Battery charge graph

[solarengineer]

Hello, I have a current sensor in series with my windmill and battery +

The current sensor is designed to be on the + line

What i dont understand is, my logger takes 1 sample per second of the battery as it is being charged. the graph jumps up and down like crazy. pls see picture.

I thought it might be noise but if i check the current sensor with a dc load it shows the current as a stable straight line.

I realise the graphs numbers are in the negative, its just the way i have the sensor connected.

I have a full bridge rectifier on the generator output and a 22,000uF capacitor connected across the  + and - of the bridge to filter the noise.

So is what im seeing in the graph normal as a battery is being charged? or any thoughts on why the graph would be so jagged?

Thanks for any help

[solarengineer]

BTW this is over the course of about an hour.

Thanks

Jamie

[Nando]

Jamie:

The graph is too compressed to see what is going on,

You need to detail how the equipment is, type of charger etc.

It seems that the charger may be going ON/OFF at high rate.

Detail the problem to have better reports ( remember that You have the equipment but We do not know what You have, though explained in the past, ASSUME that We do not know anything about Your system)

Nando

[solarengineer]

Hi Nando, thanks for the reply. Here is a section of the compressed graph, about 60 seconds worth at 1 sample/s

The charger used in this test is a standard 10A automotive battery charger.

My wind generator displays the same waveform though so not sure whats going on.

the battery is an older battery that i use for testing and i was wondering if it was varying absorbtion rates happening quickly or something. The graph never is steady during charging, but when a DC load is activated the graph is steady!

Any help is greatly appreciated.

Thanks

JAmie

[stephent]

Your sensor is reading the "ripple" of either charging source (genny or grid charger) both aren't filtered. You are reading the amps through a wire aren't you?

[Flux]

Stephent has hit the most likely cause. Your 22,000 uF should have removed most of the ripple, but the capacitor must be before the current sensor, if it is after it then it will be worse.

You don't say what is powering the data logger, if it has its own batteries then ripple seems the only reasonable answer, but if you are powering it from the main battery then many things can be happening.

I prefer to log voltage derived from the current so that you can filter the current signal to mean value. You may not have that facility.

Flux

[solarengineer]

Hey Flux, the datalogger is powered by a seperate 9v battery. The problem i'm having with the peeks is that i want to total amp hours in and out and at a sample rate of 1 second i fear i may be missing some spikes between samples resulting in an inaccurate picture.

at some point i want to power the logger directly from the batteries.

thanks everyone for the help!

jamie

[solarengineer]

BTW sorry i forgot to mention, the current sensor i a hall effect type!

[Nando]

Jamie:

There is a discrepancy because the sampling is time fixed and the input is time variable, and the window chosen allows for limited peak measurement.

When the system is loaded every thing is OK because You are stabilizing the wind mill for one ( You are placing a semi brake to it, allowing more uniform output power).

The response of the Hall Effect if too fast for You, slow it down by placing a low past filter in the output, for a wind mill 2 to 5 Hertz low pass is OK, all depending on the response time of the wind mill.

At average wind velocity, what is the generated frequency of the wind mill ?.

Divide it by 3 or 4 and that could be the Current detector low pass filter -- I am guessing here, because I do not know your over all system parameters.

Nando

[Flux]

batteries work on average current so you will be ok to filter but you must do it before the sampling to make sense.

If your hall sensor has volts out then it is easy, if it is a current balance type then you will need to convert to volts with a load resistor before you filter.

Ah in and out will give you a rough idea but without dealing with charge efficiency and Peukert effect you will not get an exact answer so perfection in the current measuring may not be that critical.

Flux

[solarengineer]

Thanks Nando, I put a low pass filter on the output of the hall sensor using a 100k resistor and a .1uF capacitor. it is a 10hz low pass filter.

This seems to have cleared things up incredibly! i still get ripple on top of the waveform but it way down from what it was. I will try a 2hz filter as you suggested I just have to calculate the resistor & capacitor values next!

Again, thanks everyone for your help!

Jamie

[Nando]

You will have ripple always.

The reason is the crest of the 3 phases that give You 6 ripples and since the frequency is sometimes low and variable, so the ripple.

The 100k and 0.1 UF is 15.9 Hz ( 3 db point)

Do not put a 2 HZ filter unless You know what you are doing with the over all system.

Tell me the frequency at cut in and the RPM the I let You know what filter to use, also let me know the peak power and the furling point (RPM)

Nando

[solarengineer]

Hi Nando, it cuts in about 150rpm.

the peek power is 1000w, this is the best ive seen from it. its a 10ft 9 coil 12 magnet per rotor with dual rotors machine. 37 turns 2 in hand #14

the furling point rpm im not sure of.

thank you!

jamie

[boB]

Can we see some more graphs when you think you have a good filter installed ??

This is ingeresting and I'd love to see more of this.

boB

[Nando]

I would like to receive more info on your data logger, why don't you connect with me directly, my email address is in the heading though change -at- and -dot- ( anti-spam) for proper sending.

Nando