Cleaning Signals from 300 ma sensor

Go To Last Post
9 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I just got my new scope.

I am breadboarding a project using an external 14.7456 crystal oscillator. as I want to use serial to communicate.

 

The top is without the crystal running

The bottom is with the crystal running and the processor not doing any pwm

 

The frequency is sometimes maxed at 200mhz 

 

How does one go about trying to read 0 to 300 ma with all that going on?

 

 

 

Attachment(s): 

This topic has a solution.
Last Edited: Sun. Oct 6, 2019 - 02:58 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Breadboards are notorious for being ferociously noisy.  Just using perfboard and soldered wires should make a huge difference.  There are lots of other things that might be acting up too - use lots of power decoupling caps, note carefully where you put the 'scope ground, &c.

 

You might want to spend some time on Google looking up 'filters'.  200MHz is easy to get rid of.   How fast does your sensor need to respond?

 

If you want more specific help you'll want to have a schematic handy, and probably a picture of the breadboard as well.  Meanwhile - just what are you measuring there?  The power rail?  The signal out of your sensor?  Where's the 'scope probe?  Where's its ground?  among other things...

 

  S.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How are you measuring it? What you’re seeing is noise and may/may not be an accurate representation depending on the exact method you are using. Where you place the ground lead has a large effect.
Note mhz is millihertz. You meant MHz which is megahertz.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


 

Grab a piece of blank copper clad PCB...solder your components there & you will have a nice gnd plane....generally nice and quiet.  The wide copper plane has low inductance, very helpful in truly grounding things (such as your xtal caps).  Thin wires don't compare.

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Sat. Oct 5, 2019 - 07:13 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Scroungre wrote:
Meanwhile - just what are you measuring there?  The power rail?  The signal out of your sensor?  Where's the 'scope probe?  Where's its ground?

+1

metron9 wrote:

The top is without the crystal running

The bottom is with the crystal running and the processor not doing any pwm

Where exactly you are measuring? 

.

Consider AN2519

See Figure 2-1 and Figure 2-2

You must use decoupling cap as close as possible to MCU.

.

avrcandies wrote:
Grab a piece of blank copper clad PCB...solder your components there & you will have a nice gnd plane....generally nice and quiet.  The wide copper plane has low inductance, very helpful in truly grounding things...

Nice method for prototyping! 

Majid

Last Edited: Sat. Oct 5, 2019 - 08:24 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I understand that Candyman developed this technique for mounting through hole components shortly after he broke his last drill bit!  cheeky

 

Nice job on those boards!

 

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I recognise it as a variant of the dead bug technique.

 

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0



Nice job on those boards

Well, I can't claim those internet samples...Bob Pease & Jim Williams were famous for espousing this method.  I use sockets & glue down smt adapter boards (or use a dremel tool to make some pads)....makes quick work to wire something up.   Here is one I did

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

This reply has been marked as the solution. 
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Wow do I feel silly nut learned a lot in the process

I'm reading an EPM-4001 sensor using two led's ELM=4002

Using a 14.7XX external crystal oscillator on the board

I had led light bulbs and my 30 inch monitor blaring the area giving me lots of noise

but the WIRE Antenna I attached to the oscillator was the real dumb thing

I used this board to test, adding a 5v regulator and a battery pack to eliminate noise

I was using a 5V regulated wall wart but that turns out to be just fine as well even when it goes through the low drop out regulator.

 

Now I have 12mv ripple and with red and uv led's off I get down to the 12mv ripple and with either on up to 330 mv

 

Now I will finish the circuit and add an active low pass filter and op amp voltage gain to get the voltage where the ADC can read it.

 

Thanks for the help, links, and just being there. 

 

Attachment(s):