cap on battery, influences charging?

[etriker]

Edit: well, after plundering more about capacitor role - I'll admit that only counteracting inductance is a too brave a statement - they're certainly used for many applications. Merely ment to point out that the role of the capacitor is to "smooth voltage" in most cases and that it, sort of, works exactly the opposite of a inductor. As for cap on a battery, no use - commercial chargers don't use them either, I've checked. You'll need a hell of a capacitor to make a difference.

I definitely agree that you need caps of several hundred farads or larger to make a difference in an e-bike app. Big, heavy, expensive and not worth it in that application IMHO. And you need HUGE cap banks for transportation uses (bus, railroad, etc.), but they are used. And (ultra)caps are used in parallel with batteries to provide better low-temperature performance than battery-only systems with pulsed loads. Even very high performance LiPo's lose a lot of their capability at low temps.

Additionally, small caps are used extensively in parallel with a battery in energy harvesting applications to provide the initial current burst to an intermittent load. This keeps the battery from having to supply all of this initial burst and measurably increases the battery's run-time since these batteries often have very high ESR and waste a lot of power if required to source current for larger pulsed loads. This is critical for remote sensors and other devices that use energy harvesting for storing power and need every electron they can find/store.

And every single commercial charger based on a switching power supply circuit (i.e., a LOT of them) uses capacitors on its input and output for ripple smoothing and as part of the compensation network that stabilizes the circuit's output when the load changes (transients). And also for decoupling/bypassing and charging-timeout functions too. They're not huge caps, but they are caps...and they're used in commercial chargers.

And an ebike controller is a switch mode power supply with caps in it.

 

[dnmun]

no it is not.

 

[etriker]

no it is not.

It does not have to be that is for sure !

Some of the newer designs will take a 48vdc input and convert that to pulses that are sent to the motor.

The pulses are created by switching transistors on and off. That is what I call a switch mode power supply.

 

[dnmun]

i cannot tell if this is a joke or if you really think this stuff is true.

 

[etriker]

i cannot tell if this is a joke or if you really think this stuff is true.

I am a power supply tech.

I have been repairing hi tech power supplies since 1986.

cap link

http://badcaps.net/forum/index.php

 

[nechaus]

iv found cement in some 1farad caps.. they like to make it feel like a heavy quality product or something


bastards

iv hooked up x4 12 volt 1 farad caps in series to a controller and some sla's..

With out the batteries hooked up, it would be enough power to pretty much Jolt the bike while your sitting on it, or lift the rear wheel and it will start spinning a wheel for a few seconds..


better off getting extra rc lipo

 

[ProDigit]

Well,
It's quite clear from this conversation, that a capacitor (even of respectable size), does not contain enough energy for acceleration. At best, it will even out pulses generated by the controller, and cause your voltage meter not to drop so quickly when accelerating, and perhaps in minimal ways extends your battery life by a bit.

Another thing is, at 1F it takes about a good 3 seconds for the capacitor to be completely empty. It's not the power within the capacitor that is driving the bike, but the battery is. The capacitor is merely affecting a few volts at initial acceleration on top of the voltage drop; enough not to see your fuel gauge go all the way to 50% of battery in acceleration, and enough to perhaps prevent the controller from shutting down due to low voltage spikes; but not enough to make a big difference power wise.

 

[amberwolf]

Doesn't matter how long it takes a cap to be "empty", as it's irrelevant when driving a load like an EV motor. Once it drops below the controller's LVC it can't do any more work.

 

[etriker]

iv found cement in some 1farad caps.. they like to make it feel like a heavy quality product or something


bastards

iv hooked up x4 12 volt 1 farad caps in series to a controller and some sla's..

With out the batteries hooked up, it would be enough power to pretty much Jolt the bike while your sitting on it, or lift the rear wheel and it will start spinning a wheel for a few seconds..


better off getting extra rc lipo

I found that A123 26650 cells even in a 1p setup will work kinda like output caps.

They will charge and discharge very fast. Fast enough for ebike power surges.

In my setup they are recharged by a laptop cell pack. Laptop cell packs hold a lot of power for their size but need to be large to supply the power surges.

The A123 cells take care of the surges and the laptop cell pack charges them back up going down hill or at stops or slower speeds.

I can not do rc lipo tests where we do tests for safety reasons but I am interested how they would work as output caps.

 

[BatteryMooch]

I found that A123 26650 cells even in a 1p setup will work kinda like output caps.

They will charge and discharge very fast. Fast enough for ebike power surges.

In my setup they are recharged by a laptop cell pack. Laptop cell packs hold a lot of power for their size but need to be large to supply the power surges.

The A123 cells take care of the surges and the laptop cell pack charges them back up going down hill or at stops or slower speeds.

I can not do rc lipo tests where we do tests for safety reasons but I am interested how they would work as output caps.

Interesting!
We've gone in the other direction...testing 4S1P A123 M1a packs with ultracaps. The A123's supply the bulk capacity and six 100F to 600F caps (in series, with balancer) help with the 350A-800A pulses we need for a particular application. The A123's can do that themselves, but the caps increase the power delivered by 30%-250% (by increasing the voltage-under-load), depending on temperature. At room temps, we could probably use LiPo's but it would take a LOT of effort to protect pouch cells from the rather severe operating environment. At cold temps, -10F to -50F, the ultracaps are fantastic. We only lose about 20% of the power (vs. room temp) at -20F.

 

[ProDigit]

Doesn't matter how long it takes a cap to be "empty", as it's irrelevant when driving a load like an EV motor. Once it drops below the controller's LVC it can't do any more work.

It does matter. If a cap can provide a voltage difference of 2Volts for 3 seconds, that might be enough for the controller not to shut down (they tend to shut down early, when voltage drops low at acceleration, but you can still drive on the battery for several minutes if you drive carefully, without acceleration.

True that it's not going to give a LOT of difference, at most it will help you accelerate one more acceleration out of your bike without the controller shutting down, but I'm not installing the cap for that reason; those are but side effects of having a cap installed.

 

[amberwolf]

Doesn't matter how long it takes a cap to be "empty", as it's irrelevant when driving a load like an EV motor. Once it drops below the controller's LVC it can't do any more work.

It does matter. If a cap can provide a voltage difference of 2Volts for 3 seconds, that might be enough for the controller not to shut down (they tend to shut down early, when voltage drops low at acceleration, but you can still drive on the battery for several minutes if you drive carefully, without acceleration.

True that it's not going to give a LOT of difference, at most it will help you accelerate one more acceleration out of your bike without the controller shutting down, but I'm not installing the cap for that reason; those are but side effects of having a cap installed.

No, it still doesn't matter how long it takes for the cap to be "empty".

Empty means 0V for a cap. Below the LVC of your controller is unusable, because the controller no longer drives the motor at that point, and is no longer pulling power from the caps or battery.

It only matters how long it takes the cap to discharge from whatever it's "full" voltage is to whatever the controller's LVC is, as that is the only usable voltage range.

In reality it only really matters how long it takes the cap to discharge from it's "full" voltage of the moment (whatever the pack voltage was just prior to that moment's discharge start) to whatever the max sag of the batteries is at that particular state of charge and that particular moment's power draw from the controller.

 

[ProDigit]

Doesn't matter how long it takes a cap to be "empty", as it's irrelevant when driving a load like an EV motor. Once it drops below the controller's LVC it can't do any more work.

It does matter. If a cap can provide a voltage difference of 2Volts for 3 seconds, that might be enough for the controller not to shut down (they tend to shut down early, when voltage drops low at acceleration, but you can still drive on the battery for several minutes if you drive carefully, without acceleration.

True that it's not going to give a LOT of difference, at most it will help you accelerate one more acceleration out of your bike without the controller shutting down, but I'm not installing the cap for that reason; those are but side effects of having a cap installed.

No, it still doesn't matter how long it takes for the cap to be "empty".

Empty means 0V for a cap. Below the LVC of your controller is unusable, because the controller no longer drives the motor at that point, and is no longer pulling power from the caps or battery.

It only matters how long it takes the cap to discharge from whatever it's "full" voltage is to whatever the controller's LVC is, as that is the only usable voltage range.

In reality it only really matters how long it takes the cap to discharge from it's "full" voltage of the moment (whatever the pack voltage was just prior to that moment's discharge start) to whatever the max sag of the batteries is at that particular state of charge and that particular moment's power draw from the controller.

Like I said, it's perhaps good enough to compensate for a short voltage dip. But the main reason for putting a cap is for a much cleaner voltage.
The only problem so far is finding large capacitors for 48V.
There are plenty of caps for higher AC voltages, but they not many have a high capacity (usually in the ranges of a few mF).
And high capacity caps for lower voltages (usually 12-24V) are also plenty.

It's more difficult to get my hands on high capacity 48V caps. Probably a series connection is necessary.

 

[ZOMGVTEK]

You want gigantic storage capacitors on your battery to clean up the power coming out?

I don't get it.

 

[ProDigit]

You don't need gigantic cap;
One of 1Farad would be nice at 48V.
It would do a good enough job with cleaning up the DC voltage.
Not to provide the motor with more voltage, but to filter out the noise on the battery.

 

[ZOMGVTEK]

Why do you care about noise on the battery? Batteries tend to handle that stuff really well. It's not like you're running an RF transmitter on an unregulated power supply.

I would consider a 1F capacitor of sufficiently low ESR to be a VERY large capacitor, especially at that voltage.

 

[ProDigit]

probably, but they're not that big in size. They can fit almost anywhere on an ebike or scooter!

And for those pulses you'll need a good cap. 1 farad cap is smaller in size than a can of axe deodorant!
Even if batteries can handle them fine, it's always better to not have them at all!

 

[Punx0r]

Why would you need to filter the output of the controller? This sounds like a solution to a problem that doesn't exist...

 

[amberwolf]

He's talking about filtering the output of the *battery*, input into the controller.

I don't see the need either, because adding more battery in the same volume and weight the capacitor would take would almost certainly do more to help than a simple capacitor would, both by storing more energy total and by supplying current to help mitigate sag.

He'll have to test his battery both ways to find out which one will actually work better.

 

[Punx0r]

I'm still confused...

AFAIK batteries output the purest DC you're likely to get, so filtering is pointless. If we're talking about combating sag, wasn't it established above that this just wasn't practical?

 

[ProDigit]

The batteries are indeed giving a clean output, but the controller (due to chopping voltages) raises and lowers the voltage several times a second; creating a very dirty output.

My only fear by connecting a battery in parallel is that I will destroy one battery because of one battery overpowering the other.
Besides when charging one battery, the other will drag on the one, and charging two batteries simultaneously, could cause damage (as both chargers will push voltage limits on the bike and might damage some stuff.
So diodes are necessary, but diodes cause a voltage drop.

I'm more concerned about battery life than anything else. However, I'm still looking for a cap that is rated above 48V, and has a high capacity.

 

[Teh Stork]

I'm still confused...

AFAIK batteries output the purest DC you're likely to get, so filtering is pointless. If we're talking about combating sag, wasn't it established above that this just wasn't practical?

You are right. Don't let ProDigit's endless mumbojumbo get in the way of common sense. Several members have done the math, he just continues to downright ignore it over and over again.

 

[amberwolf]

The batteries are indeed giving a clean output, but the controller (due to chopping voltages) raises and lowers the voltage several times a second; creating a very dirty output.

Which has absolutely nothing to do with the input or the battery, and is what teh controller is SUPPOSED to do at it's output--that's how they work!

If you want to change that, putting caps or anythign else on the input won't do it. You need to redesign the controller and add filtering on it's phase outputs, and that will depend on the actual motor you use, in addition to the controller.

My only fear by connecting a battery in parallel is that I will destroy one battery because of one battery overpowering the other.
Besides when charging one battery, the other will drag on the one, and charging two batteries simultaneously, could cause damage (as both chargers will push voltage limits on the bike and might damage some stuff.
So diodes are necessary, but diodes cause a voltage drop.

If they're the same battery type I don't see how that could happen. If they're not the same, you need to go read all the threads about this kind of thing, so you will understand how this stuff works--I'm pretty sure based on the way you state things that you haven't got a full grasp of it yet, and each time people explain things you seem to "miss" what they are trying to tell you.


Sorry if I sound harsh--I don't mean to. But it is getting quite frustrating.

 

[etriker]

There is already a cap on the input of the controller. In the controller.

That is where it is needed.

 

[ProDigit]

The batteries are indeed giving a clean output, but the controller (due to chopping voltages) raises and lowers the voltage several times a second; creating a very dirty output.

Which has absolutely nothing to do with the input or the battery, and is what teh controller is SUPPOSED to do at it's output--that's how they work!

If you want to change that, putting caps or anythign else on the input won't do it. You need to redesign the controller and add filtering on it's phase outputs, and that will depend on the actual motor you use, in addition to the controller.

And how do you think a chopped output would look to the input of the controller?
Chopped!

There's no way that the input signal of the controller is clean with that high power draw!

If they're the same battery type I don't see how that could happen. If they're not the same, you need to go read all the threads about this kind of thing, so you will understand how this stuff works--I'm pretty sure based on the way you state things that you haven't got a full grasp of it yet, and each time people explain things you seem to "miss" what they are trying to tell you.


Sorry if I sound harsh--I don't mean to. But it is getting quite frustrating.

Perhaps because the reverse is true.
You guys don't seem to grasp what I'm talking about, so I constantly have to explain myself, which may lead to both of our frustration.

I wasn't talking about a same battery, but connecting my XB-700Li's internal 20Ah LiPo4 battery, with a LiFePo4 battery that's 10 or 15Ah, and much smaller than the stock bike battery!