Simpler inverter/charger

[johntaves]

I worked for 15 years at a company that built UPS, power inverters and large power supplies. Up to 150 kW. I designed several commercial products.

I can tell you that the cost of such products is never, never, never, in the microcontroller.
The power semiconductors, the large heatsinks and/or fans, the huge bypass capacitors, the magnetic components (inductors and transformers), the protection and monitoring devices like fuses, current sensing circuitry, large capacity transient suppressors....... that is where all the cost is. Easily 95% plus.

Actually, because of all the necessary monitoring and protection circuitry, and I mean necessary and not optional, a microcontroller is actually a cost reducer.
And since the microcontroller is already gathering and manipulating so much internal state information, it only requires a few lines of code and an inexpensive LCD display to provide all sorts of good information to the user.

Oh, did I forget that additional lines of code can also provide very useful diagnostics? Like how many charge/discharge cycles the battery has had, the actual run time, overload events, and many more.

Excellent information.

Thanks!

jt

 

[johntaves]

0.2 x 280 = 56A charge.
Than needs some serious balance and monitoring to prevent imbalance!

Not really. .2C is squat for lithium. The fact that these are bigger cells that are taking 56A which requires a fat wire, is irrelevant. The cells should be from roughly the same production batch and thus will be very similar. The BMS will be balancing 24/7/365. It might be several days before the last .0x of volts match on the initial install, but so what? From then on, there's little chance the puny resistors cannot handle the differences. The BMS will shut off charge and loads when any one cell is at the limit. Thus, if they get out of whack, it will be fine, but the total charge the pack can provide will be lower. The UI will tell me this, and I can do something about it.

I expect that the puny resistors will keep this thing in balance for many years.

(I am assuming that "serious" means that the BMS must have either active balancing or just big resistors with heat sinks)

BMS boards as typically sold are last-ditch safety devices only. they are not charge controllers.

True, and that is all that is needed.

The charge section of the system must have both voltage and current control for optimum charging and battery life.

Not really. Just apply 3.6 v (don't quote me on the exact amount) and limit the current to whatever makes sense for your pack. If the BMS shuts off the charge and loads from the extreme ends of capacity, it makes little difference what voltage and current. I believe this is essentially the bulk phase of a lead-acid. As soon as the volts start climbing, as the battery gets near capacity, just stop. There is no need for an acceptance phase. For off grid/RV applications, it does not matter if the battery never sees above say 90%.

Indeed, you can basically hook up a matched solar panel straight to the battery. A 60 (or 72) cell panel will have voltages that seem totally wrong, but since the battery has nearly zero impedance, the volts end up being set by the battery and the panels dump in whatever current they have. A solar controller just needs diodes to stop reverse flow and a relay so that BMS can shut off charge to keep from getting into the high state of charge area.

 

[Nigel Goodwin]

I'm not sure why you concluded I needed to do more research from this statement: "With lithium, generally stay away from fully charged and zero by say 10%."

Because you don't discharge below a certain (fairly high) voltage with Li-Ion, certainly nowhere near zero, actually about 2.75V per cell.

 

[johntaves]

Because you don't discharge below a certain (fairly high) voltage with Li-Ion, certainly nowhere near zero, actually about 2.75V per cell.

Ahh, sorry, "fully charged" and "zero" were meant to be referring to the state of charge, not the voltage.

This was in response to someone asking why I was comparing a LiFePo4 pack with nominal capacity of 6.7kwh vs a FLA pack of 10.8kwh.

 

[rjenkinsgb]

The UI will tell me this, and I can do something about it.

And the UI is driven by a CPU. That's not a "dumb" system.

As you seem absolutely intent on ignoring all points, just do it - it's your money, not mine. I'll wait to see them in the news.

I'm out.

 

[johntaves]

And the UI is driven by a CPU. That's not a "dumb" system.

As you seem absolutely intent on ignoring all points, just do it - it's your money, not mine. I'll wait to see them in the news.

I'm out.

I'm sorry, what point did I miss?
What do you think I am going to do that you think is bad?

Not once have I suggested that the BMS (cell balancer and pack protector) will be a dumb system.

Again, the whole point of this thread is that the BMS will certainly have a CPU and a UI. The question in this thread is whether a CPU in the inverter/charger can be eliminated because the BMS already has the smarts necessary.

 

[Nigel Goodwin]

I'm sorry, what point did I miss?
What do you think I am going to do that you think is bad?

Not once have I suggested that the BMS (cell balancer and pack protector) will be a dumb system.

Again, the whole point of this thread is that the BMS will certainly have a CPU and a UI. The question in this thread is whether a CPU in the inverter/charger can be eliminated because the BMS already has the smarts necessary.

As has been pointed out endlessly, WHY do you want to do that?, removing it will make your design more complicated and more expensive - but you can certainly remove it if you wish.

 

[johntaves]

As has been pointed out endlessly, WHY do you want to do that?, removing it will make your design more complicated and more expensive - but you can certainly remove it if you wish.

I see the confusion. I should have written:

"Again, the whole point of this thread is that the BMS will certainly have a CPU and a UI. The question in this thread is whether a CPU in the inverter/charger can be eliminated, to save costs, because the BMS already has the smarts necessary." If I had written it that way, the "again" word would properly fit with my question in the opening post: "In other words is there a separate communications/UI processor that can be ditched to save money? "

I apologize for not making it clear that I was hoping to make a cheaper inverter.

I also apologize for persisting in making sure that the feedback I was getting was properly considering the situation where there is no need to charge lead-acid and that there will be an external CPU that will shut it off if necessary.

I appreciate the education. Thanks.

 

[Nigel Goodwin]

I'm rapidly coming to the conclusion you're just a troll, posting rubbish and ignoring any replies, I can see this is probably going to get deleted in the not so distant future?.

 

[Reloadron]

If all I want is an inverter I can buy just an inverter on flea bay bare bones. I can also buy an inverter at any number of retailers with my choice of TSW (True Sine Wave) out or MSW (Modified Sine Wave) out. I can add any features (whistles and bells) I want. I can include a uC or just use simple discrete components. Point being I can have whatever I want including kits available. You can have whatever you want.

Ron

 

[johntaves]

I'm rapidly coming to the conclusion you're just a troll, posting rubbish and ignoring any replies, I can see this is probably going to get deleted in the not so distant future?.

WTF! Why? What the hell did I do wrong for christ's sakes? My only guess as to why you are giving me **** is because I did not take your response "Everything you've just described, LCD, buttons, serial, and everything else - microcontrollers make things possible and cheap - using a microcontroller drastically reduces your costs in almost all cases. " as some sort of gospel. Give me a break, of course I don't accept that at face value. How do I know you have a clue? Of course, I am going to ask for more clarification, more detail, more substance.

Sorry, but I don't see how I am behaving like a troll at all. I sincerely wanted to know more about inverters. I have learned, and I appreciate the time you and the others have spent explaining the details.

 

[johntaves]

If all I want is an inverter I can buy just an inverter on flea bay bare bones. I can also buy an inverter at any number of retailers with my choice of TSW (True Sine Wave) out or MSW (Modified Sine Wave) out. I can add any features (whistles and bells) I want. I can include a uC or just use simple discrete components. Point being I can have whatever I want including kits available. You can have whatever you want.

Ron

I have found almost none that will accept an on/off signal for the inverter and separately for the charger. The Victron Multiplus will do it, but it costs over $1k. I bought a WZRELB 4kw 24v unit ($450) and will plop the on/off signal inline with the mechanical switch. I am also buying a mean well RSP-750-27 ($180) charger that will accept an on/off signal. In other words, I am making an inverter/charger from an inverter and separate charger.

So, yea, I agree that there are a ton of inverter/chargers out there, but there are near zero that can function with a BMS that provides on/off signals for charge and loads separately, because the current does not run through it (thus does not have it's own expensive power control stuff).

I think some products need to be updated to handle this sort of BMS, and I was hoping that I might be able to facilitate getting cheaper/simpler inverter/chargers to market.

 

[Reloadron]

In other words, I am making an inverter/charger from an inverter and separate charger.

That's the idea. Good inverters, especially true sine wave inverters come with a price tag. No getting around that but anyway back to the beginning there is no need for a uC or uP. Just a matter of features. Add enough features to a basic inverter and a uC or uP becomes practical over a pile of discreet components. That's about all there is to it.

Ron

 

[johntaves]

That's the idea. Good inverters, especially true sine wave inverters come with a price tag. No getting around that but anyway back to the beginning there is no need for a uC or uP. Just a matter of features. Add enough features to a basic inverter and a uC or uP becomes practical over a pile of discreet components. That's about all there is to it.

Ron

That's not what I was understanding from the others on this thread. They seemed to be saying that if you are going to have overvolt, undervolt, current limit, and auto transfer, then a uC/uP will provide a cheaper implementation than a bunch of analog discrete components.

The prices of these things are all over the map. My 4k WZRELB pure sine inverter (no charge functionality) is $450 to my door. The separate charger is $180. There are low frequency inverter/chargers of say 3kw that will do both functions and should handle inductive loads better than my 4kw high frequency unit for about the same cost ($660). Or I can pay more than double for a victron, magnum, of comparable features (I might get longer lasting components, but I cannot measure that).

My interpretation of the others is that they are essentially saying that I should find a uP/uC in the WZRELB that is handling the under/over volt and current limits (even though I cannot set those values). They are saying that you cannot build a cheap inverter without a uC/uP to do those functions even if those values are hard wired.

 

[Reloadron]

My interpretation of the others is that they are essentially saying that I should find a uP/uC in the WZRELB that is handling the under/over volt and current limits (even though I cannot set those values). They are saying that you cannot build a cheap inverter without a uC/uP to do those functions even if those values are hard wired.

Since I never was involved with design of inverters beats me. I would guess using a uC or uP in most cases would reduce cost. When things like features go in at design time is when the cost analysis is done. Inverters are nice to have and many added features are nice to have depending on what the end user wants and what they are willing to pay.

I have three ancient APC UPS 1500 models here and I forget what the remaining unit my wife's stuff is on even is. Every several years I replace the batteries. Power fails I only need them less than a min before my backup power comes online. The downside to any inverter is the need for large batteries. Just a 1 KW true power inverter on a 24 volt battery (two X 2 in series) is over 41 amps plus any inefficiently. You build (design) an inverter for an intended application much as you buy an inverter for an intended application.

My interpretation of the others is that they are essentially saying that I should find a uP/uC in the WZRELB that is handling the under/over volt and current limits (even though I cannot set those values). They are saying that you cannot build a cheap inverter without a uC/uP to do those functions even if those values are hard wired.

I would likely agree with that. I can do under/over volt with discrete components using a discrete component window comparator and I can also monitor over current using a shunt and comparator circuit but I can do those things using a single small uC having a few A/D channels. I can have adjustable trip limits using either method.

My older APC 1500 units work fine for their intended application. I doubt there is a uC or uP in them The newer unit my wife has on her stuff is nice, has all sorts of things including a nice LCD display which I can scroll through and it tells me all sorts of stuff I have no desire to know. It sits under her table and I have not looked at it in years. About once a month I make a journey down the basement and turn off mains to make sure our backup generator transfers so I figure I would know if any UPS wasn't working. Guess it just depends on what the user is interested in.

Ron

 

[johntaves]

I would likely agree with that. I can do under/over volt with discrete components using a discrete component window comparator and I can also monitor over current using a shunt and comparator circuit but I can do those things using a single small uC having a few A/D channels. I can have adjustable trip limits using either method.

Thanks for that info. I am concluding that there is no decent cost savings by attempting to dumb down the inverter.

 

[Reloadron]

Thanks for that info. I am concluding that there is no decent cost savings by attempting to dumb down the inverter.

None that I can see but again, not my forte.

Ron

 

[ChrisP58]

I just bought 8 cells of 280ah for $1000 from Xuba in China delivered by slow boat. This will be 6.7kwh capacity. I figure that's similar to 8 trojan 6v 225ah for $160ea., assuming we have an effective usage of say 85% of nominal from lithium and 50% of nominal from lead acid.

Sorry about my comment. On the page in your link it says "1Pieces = $330.68". Elsewhere it does imply that you're buying "4PCS". For this to make sense you have to understand that 4PCS = 1Pieces.
I would have written it to say that the $330.68 price is for a 4 piece lot.