Pages: [1]   Go Down
Author Topic: noise  (Read 344 times)
0 Members and 1 Guest are viewing this topic.
0
Offline Offline
Full Member
***
Karma: 0
Posts: 151
Arduino rocks
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

I've got an arduino that I built as a standalone, and I'm wondering why is it when a pump status comes on, I get a lot of noise on the temperature sensors.
When the pump status point is off, there's no problem?

The pump status input to me is soldered to a PIC pin which switches a relay on / off.
The temperature sensors are simple thermistor, resistor divider to input.
The ground path is common all ground requirements are soldered back to the one location.
Logged

Manchester (England England)
Offline Offline
Brattain Member
*****
Karma: 597
Posts: 33314
Solder is electric glue
View Profile
WWW
 Bigger Bigger  Smaller Smaller  Reset Reset

You get noise because the motor generates it. You need to reduce the noise generated with snubber circuits, reduce how it can get into the measuring circuit with supply decoupling and reduce its effects by smoothing on the sensors using capacitors.
Logged

Valencia, Spain
Offline Offline
Faraday Member
**
Karma: 143
Posts: 5309
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Power supply/decoupling?

Logged

No, I don't answer questions sent in private messages (but I do accept thank-you notes...)

0
Offline Offline
Shannon Member
****
Karma: 200
Posts: 11672
Arduino rocks
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

When you say "ground path is common" I hope you don't mean that.  Common star ground is specifically to
avoid any common wiring between diffferent uses (particularly sensors and high-current paths).  There is
a common point, not path.

Sensor cables shielded?  analog pin decoupled?
Logged

[ I won't respond to messages, use the forum please ]

0
Offline Offline
Full Member
***
Karma: 0
Posts: 151
Arduino rocks
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

What do you mean analog pin decoupled?
Sensor cable isn't shielded, just security system cable.

The ground point is all soldered back to a single ground point.

I think from what you are saying, I should have multiple ground paths to the single point - i.e. the ground reference for analog input should be direct to the ground point? and independant to the chip's supply ground.

It is just thermistors, but the graph I've been getting is bouncing like crazy when pump is running and stable without.
Logged

0
Offline Offline
Shannon Member
****
Karma: 200
Posts: 11672
Arduino rocks
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Try decoupling the analog pin with 10nF or so (between pin and ground - thus shorting a lot of the pickup straight
to ground - your thermistor signal is very low bandwidth I assume).

Google is up today BTW  smiley
Logged

[ I won't respond to messages, use the forum please ]

Manchester (England England)
Offline Offline
Brattain Member
*****
Karma: 597
Posts: 33314
Solder is electric glue
View Profile
WWW
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
I should have multiple ground paths to the single point - i.e. the ground reference for analog input should be direct to the ground point? and independant to the chip's supply ground.

Look at the ground path from the motor and the sensor. You need the ground path of the two not to be common untill you reach that common point, this should be at the arduino ground pin.
Logged

the land of sun+snow
Offline Offline
Faraday Member
**
Karma: 158
Posts: 2882
View Profile
WWW
 Bigger Bigger  Smaller Smaller  Reset Reset

Quote
The pump status input to me is soldered to a PIC pin which switches a relay
on / off.
Since this is an Arduino forum, we should also recommend that you can the PIC, and
switch to Arduino.

Besides what everyone else mentioned, you also need a reverse diode across the relay
coil, to prevent inductive spikes on switch-off.

Also, for situations like this, you need to implement several of the common fixes to get
the best results = diode across relay coil, snubbers on motors, star-grounding, low-pass
filters on ADC channels, extra decoupling [bypass] caps on the microcontroller. Each of
these solutions helps in its own way, and there is no magical one-thing-fixes-all.
Logged

Pages: [1]   Go Up
Jump to: