Solar to battery direct - problem

I'm asking here because there are sensible real humans who might be able to help. The rest of the internet tells me "danger, can't be done".

I have a 9series 32.4v nominal/37.8v fully charged Li battery for a small boat, I've used it with great success on the boat; a few posts about it on this forum. I'd like to use solar to charge the batteries and have searched for MPPT controllers that cover this voltage but none are available. Two flexi solar panels I have in series give (according to specs) about 40v open circuit (exactly what I measure with a wattmeter between panels and battery) and 6.3A maximum current short circuited. 6 amps is far below what the battery is capable of being charged at so extremely unlikely to overcharge, and the battery voltage is monitored, so direct connection to the battery should be possible .
Problem is, when I connect it all together in bright sunshine the wattmeter goes from 40v to 32.4v - but zero current!

The plot thickens:-
When I added another wattmeter to check that the first wattmeter wasn't faulty the first one now shows 32.4v and about 3A - perfect - and the second shows 32.4v and zero amps! I've used the second wattmeter hundreds of times and trust it.

Any ideas?

 

Not sure what exactly you are asking, but you have to have a higher voltage at the panels when under load to be able to move power into your bank and of course the bank has to be low on charge to take any power.

With regard to the MPPT, why not just use a smart 48V controller and throttle it back so the settings work well for your particular battery?

 

My suggestion would be to fully charge the battery separately with a conventional charger, which is set to not overcharge. Then remove and immediately connect to your solar panels, which should show 0 amps of charging continuing after a few minutes. That's what should happen if you want to protect your battery from overcharging.

Solar panels are notoriously sensitive to the amount of sunlight because of the series connection of individual cells. If even one individual cell is blocked from sunlight it can make a big difference in performance, when connected directly to an operating motor, without a battery buffer in the circuit. Even a fist shading a 4 ft panel can bring the motor power disproportionately way down, ie 50% shading May mean close to 0% power, instead of the expected 50%. And it's almost impossible to have identical sunlight outputs minute by minute as regards atmospheric or immediately localized sunlight illumination. That's what happens with my 20 year old panel anyway, it may not be the case for the more recently manufactured panels? I only have one solar panel, putting another one in parallel may help considerably in terms of the voltage drop.

 

Does the battery come with a built-in BMS in a marine enclosure? It may be an electronic chopper and many meters won't work on circuits with choppers. (Chopper = PWM controller). At any rate, call the battery mfg and confirm it is safe to use a direct solar connection. You run a real risk of damaging the BMS on undervoltage with some of these.

 

Solgato - I can't find a 48v charger which can be programmed down to the charge voltage of a 9S battery. Also it would be physically difficult to reconfigure to 10S (36v) for which MPPT chargers are available (but becoming rarer).

Portacruise - I think you are right, these batteries consist of 8 robot lawnmower battery packs in parallel, they hold their voltage forever, simply do not seem to degrade in storage, so I am probably trying to charge fully charged batteries, although I did leave the set-up charging for a few hours at about 3 Amps (when both wattmeters were in the circuit in series) and the finish voltage was 34.something so they did take some charge. I may also have underestimated the effect of minor shading.

Philsweet - No BMS at all, and home made Perspex and oak case. Each battery pack can be unplugged easily to check individual modules. Maximum charge rate, even on my proper charger is so far below "safe" charging that I'm not expecting problems.

 

What chemistry are these batteries?

Are you sure the packs don’t have a BMS?

What are your required charge and float voltages?

You can setup a Smart 100/20 48V Victron MPPT to 36V mode and use the Expert Mode option to bypass the preset charge profile and set up your specific charge and float voltages.

You can download the Victron Connect App and use the Demo Library option to select the Smart 100/20 48V charger to see what settings you can manipulate before purchasing if you like.

 

Solgato - batteries are lawnmower batteries that look exactly like this: Amazon.co.uk https://www.amazon.co.uk/Shentec-Battery-Compatible-Replaces-F016104299/dp/B0CHF2ZQ3C?th=1
I got them several years ago and much cheaper which is why I've ended up with an odd voltage battery. In my tests I could discharge these at 1kW for a few seconds without the voltage going below 3v/cell. I have eight of these in parallel to make my boat battery (I have two boat batteries). My 14ft rowing skiff goes nicely on 250 watts. The only management is a temperature cut-out in each module.

I think you've solved my problem as the charger mentioned is now only about £60 which is fine, last time I looked they were out of my price range, and I already have the Victron app, for a 12v/24v charger.
Thanks all, for your helpful suggestions.

 

Well the good news is according to this image, the batteries do have an internal BMS, so leaving a constant charge or discharge source on them and wiring them in parallel shouldn’t be a problem.

 

How many years ago?

 

These exact batteries, bought early 2021:
BULK Box of 8 - Genuine Samsung 32.4V 3.0Ah Li-ion Battery for Bosch Indego | eBay UK https://www.ebay.co.uk/itm/146486609676?_skw=indego+battery&itmmeta=01K2TD90J0X0XZEG9W22NNSERT&hash=item221b48470c:g:6b0AAOSw90Bn6rXs&itmprp=enc%3AAQAKAAAA0FkggFvd1GGDu0w3yXCmi1eHLKd6Gca7bMlw6AbYuEWQseGAyWdG2%2FO%2FapRNUD9K8WR6M%2FeqU7ob1E%2F9fPCClSLK%2Be5G%2BXxhoBd2%2FH1ppXn3SP4sYjbY2RXID%2B%2FK%2F9P%2Fe0V98cwGmY3kwJJ668ioDbsPsnl2Ykso865YqkacZXr2FT90B%2B5cmhdavAaRRtmRyfOC3GCSq80T4KjjgRRX2ZUQ54cHWL2YFs1gGY1i0hXl%2Bowo16k75nq%2FdNC3f48dlqag0Q%2F9y42GcEAJs2qQ%2FEI%3D%7Ctkp%3ABk9SR_qJpM2WZg
Other sellers show the internals, definitely no BMS, only a temp. switch. Picture shows 8 modules assembled into one battery:

Looks like a 19th century lead-acid! Mess of cables go via a separate connector for each module, so they can be individually checked manually, to busbars for +ve and _-ve. 9kg. 777Wh. Anyway, I think I just need to shell out for that Smart 100/20 charge controller now.

 

I would be really surprised if a company like Bosch didn’t integrate a BMS into that pack, especially given the voltage and energy density.

They have BMS’s in their batteries for other products like power tools and vacuum cleaners.

If the plug has more than 2 or 3 pins, then it is possible the BMS is external, but most manufacturers don’t design their battery systems that way as it’s more costly.

And typically a 3rd party manufacturer/supplier wouldn’t bother including a BMS if the OEM didn’t originally do so

Have you actually taken one part?

I’m always interested in identifying compact pre enclosed/housed higher voltage batteries to play around with, although in this case EBoke batteries might be better suited with regard to a more waterproof enclosure and standard voltages.