I need a voltage follower with >= 1Amp output for experiments

[(*steve*)]

It looks to me like the LTC3300 is exactly what you need.

It can transfer (via a transformer isolated power supply) charge from stronger cells to weaker cells (during discharge) and from weaker cells to stronger cells (during charge). This is exactly what I mentioned in post #17.

A single LTC3300 works for up to 6 cells. For more cells you can daisy chain them.

 

[Gazza_AU]

I also read the data sheet, as said. Problem solved.
Seems we are not paying attention to each others posts.

Firstly, I have no idea why the current average cell voltage is particularly meaningful.

The interest is in the equilibrium created by 2 voltages in parallel, regarding work.
Also, how this knowledge can be used further with comparator's and gates.

I would explain as follows:
The 17 cells in series are connected to solar Charger.
These cells each have a voltage across that adds up to the Bank Charging voltage whilst connected?

The Injection Voltage is Bank Voltage divided by number of cells.
Where as, the voltage across each cell will be above, bellow or equal to bank voltage division.

Current is entering the series cells via Solar Controller.
If the injector is connected to a cell, and the injector voltage is higher than that across the cell,
then current enters the cell via injector.

The cell voltage across a cell, connected to injector, will now be the result of:
Cells series voltage from the Solar Charger, in parallel with injector voltage across cell.

Are you saying the injector becomes a sink for cell,
when it is lower then the Solar Controllers voltage across cell?

I am curious to know more, if you are interested.
If not, just let me know.

Cheers

 

[(*steve*)]

The interest is in the equilibrium created by 2 voltages in parallel, regarding work.
Also, how this knowledge can be used further with comparator's and gates.

That doesn't make sense. Two voltage sources in parallel will cause zero current to flow if they have the same voltage and an infinite current (and an undefined voltage) if the voltage are different.

To make sense of your question we have to assume some properties additional to a perfect voltage source. Without knowing what these additional properties are, the question cannot be answered.

If it can't be answered then anything based on it is likewise unknowable.

 

[Gazza_AU]

Two voltage sources in parallel will cause zero current to flow if they have the same voltage and an infinite current (and an undefined voltage) if the voltage are different.

I am not saying you are not perfectly right, that's just the question.
Thanks for your input.

How do I calculate what the voltage across cell would be with Injector connected.
For a cell with 3.3V across, connected in parallel with 3.4V injector.
And,
For a cell with 3.5V across, connected in parallel with 3.4V injector.

I thought the effect of such parallel voltages,
would be similar to 2 cells in parallel, in the first instance.

The two cells in parallel eventually come into full balance.
Whole Banks can be prepared in this way.

My earliest thinking on this concept was to have a spare separate cell as the injector,
not to say this concept has any merit either.

I have done some more research, on BMS "Chinese sellers" are claiming to be Active,
they are just passive and do not start balancing by limiting charge to high cells until after 4v..
Making LTC3300 balance circuit the only real option I can find!

 

[(*steve*)]

I thought the effect of such parallel voltages,
would be similar to 2 cells in parallel, in the first instance.

That requires both voltage sources to have some internal resistance and the feature of having the voltage drop as charge is removed, and voltage rising when charge is added.

Those features are not those of a theoretical voltage source, not are they what you see from a voltage regulator.

I think the LTC3300 is a really good solution. Keep us updated as to how you get on because I'm certainly interested.

 

[Gazza_AU]

Let you know what eventuates with LTC3300.

Your explanation is intuitive regarding internal resistance, makes sense.
So it would require something with higher internal resistance then cell.
http://electronicdesign.com/power/can-supercapacitors-surpass-batteries-energy-storage
Keeping an open mind.

 

[Gazza_AU]

What could I try in > or = 25F Super cap.

Anybody up with the best dollar value for large caps?
Need to handle 3.5v

How would you go about it?

EG. 2x 2.8v 25F Caps in series would give a 5.6v rating?
Supercapacitors / Ultracapacitors 25F 2.8V EDLC HB SERIES CYL
http://au.mouser.com/ProductDetail/PowerStor-Eaton/HB1625-2R5256-R/?qs=sGAEpiMZZMuDCPMZUZ%2bYl5pHcBUbMqmZb28d98DBxo8=

Or 300 F :

2x BUSSMANN BY EATON XV3550-2R7307-R Supercapacitor, Supercapacitor, 300 F, 2.7 V, +10%, -5%, Quick Connect, Snap-In
http://au.element14.com/bussmann-by-eaton/xv3550-2r7307-r/cap-super-300f-2-7v-snap-in/dp/2148519

What do I need to know when choosing?

 

[(*steve*)]

EG. 2x 2.8v 25F Caps in series would give a 5.6v rating?

Yes, but remember that if you want 25F and you are placing two equal capacitance capacitors in series, they will have to be 50F.

 

[Gazza_AU]

Yes, but remember that if you want 25F and you are placing two equal capacitance capacitors in series, they will have to be 50F.

Yes, I was not thinking it through.

Do you know much about large capacitors to help in choice.
What would give the best chance of equalizing with 100Ah cell?

I have been doing some research into the sequencer, and keen to have a play with the idea.

Will not be connecting this to the actual cells, this is hypothetical to explore the idea.
And I have other uses for the equipment.

 

[(*steve*)]

What type of cell?

 

[Gazza_AU]

What type of cell?

3.2v lifepo4 100Ah Prismatic.

cell 3.2V 100Ah for different packs GBS-LFP100Ah-A
Internal impedance ≤0.6mΩ
Although the method is not stated in Data sheet, It is common to measure impedance using AC..
This value is not believed to be of great value, but that's whats available.
3.6v is the upper charge voltage limit, for my example using PSOC.

The idea is in flux, nothing set in stone.

But we start with the Bank divider that gives an average of voltage across cell.

Scenario A: The capacitor starts by receiving this charge voltage, adjusted for losses + a few milli-volt.
So assuming we have already addressed current inrush. The cap then charges to voltage.
Is disconnected from divider source and connected to first cell.
Is disconnected from the first cell and connected back to divider source.

Or Scenario B: The cap is connected to voltage divider source constantly, and jumps from cell to cell.

The later needing control circuit like is used in modern electronics needing higher current then supply.
Storing Power with Super Capacitors: http://www.skyworksinc.com/uploads/documents/202377A.pdf
or just a resistor and diode?

Power relays are 10A, AC 250V, DC 30V.

 

[(*steve*)]

I'm not really sure what you are trying to do. Charge the lithium cells? charge the capacitors from the cells?

Either way, both scenario A and B seem to be poor choices.

 

[Gazza_AU]

I'm not really sure what you are trying to do. Charge the lithium cells? charge the capacitors from the cells?

Either way, both scenario A and B seem to be poor choices.

The idea began with a claim that jumping a capacitor around from one cell to next in a series string,
would balance out cells, evidently because the Cap would either give or take depending on cell voltage in string.

I think this is crap, that it did not work!

But I would like to hear if you agree it could not work.
Or if you think it could work under certain conditions.

This variation differs, in that the cap is held at average cell voltage before being applied to cell.
The result should be, the cell and cap equalize. The question is not will it work.
More, what design gives the best possibility of doing something.

The other aspect to note, is the Bank divider source, follows changes in different charge settings.
The norm with resistive dissipation is balancing can only be done at just before full capacity, ruling out PSOC.

The individual cell voltages; that add up to bank voltage, are always going to be different.

The standard methods use many forms of measurement to estimate cell capacity.
This approach doesn't sense cell capacity, the cap is topping low cells with a higher then average voltage, because their share of the bank voltage was already lower then others.

 

[(*steve*)]

The LTC3300 solution allows charge to be moved from one cap to another. It is a really good solution.

Any solution which is not totally or partially dissipative will employ inductors in some manner.

 

[Gazza_AU]

I think the LTC3300 is a really good solution. Keep us updated as to how you get on because I'm certainly interested.

I like to contribute to all posts, with ideas and discussion.
It doesn't matter who posts, or what they ask.

Ummmmm, because there's a transistor in there perhaps.

For some reason you are arrogant, and unwilling to give anything to discussion other then negativity or criticism.

If you are certainly interested in the LTC3300, why not study the IC.
Make inquires with Linear/Analogue and share your findings for others to benefit?

Seems like a good argument for "get what you give".
Short sharp replies?

 

[(*steve*)]

The problem is that you are asking about stuff that requires more knowledge to understand than that which you have.

Your complaint about one of my replies is somewhat ironic considering it was an exasperated response to you not recognizing what was an example of that which you initially requested by name. It's like someone wanting a car and then when shown one, pointing to the wheels and asking "won't they stop it moving?". It's kinda a reasonable assumption that if you ask for a car by name that you know the basics of the operation of wheels.

Even though you lack the understanding of current good solutions, you try to come up with other solutions which either won't work or will be grossly inefficient.

It's almost impossible to explain why because of the gaps in your knowledge.

You're trying to solve a hard problem. You've found a good solution that will do exactly what you want, yet you feel that (even without the knowledge to understand the problem in any depth) you can come up with workable alternatives.

Sometimes you need to realise that some additional background knowledge might come in handy.

 

[Gazza_AU]

You seem obsessed with selling me a Linear IC, because you lack any imagination to consider anything different.
You have preconceived ideas that cloud your ability to think outside.

The project is going well.

Take care

 

[(*steve*)]

Best of luck.