Searching \ for 'How to deal with a noisy environment' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/language/index.htm?key=environ
Search entire site for: 'How to deal with a noisy environment'.

Truncated match.
PICList Thread
'How to deal with a noisy environment'
1998\07\11@172318 by Bob Bullock

flavicon
face
My project envolves using a PIC to control a stepper motor.  Also envolved
is a high speed drill, similar to a "dremol".  It seems the drill is
generating huge amounts of noise on the AC lines and is screwing up the
external interrupt line of the PIC as I am getting continous interrupts.
The interrupt line is being used to monitor for a relativly infrequent, but
quite fast transient, hence the use use of the irq line.

I have tried running the drill on a seperate ac breaker, have installed
snap-on chokes on the drill line, inserted an opto-isolator on the irq
input and it still screws up if the drill is running.  It would appear to
be strictly a problem with the noise on the power as the spurious irq's
still occur even if the irq source line is removed.

The power supply is a conventional step down transformer, bridge rectifier
and fixed voltage regulator. I have added lots of filter capacitors of
various small and large sizes to no avail. You can still see the noise from
the drill on the Vcc with a scope, its quite pronounced.

Any ideas on how to clean up the supply.

Thanks!

Bob Bullock

1998\07\11@213102 by Philip Starbuck

flavicon
face
>My project envolves using a PIC to control a stepper motor.  Also envolved
>is a high speed drill, similar to a "dremol".  It seems the drill is
>generating huge amounts of noise on the AC lines and is screwing up the
>external interrupt line of the PIC as I am getting continous interrupts.
>The interrupt line is being used to monitor for a relativly infrequent, but
>quite fast transient, hence the use use of the irq line.

Make sure that the interrupt line is shielded and is not routed near any AC lines.  The interrupt line should be terminated at its source and not allowed to float.

{Quote hidden}

I have had good luck using COMMON MODE EMI INDUCTORS on the line input with a cap across the input to the step down transformer.  Try TRIAD "E-CORE" inductors, DIGIKEY p/n 10523 to 10541 (page 254 of the current catalog) or PANASONIC Line Filters (Page 269 of the current DIGIKEY catalog).  BG Micro ( http:/http://www.bgmicro.com  had some at a good price ($0.99 ea.) on page 18 of their current catalog.

Good Luck, there are those amoung us that make a very good living solving these kinds of problems.

cheers,
Phil

Philip Starbuck
(909) 792-7917

"There are three principal ways to lose money.  Wine, women, and engineers.  While the first two are more pleasant the third is by far the more certain."
                                               -- Baron Rothschild
                                                       ca. 1860

1998\07\11@215008 by Reginald Neale

flavicon
face
>>My project envolves using a PIC to control a stepper motor.  Also envolved
>>is a high speed drill, similar to a "dremol".  It seems the drill is
>>generating huge amounts of noise on the AC lines and is screwing up the
>>external interrupt line of the PIC as I am getting continous interrupts.
>>The interrupt line is being used to monitor for a relativly infrequent, but
>>quite fast transient, hence the use use of the irq line.
>
>Make sure that the interrupt line is shielded and is not routed near any
>AC lines.  The interrupt line should be terminated at its source and not
>allowed to float.
>

Bob:

As Phil says, termination of the interrupt line is important. To minimize
noise susceptibility, the pull-up or pull-down resistor should be as low in
value as possible consistent with power requirements and PIC output
ability, i.e., anything from 250 ohms up will be within the 20 ma current
sink/source ability of most PICS. Should be easy to test the effect if
you're getting continuous interrupts.

Good luck!

Reg Neale

1998\07\12@031145 by Daniel Staempfli

flavicon
face
-----Message d'origine-----
De : Bob Bullock <spam_OUTbobbTakeThisOuTspamUNISERVE.COM>
@ : .....PICLISTKILLspamspam@spam@MITVMA.MIT.EDU <PICLISTspamKILLspamMITVMA.MIT.EDU>
Date : samedi, 11. juillet 1998 23:20
Objet : How to deal with a noisy environment


{Quote hidden}

Hi Bob!

I have got exactly the same problem and made the same trials you did.
In case you get an answer, could you re-direct it to me?

Thank you

Best regards

Daniel Staempfli / HB9IIU

1998\07\12@085221 by Coetzee, Morne

flavicon
face
I think the noise you are experiencing has to do with the 3rd harmonics
generated by your drill. These harmonics are called triplens and can
actualy add to each other in your neutral conductor (instead of being
cancelled out). They are mainly generated by AC speed controllers and
switch mode power supplies. Generaly manufacturers specifify that you
should use input line filters to overcome this problem.

That's about all I know about this subject...maybe someone else has a
suggestion for a good filter circuit.

MornŽ Coetzee


> {Original Message removed}

1998\07\12@103001 by Brian Aase

flavicon
face
It seems to me that the total current drawn by something like a
Dremel tool is much too small to begin worrying about x3
harmonic generation.  Besides, the main bad effect of this
phnomenon is to create recirculating currents in 3-phase
building supply transformers, as well as (as you correctly
point out) energize the neutral conductor thus resulting
in a potential safety hazard.

My experience has taught me that the main cause of the "buzz"
type of interference that results in nasties like unwanted IRQ's
is excessive mid-frequency coupling from the mains supply to
the uP supply.  To isolate the root of this problem, point your
finger straight at the power transformer in the +5V (or whatever)
uP supply.  Regular E-I and toroidal transformers have a
surprising amount of primary-to-secondary capacitance to
establish effective transfer of mid-frequency spikes etc. right
into the Vcc line.

One might think of line filters, ferrite chokes, and the like as good
solutions to this problem, but in fact they are not.  These devices
are excellent at removing RF-type interference, but their effective
range normally does not extend downwards into the kHz region
where this kind of problem is happening.

In my experience (FWIW), the best countermeasure for this kind
of problem is to switch to a power transformer that has less
primary-to-secondary capacitive coupling.  The split-bobbin
transformers such as from Signal, Prem, and a bunch of other
places are excellent in this regard, and give the added benefit of
greater arc-over insulation as well, thus making a safer product
overall.  Since the primary (mains) coil is wound totally on one
side of a plastic barrier, and the secondary is totally on the other
side, coupling between them is on the order of only a few pF or
even less.

Now if you have followed good construction practices elsewhere
in the circuit (i.e. prudent bypassing, EMI filtering, etc.),
suddenly most of the mysterious IRQ's, false input triggers,
A/D noise, and all the other things we have come to love so
dearly will suddenly be gone or at least greatly reduced.  Gee.

Of course if you are using a modular switching supply,
you can ignore all of the above and go talk to your supply
manufacturer instead  ;-)

Just my daily dose of hot air...

> I think the noise you are experiencing has to do with the 3rd
> harmonics generated by your drill. These harmonics are called
> triplens and can actualy add to each other in your neutral conductor
> (instead of being cancelled out). They are mainly generated by AC
> speed controllers and switch mode power supplies. Generaly
> manufacturers specifify that you should use input line filters to
> overcome this problem.

1998\07\12@143747 by Peter L. Peres

picon face
Corcom line filter directly at the drill motor (physically bolted to it if
possible). I will buy one for the jeweller next door soon, I can see his
drill up into UHF on top of the local station, and in every scope and
meter in the lab, and he uses no switching controller ;( 50.000 rpm draw
nice arcs...

Peter

1998\07\12@230901 by John Fimognari

picon face
I think the noise you are experiencing has to do with the 3rd harmonics
generated by your drill. These harmonics are called triplens and can
actualy add to each other in your neutral conductor (instead of being
cancelled out). They are mainly generated by AC speed controllers and
switch mode power supplies. Generaly manufacturers specifify that you
should use input line filters to overcome this problem.

That's about all I know about this subject...maybe someone else has a
suggestion for a good filter circuit.

MornŽ Coetzee

A few suggestions. They may work. I have had similar problems with
electrical equipment. You problem is in the AC supply. Are you using an
earthed power supply? Have you checked you neutral and earth conductors
are at the same potential? In Australia we use an MEN system - main
earth neural - The neutral and earth are bonded at the main switchboard
of every building. An earth - neutral differential can cause it. Try
proper series chokes rather than snapon types on your power supply. A
good quality earthed pi filter may help. Check yor drill's supression
filter it may be defective, or you may not have one installed. This is
usually the problem. A pi filter in series with the drill will confirm
this. Check that you are using a double wound transformer and not an
autotransformer type. Try a double wound transformer with an earth
shield wound between the primary and secondary windings. If you have
good test equipment, conduct tests on the AC mains using a stepdown
transformer and borrow a number of drills. Compare the RFI. Separate
supply circuits won't help. Once RFI is in the mains supply it will
radiate quite a distance, sometimes miles.
Regards,
John FImognari


> {Original Message removed}

1998\07\13@070258 by g.daniel.invent.design

flavicon
face
Reginald Neale wrote:
{Quote hidden}

Hi Bob,
Adding to the solutions above, the noise you are getting from the high
speed drill is going to be inductive spikes produced by the commutator
repeatedly breaking the field in the rotor windings. If this is a D.C.
motor, being driven only one way, then you can probable solve your
problems simply with a diode pointing from motor negative to motor
positive. If you are using bidirectional drive on the motor, then use a
varistor in parallel with motor.   These solutions will act at the
source of the problem.

Additionally, motor startup for a D.C. series wound or permanent magnet
motor involves a high start up current, by using a resistor to power
your logic rails you can retain power in your logic storage capacitors
better during start ups.

regards,
Graham Daniel.

1998\07\14@005310 by Dwayne Reid

flavicon
face
>My project envolves using a PIC to control a stepper motor.  Also envolved
>is a high speed drill, similar to a "dremol".  It seems the drill is
>generating huge amounts of noise on the AC lines and is screwing up the
>external interrupt line of the PIC as I am getting continous interrupts.

<snip>

The problem is most likely coming from the brushes in the drill motor.
These motors are HUGE emi (Electro Magnetic Interference) generators and
radiate all over the place.

There are a couple of things that you can do to reduce the problem.

The first approach is to quiet the motor.  If you can find an old PC power
supply that you can strip for parts, grab it.  Open it up and trace the
wiring from the AC input connector.  This connector should go to a choke
(small square block) with two windings on it - the windings should appear to
be identical - and one or two capacitors.  Jot down a sketch of the
connections and remove the choke and capacitors from the supply.

You need to install this filter network as close to the drill motor as
possible.  The wiring should like something like the following (awful) ASCII
diagram:

       --------------oooooooo---------------
             |                    |
             |        this        |
power        |         is         |             to drill motor
 in     cap ===       the        === cap
             |       choke        |
             |                    |
             |                    |
       --------------oooooooo---------------
You need to ensure that the phasing on the choke is the same as was in the
power supply: usually, one side of the choke is input, the other side is
output.  Don't cross (mix) an input pin with an output pin.

I usually just solder the capacitors right to the choke pins, then solder
the source and load wires on top of the first solder joints.  Be careful and
make sure the joints are solid.

Try this - it may be all that you need.  If not, you will have to spend some
time trying to reduce your circuit's sensitivity to the noise.


This can take a long time to discuss - I'll touch on the highlights.

Reduce the loop area of your sensitive circuits.  Basically, this means keep
your circuit physically small.  Think of the wiring as a loop antenna - the
larger the loop, the more interference it can pick up.

Your circuit has inputs and outputs.  Group common signals together and have
a ground lead for each group.  Twist the ground and signal leads together.
Connect the ground lead as close as possible to where the signal lead ties
to the circuit.  If you have a single lead that goes off to a switch or
sensor, give it its own ground lead.  If you have another lead that goes off
to a LED or relay or solid state relay or opto, give it its own ground or
VCC connection.   If you have a group of several leads that goes to a motor
or stepper driver board, give it its own ground lead, then run a heavy wire
from the driver board to the power supply ground.  In other words, use the
driver board ground as your high current star point connection.

This should give you a starting point.  Let us know how you make out!

dwayne


Dwayne Reid   <.....dwaynerKILLspamspam.....planet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(403) 489-3199 voice     (403) 487-6397 fax

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...