Understanding ripple voltage?

[DrKnow65]

I have just begun using the new Mamba XL with data logging and one of the parameters is ripple voltage, I am curious about what to make of this data?

What is a good voltage vs. bad voltage, understanding that lower is better.

[What's_nitro?]

Ripple describes how the voltage varies under load. Less ripple is always better, since we want the voltage to be as consistent as possible.

You might notice as your packs age, and their internal resistance increases, that the ripple also increases. Because of their higher resistance, their voltage sags more under load. This is also why using sub-par batteries with high-power systems is a bad idea.

[What's_nitro?]

Some light reading if you wish.

http://en.wikipedia.org/wiki/Ripple_(electrical)

Not exactly the same, but I think you'll get the idea.

[DrKnow65]

I do understand that the ripple is electrical feedback from the ESc to the battery with the resistance of the battery working to sink (or stabilize) the voltage.

What I want to know is what is an acceptable volatge? and what all effects this value?

[What's_nitro?]

Ripple itself isn't actually feedback. It can cause feedback, and that feedback can damage components.

So a BL motor has 3 phases... Every time a phase is sent power by the ESC a high load is placed on the battery. If the battery has trouble supplying the current that the motor phase wants, the voltage at the battery drops for a split second while that phase gets power. Just before the next phase of the motor recieves power, the voltage at the battery jumps back up. Rinse. Repeat. Imagine this happening at 3000 Hz, as it would inside a 2-pole motor revving at 30K rpm. You basically have an AC waveform being fed to the ESC that wants DC power. Bad stuff...if the ripple amplitude is high enough.



I'm not sure what an acceptabe value would be for the MMM... Maybe Patrick will answer that one.

[DrKnow65]

I'll see if I can figure out how to post my first data captured, perhapse it will illustrate specifically what I'm after.

Thanks for taking the time to explane that nitro :-)

[DrKnow65]

Pack voltage on the top (8S lipo), ripple voltage on the bottom...

[What's_nitro?]

So by ripple they actually mean simple voltage drop? If that entire graph represented 0.1 seconds it would be showing ripple...

Looks like the logger missed a big spike at around the 42 second mark.

[DrKnow65]

yah, kinda confusing... hense the thread

[What's_nitro?]

Originally Posted by DrKnow65

yah, kinda confusing... hense the thread

Now I see where you're coming from...

[Pdelcast]

Let me see if I can explain--

The ripple voltage graph shows the difference between a voltage sample that is taken in the middle of the "on" PWM cycle, vrs a sample taken in the middle of the "off" PWM cycle.

The PWM (or throttle) runs asynchronously to the commutation cycle... so don't get the two confused.

So, for example, if the battery voltage sags to 6V when the motor is switched on, and rebounds to 8V when the motor is switched off (which happens 12,000 times a second when not at full throttle) then that would be 2V of ripple, and the graph would show 2V.

The data log shows the battery voltage, but it's averaged.

Usually the ripple voltage will drop a LOT when at or near full throttle, or when at or near zero throttle. And it will be the worst at or near 50% throttle. This is because the pack voltage either drops (near full throttle) to the minimum point, or rises (near zero throttle) to the maximum point.

A good place to be is at less than 5%-7% of pack voltage for the worst-case ripple voltage.

An OK place to be is 7%-10% of pack voltage for worst-case ripple.

10%-12% is marginal, and is risky.

If ripple voltage is above 12-15% (or more) of pack voltage at any point, it's likely that the battery isn't up to the task, and you run the risk of damaging the battery and the electronics.

The graph shown here shows a worst-case of about 2.2V of ripple on a 34V pack -- about 6.5%-7% ripple. Not a bad place to be.

PLEASE NOTE -- RC Cars are pretty hard on the system during acceleration, so very quick excursions above the "marginal" point is OK -- as long as the system doesn't spend any significant time above the "marginal" or "risky" range, it will usually be OK.

Hope that makes sense!

Patrick

[What's_nitro?]

That does make sense now that you say it's based on the PWM signal... I now understand why it can be harmful to the ESC to limit the truck's speed with Tx trim. Thanks!

[DrKnow65]

Patrick once again you prove just how awsome you and your company are.

I understand now what the graph is showing and how to quantify the results. I'm glad the packs are doing ok, and it will be cool to watch as they progress.

I'm assuming the ripple shold get better as the packs wear in (new) and then worsen as they age and the internal resistance climbs.

Thanks for putting up percentages, that gives me a good idea of what to look for and it should prove to be a handy tool to indicate the quality of the cells I'm using as I grow an inventory of packs.

[Pdelcast]

Originally Posted by DrKnow65

Patrick once again you prove just how awsome you and your company are.

I understand now what the graph is showing and how to quantify the results. I'm glad the packs are doing ok, and it will be cool to watch as they progress.

I'm assuming the ripple shold get better as the packs wear in (new) and then worsen as they age and the internal resistance climbs.

Thanks for putting up percentages, that gives me a good idea of what to look for and it should prove to be a handy tool to indicate the quality of the cells I'm using as I grow an inventory of packs.

We've noticed that Lipos really don't change much at all, until they start to get seriously old, and then they go south really quickly.

You may not notice any break-in at all.

We also have noticed that Lipos are REALLY BAD when they are cold. So you should keep 'em warm before running them in really cold weather.

[Kirkinsb]

To calculate ripple from the Castle Data Log Viewer...I took Max voltage 33.5 volts divided by Max ripple of 2.53 volts. That gave me a ripple of about 7.5% based on 33.5volts or 6.3% based on 34 volts. Is this correct to us Max volt numbers? Also I noticed the peak ripple is for only for an instant...So am I ok on ripple? Do I use 33.5 or 34 as the max...or does that matter so much? I am either 6.3% or 7.5%