# moving the B down the curve

• posted

I am wondering if it is possible to use an anti-current to push the B down the curve. Let me explain.

I need a large inductance at a high current for a 60 Hz battery charger. 5

0+v at 80amps should do fine. Of course, We =LI^2 is going to determine the physical size of this unit, and it will be large. (Obviously PFC would be a natural way to go, but time is a factor. )

I suppose it would be possible to inject an opposing current to push the fl ux down so that it is almost centered on the core. Would this allow me to u se a smaller core? I am also aware that you don't get something for nothin g - so I expect the pain is just moved from the inductor to the power suppl y I need to inject the counter current.

This is similar to the closed loop LEM sensors, without the large inductanc e.

Thanks Bob

• posted

You can use many turns at much lower current to do that. Unfortunately that makes a high-ratio step-up transformer, so that any quick changes in the high-current side will tend to blow up the low-current side.

Cheers

Phil Hobbs

```--
Dr Philip C D Hobbs
Principal Consultant ```
• posted

pretty sure I saw a configuration like that for tube circuits was patented circa 1930 where it ran the appropriately-phased tube filament current thru the B+ filter choke, but for the life of me can't remember the proper search term to bring it up right now.

of course that situation is easier because the filament current side is the high-current one in tube circuits so the step-up works out a more appropriate way.

• posted

Ah! Op's situation is easy then, just need a high-voltage low current AC "dummy" supply and do the same thing but in reverse. Like a neon sign transformer a few kV at a few mA.

I'm only half-kidding the combined volume might actually be somewhat smaller than a lone big-ass inductor, have to do the math and see :)

• posted

Sure. But then the ripple voltage is multiplied, and it's equivalent to pass-regulating the supply with a transistor and no inductor. You're just using the inductor as a transformer, and its finite inductance only makes things worse.

What's wrong with, say, buying an off-the-shelf battery charger? Or a e.g. MeanWell something or other, 48V supply with PFC, and either tweaking the output voltage, or hacking it to control via custom interface, or using it as a hybrid supply with your own postreg controller, or..?

Personally, anything with heavy iron is way down my list, even if I were told it has to be done "fast".

Tim

```--
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Design ```
• posted

n the curve. Let me explain.

50+v at 80amps should do fine. Of course, We =LI^2 is going to determin e the physical size of this unit, and it will be large. (Obviously PFC woul d be a natural way to go, but time is a factor. )

flux down so that it is almost centered on the core. Would this allow me to use a smaller core? I am also aware that you don't get something for noth ing - so I expect the pain is just moved from the inductor to the power sup ply I need to inject the counter current.

nce.

I suspect a lot of us have cogitated over schemes like that. In principle y es it can be done. In practice it creates problems & the end solution is fa r less preferable than just getting a bigger inductor. If you're not findin g hat you need off the shelf, put some in series/parallel.

NT

• posted

You could use a permanent magnet to offset the flux, and that has been done. Sounds complicated.

Cheers

```--
Clive```
• posted

it is, but it's not practical, your current source needs high voltage compliance.

This can be mitigated by putting an inductor in series with current source, but now you've just moved the iron from one place to another.

inductors with permanent magnets built in are another thing. maybe that can work for you?

```--
When I tried casting out nines I made a hash of it.```
• posted

How is your inductance requirement being determined?

Batteries are often quite happy to absorb lower frequency ripple.

RL

• posted

Ok. Thank you all for responses. I am hearing what I expected. Doable, b ut not easily - as I figured, the pain moves from large(reliable)magnetics to something that would take more time to develop than a simple PFC solutio n.

The situation is a 48V 4x8 array of deep cycle batteries to run my house. I have plenty of solar, but does me no good on ost winter days in New Engl and. I bought a pair of 25A/48v chargers but they don't put out enough curr ent (never get more than 30A out of them in parallel, and they can't seem t o figure out when to de-sulphate.

I already have a large tape wound core that I will use to get me to about 6

0vdc with 200v in. Generator is set to 200vac/50Hz. I also have diodes and heatsink. I will buy some caps and follow it with a Chinese charge contro ller. I have one on the panels and it works very nice. I do not know how much ripple voltage the MPPT charge controller will take before it gets con fused. Caps seem to be cheap these days so inductor may not really be neces sary.

Tim - Yes, Meanwell is doing a nice job these days. I have been using them for a while. Seems every time I look they have expanded their product lin e up

• posted

wn the curve. Let me explain.

50+v at 80amps should do fine. Of course, We =LI^2 is going to determi ne the physical size of this unit, and it will be large. (Obviously PFC wou ld be a natural way to go, but time is a factor. )

flux down so that it is almost centered on the core. Would this allow me t o use a smaller core? I am also aware that you don't get something for not hing - so I expect the pain is just moved from the inductor to the power su pply I need to inject the counter current.

ance.

They don't seem to be bothered with a lot of ripple. I would think they mi ght help with keeping cells desulphated as well. I have seen a NaVY spec f or a battery charger with ultra-low ripple. They seem to think ripple wea rs out battery?? Inductance requirement would probably be determined by no t confusing the MPPT charge controller. Since I am running 5kW on 12kW gen erator I would like to keep some control of harmonics too.

• posted

Actually, charging a (lead acid) battery works better with ripple - up to a point the more ripple the better; tends to desulphate.

• posted

I think only in the sense that batteries have a finite amp-hours in and amp-hours out before they are wear out and ripple is kind of AH in and AH out at a higher rate. Good question !

As long as the controller measure the Max power point and keeps the PV panel running at that voltage, then it can't be confused. Ripple current can certainly make the PV voltage move around and not stay on the MPP so it "should" ideally be able to regulate the voltage at the ripple frequency (usually 50 or 60 Hz)

• posted

Supposedly that's how these HF pulsating desulfators work. If the batteries aren't already sulfated, it's best to not let them get that way which is done with a periodic absorb charge stage (raised charge voltage)

• posted

The situation is a 48V 4x8 array of deep cycle batteries to run my house. I have plenty of solar, but does me no good on ost winter days in New England. I bought a pair of 25A/48v chargers but they don't put out enough current (never get more than 30A out of them in parallel, and they can't seem to figure out when to de-sulphate.

If they're always sulfated, then replace them.

The best you can do, AFAIK, is a lengthy (days, weeks?) float at the recommended float voltage (which for those I think is 14.3V or thereabouts). This will charge away as much lead sulfate as the electrodes can reach. If they are well and truly gummed up, there's nothing more you can do: if the remaining capacity is insufficient, replace them.

The last thing you want is excessive charge current. That pushes the voltage up, activating the next reaction, water electrolysis. Bubbles disturb the electrodes and make things worse over time.

Don't forget to check the cells' level and top up with DI water as needed.

Tim

```--
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Design ```

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.