If the voltages of the internal charger is adjustable, what you can do is set it to something low. Then it sommer acts as battery maintenance too. Eg, on a 24V system you might set the internal charger's float voltage to 26.5, and the external one to 27.6. That way the internal one should shut down completely during the day, and at night it will "maintain" te batteries. Not perfect, but an option if you want to keep the batteries charged to a minimum SoC at all times.

 

• I seem to understand that the Axpert inverters will always use Eskom to charge the batteries. You cannot make them only use solar only to charge  the batteries. Why this came up was that if you invest in solar panels, you need to get a ROI from them and using Eskom to charge batteries, is defeating that purpose.

 

Not sure if this is true

 

edit

Checked the manual

2 options on the 5KVA

1.Solar energy and utility will charge battery at the same time.

2. Solar energy will be the only charger source no matter utility is available or not.

If the voltages of the internal charger is adjustable, what you can do is set it to something low. Then it sommer acts as battery maintenance too. Eg, on a 24V system you might set the internal charger's float voltage to 26.5, and the external one to 27.6. That way the internal one should shut down completely during the day, and at night it will "maintain" te batteries. Not perfect, but an option if you want to keep the batteries charged to a minimum SoC at all times.

 

That might be a way of temporarily "disabling" the internal charger.

2. Solar energy will be the only charger source no matter utility is available or not.

 

.AHA ... The 800w range does not have that feature. To stop if from charging using Eksdom, you need to unplug it from the wall socket.

Once a week, because my friend only has 1 x 255w panel, he plugs it back into Eskom to top up. 

 

 

I just wish the extra interconnecting stuff didn't cost as much as it does...

 

 

I have a USB MKII cable for my inverter and a USB VE.Direct cable for reading the values from the BMV702 and MPPT Charge Controller.

Would it be possible to reverse engineer them?

Would it be possible to reverse engineer them?

I have a USB MKII cable for my inverter and a USB VE.Direct cable for reading the values from the BMV702 and MPPT Charge Controller.

 

I wouldn't bother with the MKII. The mere fact that it has firmware that is periodically updated when you start VEConfigure indicates that there is a microcontroller in there, and I'm willing to bet that most of the magic is in the source code :-) There might be some interfacing comms chip in there, and that might be interesting to look at, but it's probably more effort than it's worth.

The VE.Direct, as far as I know, is just a serial cable with some added galvanic isolation. If you know the baud rate, and you know which pins is RX/TX, this would be relatively simple to reverse engineer. Still... that VE.Direct cable is a thousand bucks... which is even more than I paid for my MKII. Similarly, the CANbus on the 150/70 and 150/85 is standard enough that you can use other equipment to read it. As I understand, they use NMEA2000 (does anybody REALLY spend thousands on a spec document, just for the PRIVILEGE... that's insane), which is in turn layered on top of J1939 (that at least is more accessible). Now as I understand it it is not completely unworkable, at least one French guy on hackaday.io made his own system that reads the CANBus... however...

The issue that's really at stake here is that I think most of us understand that if you spend 30k on an inverter, you really rather should spend the 4k on the genuine cable (with the proper galvanic isolation). But it all adds up, and at least one local installer has expressed dismay as to how difficult it is to sell a Victron system to someone now that these Axpert's are out there. Sure, the Victron is the Rolls Royce and the Axpert is more like a Ford Figo, but seriously... 8k vs around 25k for the inverter, and then you still have to add on the mppt controller (around 5k for a 100/50, that's if you want something better than the axpert) and the VE.Direct cable (another 1k, just to add insult to injury)... well I understand why the Axpert/RCT/MustUPS is so darn popular.

On the other hand... I've been messing around with some of the tech on my own, and you know... just an ACS712 current sensor is around R80, a cheap microcontroller is around R50, an inductor for your buck converter would probably set you back around R150, MOSFET... if you want a nice high open circuit voltage is going to set you back R50 or more... then there's the Schottky diodes which by their very nature don't like high reverse voltages, so they cost more if you want that... so I pretty much see why an MPPT controller at under 2k is pretty much a bargain :-)

:-)

I knocked together a ve.direct cable yesterday it's a serial cable with an FTDI serial to USB converter very easy to make and cheap. Probably cost me R100?

I knocked together a ve.direct cable yesterday it's a serial cable with an FTDI serial to USB converter very easy to make and cheap. Probably cost me R100?

 

I've also googled a bit just now and came across a few posts, but none mentioned any detail except for pinouts when connecting it to a PI. I've built a USB to DMX "cable" a while back and used a FT232, but I've also added an ISO7220 for galvanic isolation. I think it would maybe be a good idea to do the same in this instance. Which FTDI chip did you use?

Good for you John. I did same, or shall I rather say, I had one made. I tend to let a lot of smoke out. SWAMBO had to intervene.

 

What pops my doppie is WHY does Victron not make all the units CAT5 accessible!? 

Serial, who still uses serial ports today? USB is fine but distance.

 

I offered them once to contract for them to do it. 

 

They seem to be focusing on the high quality parts, not the ease of access to their devices. IMEON and the rest are kicking their arses good and solid on ease of accessibility.

 

 

 

... just an ACS712 current sensor is around R80, a cheap microcontroller is around R50, an inductor for your buck converter would probably set you back around R150, MOSFET... if you want a nice high open circuit voltage is going to set you back R50 or more... then there's the Schottky diodes which by their very nature don't like high reverse voltages, so they cost more if you want that... so I pretty much see why an MPPT controller at under 2k is pretty much a bargain :-)

I think there is a wee bit of software somewhere involved, to calculate the best sun input?

 

I agree, Victron's are the Rolls Royces. I think the proof will be in how long the hybrids last. Victron has been in this game for a very long time, starting back then in specializing on electrifying ships. So I am hoping on 15-20 years for the inverter, and then be repairable.

 

I think there is a wee bit of software somewhere involved, to calculate the best sun input?

-- snip - -

 

I agree, Victron's are the Rolls Royces. I think the proof will be in how long the hybrids last. Victron has been in this game for a very long time, starting back then in specializing on electrifying ships. So I am hoping on 15-20 years for the inverter, and then be repairable.

 

I've been reading and watching a lot of youtube on this topic :-) There's two ways to do MPPT in software.

The one is to do a period sweep across the whole voltage range, and inbetween you do "perturb and observe", that is, you tweak your mark/space ratio on the PWM a bit and check if the power goes up/comes down, if it goes the wrong way, you turn around and go the other way. By repeatedly doing that you eventually end up on the MPP, basically hovering just before and just after it.

The second is more complex. Incremental conductance. Far as I understand, this attempts to predict the shape of the power curve... which if you think about it, and if you've had some training (or an applied mathematics class at uni), is almost obvious: We know more or less what the power curve of a solar panel looks like, so with a little computation, and three or more data points, you can fit a function through your data points and use it to predict the MPP.

I suspect the Victron controllers use incremental conductance. Their sales talk (they say it is fast and gives you 10% more than other MPPT controllers) sound a lot like the literature about incremental conductance. I know the Microcare uses Perturb and Observe. Most controllers probably use POA.

 

The software involved in Perturb and Observe isn't too complex really (assuming you are at least skilled enough to write embedded software). The other one is apparently a bit more work, or that is what I heard. I suspect it's more work because you need to understand calculus... :-)

@superdiy I used a damaged x bee explorer usb from a friend. Just checked price has tripled on netram.co.za since I last bought em. It just converts USB 5v to 3.3 volt that the bmv uses and creates a virtual serial port . I soldered a 2.5 pitch connector for the bmv and connected ground and to rx and tx. (obviously switching them). It uses an ft231x I think. It then just transmits every second. The grand plan was to use this with a serial board for the axpert on a pi to monitor and control the system. USB would be great for the axpert but there is no api.

@superdiy I used a damaged x bee explorer usb from a friend. Just checked price has tripled on netram.co.za since I last bought em. It just converts USB 5v to 3.3 volt that the bmv uses and creates a virtual serial port . I soldered a 2.5 pitch connector for the bmv and connected ground and to rx and tx. (obviously switching them). It uses an ft231x I think. It then just transmits every second. The grand plan was to use this with a serial board for the axpert on a pi to monitor and control the system. USB would be great for the axpert but there is no api.

 

Thanks John, when I (one day) develop the requirement to log the BM, I'll experiment a bit more. 

Hi guys

 

I just had a Victron BMV-702 installed on my battery bank. I still need to source an AC contactor with DC coil to get it to switch off the load at a specific DOD.

Where can I buy the Ve.direct cable or who can I contact to have them made? I understand the Victron cables are $%#$ expensive but unfortunately there's no way I can build one myself. I would like to use the Victron software to monitor the DOD of the batteries on the laptop connected to the inverter.

 

Thanks

C

I know ExSolar has stock (here in Somerset West). Don't know if that helps you :-)

Where are you situated? You can try www.ontrack.co.za, they had the best price on the BMV702, www.rectifier.co.za came in with a R100 more. I don't know whether any of the 2 sell the cables, but both are local Victron agents.

 

Looks like I need to start assembling these cables - opportunity to make some money. 

I know ExSolar has stock (here in Somerset West). Don't know if that helps you :-)

 

Apparently they don't sell the 702, only the 700 when I inquired.

 

What do they ask for the cables?

The VE.Direct cable is R1050 as I recall at ExSolar. Or it was last month when I decided that extra 1k tips the scale just enough that the difference between the 150/70 and 100/50 isn't enough anymore... so I bought the larger MPPT (which has an LCD, no need for the cable).

I'm in Langebaan on the West Coast. Will contact them.

Looks like I need to start assembling these cables - opportunity to make some money.

Great idea, I would certainly be interested in one.

Sent from my iPad using Tapatalk

If superdiy can make you a cable, go for it!!!

 

Retail prices:

VE.Direct to USB interface R 809.93 ex

Interface MK2-USB (VE.Bus to USB) R 809.93 ex

Can find out re stock and what the best price can be if you want new.

The prices from Battery Pit Stop are 10 to 1 old stock. The new stock prices are more and the R14/$ is going to push them up even more in Sept.

 

And I got some info on the T105 versus T105RE's. They are nowhere the same. This is the feedback I got:

Battery Pit Stop s

The prices from Battery Pit Stop are 10 to 1 old stock. The new stock prices are more and the R14/$ is going to push them up even more in Sept.

 

And I got some info on the T105 versus T105RE's. They are nowhere the same. This is the feedback I got:

Battery Pit Stop s

Cool! Posted that feedback because of the comment to me from Battery Pit Stop, claiming the batteries are the same, bar the terminals.

 

The Afrikaans part was feedback from a Trojan email address here in SA, for I wanted to know more on the differences.

 

Am told there is only one main distributor for the Trojan T105 RE's in SA ... and I am one of their smaller dealers with low overheads, no onsite stock to push, so have more room to maneuver. :-) 

@cvzyl got that setup at the moment using a contactor via a 24v relay (cheaper than dc coil) to disconnect the inverter output via the bmv at 40% soc as an emergency measure, the bmv itself stores number of charge cycles, deepest discharge etc as long as you don't disconnect it, it'll store the data for you.

I bought all of my T1275s from battery Pitstop. At their prices, there is no old stock in that shop.

Hi John

 

I don't have 105's or an Axpert. I did an installation for my dad where I've used T875s on a 24V Axpert.

 

Unfortunately the Axpert inverters are not the best type of "chargers" to use with Trojan batteries and not with small battery banks (200Ah-300Ah) either.

 

According to the Trojan datasheets the bulk charge setting for a 24V bank should be 29.6V and for a 48V bank it should be 59.2V. Unfortunately the 24V Axperts can only go up to 29.2V and the 48V Axperts can only go to 58.4V. You'll have to set setting 5 (Battery type) to USE (User-Defined) and setting 26 (Bulk charging voltage) on the Axpert to its highest possible value and setting 27 (Floating charging voltage) to 26.4V on the 24V Axperts and to 52.8V on the 48V Axperts according to the Trojan datasheets.

 

The ideal charge rate should be between 10% and 13% of the battery's 20-hour Ah capacity. For one bank of T105s, that should be between 10% of 225 => 22.5A and 13% of 225 => 29.25 - I would suggest to set it to 25A, but since the Axperts only have 20A or 30A options, I'll use 20A. Setting 2 (Max charging current) should be set to 20A or 70A  for the 4KVA and 5KVA Axpert models. Setting 11 (Max Utility charge current) should then be set to 20A. Hopefully you don't have (a lot of) PV panels installed, because you cannot lower the solar charge current and the 60A (minimum) is extremely high for a 225Ah battery bank.

Hi Superdiy,

could you please clear for me the problem with charging Trojan T105 with Axpert's charger. Does it mean the batteries won't be fully charged and will not be able to deliver all capacity of 225AH and does it also mean that will shorten battery's life?

I have 2*3.2kw Axpert, 20*250w Renesola pvs and would like to have really  good batteries which ( I believe) Trojans are.

Hi Superdiy,

could you please clear for me the problem with charging Trojan T105 with Axpert's charger. Does it mean the batteries won't be fully charged and will not be able to deliver all capacity of 225AH and does it also mean that will shorten battery's life?

I have 2*3.2kw Axpert, 20*250w Renesola pvs and would like to have really  good batteries which ( I believe) Trojans are.

 

I'll jump in so long, but I'm sure Superdiy will also answer. The Trojan's absorption voltage is 14.8 (for a 12V bank) instead of the usual 14.5 or so that is common for lead acids. This means for a 24V bank, you need to take them up to 29.6V for proper absorption. To make it even worse, you need to take them to 31V for equalisation, and the axpert will turn off if you do that! Thankfully you don't have to do that often.

 

So this means the Axpert cannot properly charge a Trojan bank. This might not be a complete deal breaker, but it is an obstacle. If I am wrong, I hope the others will correct me, but as I understand it, at the end of the absorption phase the battery will be at 85% SoC more or less. Chargers will usually switch to "float" at this point, which is normally 27.6V for a 24V bank. 27.6V is however still a "gassing voltage", it is higher than the sit-open-circuit voltage of the battery, and will still force current through the battery in "the wrong direction" causing it to absorb more energy, so you will eventually get them all the way to 100%.

 

(Ie, contrary to popular belief, the batteries aren't "full" when the charger switched to float...)

 

Now if you're absorption voltage is too low, your charger is going to switch to float at a lower SoC. You might still be able to fully charge them, but it's going to take a lot longer to get there. So this might not be a complete deal breaker if you use it as a UPS, but it sort of is if you cycle it daily.

 

At least, that's how I understand it.

 

There is another thread here about calibrating these inverters. It might be possible to adjust them with a deliberate error of 0.4V... but I don't know if the originator of that thread had much luck with that idea.

Edit: To answer your question, unless you spend enough time floating it all the way to 100%, then yes, it won't have the full 225Ah capacity, and sulphation will begin to set in and reduce the life of the batteries.