rings on the pos end and if that end is in a plastic holder, then it's even more protected from shorting out. But, I am also respectful about the possibility of shorts occurring.
 
Yeah I see your reasoning. I'm still a scaredy puss. If wraps were more expensive, perhaps, but it's kinda like cell level fuses - cheap insurance.

Bare metal batteries do look cool tho
 
Had some family issues cause pause for my build, but I'm back at it! I've broken down several of these hoverboard packs now, and am slowly working through the rest. It's taking me about 30 minutes per pack for breakdown!! Yikes. I'm hoping I can speed up, but so far every time I try to get faster, I end up causing some sparks :)

An interesting note about the hoverboard packs! They have some phantom power draw on the BMS. The packs I have are all consistently0.2v lower now than they were in September, so around 0.04v droop per month. Since they have BMS's, they shouldn't go lower than 3v per cell, but I thought this might be an interesting factoid for anyone else with these packs.

Also, I've decided to start streaming on YouTube some parts of my RVbuild (including this very long and boring process of breaking apart the cells). Feel free to drop by my channel if I'm streaming, say hi and talk about stuff. I can't do much else! Hands become occupied :-D

https://www.youtube.com/channel/UC_EiBWXUd_JM5XCVXii4uKQ
 
TravisWatson said:
Yeah I see your reasoning. I'm still a scaredy puss. If wraps were more expensive, perhaps, but it's kinda like cell level fuses - cheap insurance.

Bare metal batteries do look cool tho
Click to expand...

Clear heat shrink? ;)
Subbed to your channel :)
 
Yeah they do! Toochi carries the clear ones too ;)
 
I actually do have some clear heat shrink tubing! They look wild. Really unnatural. Like a hairless cat.
 
You must log in or register to reply here.
Back
Top