Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Electrical noise generated by sealed lead-acid batteries

Status
Not open for further replies.

a.down

Newbie level 1
Joined
May 25, 2018
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
17
Good afternoon. I'm designing a low-noise analog front end. Nothing fancy, i am shooting for a 0.5uV baseline noise after digitization with a +/-5V dynamic range. All within the capabilities of the latest ADCs like the LTC-2508 IF i can provide a quiet enough environment. The application needs to be battery powered, which i figured would probably help my cause, eliminate any switching regulators and stick to really low noise LDOs. I'm chasing some noise around and i noticed that if i measure the terminals of my lead-acid battery directly with a tek TBS2000 scope i get alot of noise. This isnt an isolated scope, so the noise floor already looks like 10mVrms across its 70MHz bandwidth, but with the probe on the battery i get almost 30mVrms across 70MHz. That works out to something like 3.5 uV/sqrt(Hz), i was under the impression the chemical noise of an SLA was about 1000x quieter than that.

Anyone else stumble across something similar?

Thanks kindly,
-Alexander
 

Hi,

0.5uV in a 10V range makes a 1:20,000,000 resolution.
This means at least 24 bit (noise free) resolution.
It's a hard job. (I doubt I could achieve this, although I have quite good experience with high resolution circuits)

Now it mainly depends on output data rate you want to achieve.
I assume a delta sigma ADC with high oversampling rate is the best choice.
For sure it's not only the ADC...
* your input signal needs to be noise free, too.
* and - many designers forget - the reference voltage needs to be stable and noise free in the same magnitude.

LTC2508...is at it's specification limits, and this only at the highest oversampling rate.
Very careful analog circuit design is essential:
* the simplest task will be to choose a low noise voltage regulator (doesn't need to be 23 bit stable and noise free)
* use effective filters for the power supply to suppress HF noise (then I don't think that the battery noise will influence the conversion)
* any Opamp circuit needs to be very carefully designed
* choosing low noise resistors and low tempco
* use guarding technique for the analog signals
* very symmetric design (thermal) is essential to avoid thermocouple voltages to create (variable) DC offset.
* you need a very good GND concept
* decouple the complete ADC curcuit, maybe use a shielding case across the analog and ADC section
* decouple even the data lines with resistors in the 100 Ohms range. Keep them far away from analog signals.

****
How do you decide to process / transmit the digital data?

Klaus
 

guess what happens to the noise on your battery when you start drawing a load off it, or when re-charging... they are inherently noisy, try Li-Fe-PO4 for a bit more quiet ...
 
  • Like
Reactions: d123

    d123

    Points: 2
    Helpful Answer Positive Rating
May be it's the battery. But before blaming it, I would expect significant measurements verifying that it's not just the load current ripple or some measurement artifact, e.g. by spectral analysis.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top