SX Resetting itself in Electrically Noisey Machine Environment

Tom M

I'm having a problem with SX Resetting itself in an Electrically Noisy Machine Environment.

I'm using SX48 and SX28 to control a large hydraulic press.
I'm activating two large solenoid valves that control the direction of a hydraulic cylinder.
This is a circuit that allows control of the Up and Down Limits of the Cylinders Ram.
It utilizes a Glass Scale for Position Feedback.

When I first tested the setup, I found that the SX would reset almost every time I turn off one of the solenoids.
So I've redesigned with the following ...

Added Optical Isolation to Inputs from the Glass Scale.
Added Optical Isolation to the two Relay Drivers that in turn switch the solenoids On/Off.
All circuitry is Isolated from the Chassis.
All Cable Shields are connected to Chassis.
AC Power Ground is connected to Chassis. ( By Chassis I mean the Enclosures. )
The Machine, its Electrical Box, the SX Control Box, and Relay Box are all grounded together.
The 5V DC SX Power supply supplied from a wall transformer and regulated with a 3-terminal regulator with a large Filter Cap.
The Glass Scale is powered from it's own wall transformer.
And the Relays are powered from their own wall transformers.
None of these circuits are connected in any way.
The Wall Transformers Outputs and all wiring to and from the control circuit, wrap through Large Ferrite Cylinders.
All IC's and the SX are Bypassed by 0.1uf caps.
The SX uses a 16 mhz crystal.
The SX Control Unit is in it's own Die Cast Aluminum box, separated from the box containing the Relay Drivers and Electrics.

The Reset Rate has decreased to once in every hundred or so operations.

I need to get this to zero resets, for safety reasons if nothing else.

Any recommendations or anyone have experience in industrial machine control?

Thanks All.

Tom M.

Bean

First run the SX from batteries to determine if it's noise on the power supply that's causeing the resetting.
It's got to be noise getting in from somewhere.

Bean.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
“The United States is a nation of laws -· poorly written and randomly enforced.” - Frank Zappa

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
www.hittconsulting.com
·

Tom M

Good Idea Bean.

Thats the first thing I'll try.

Tom M.

nick bernard

your coils could be producing a flyback voltage. try placing a diode across the dc coils; anode to negative, cathode to positive.

also your coils could be producing an emf. if the flyback diode does not suffice, try placing the control circuitry farther from the coils and/or in a grounded metal enclosure.


its good to be back...
nickB

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Let each man exercise the art he knows.
Aristophanes, Wasps, 422 B.C.

SamTheMan

try to use a smaller bypass cap(S). 0.01uf

william chan

Did you connect the aluminium box that encloses the SX to the SX's ground?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my

Jim Richey

That's a good idea Bean. It has the symptoms of a power supply voltage drop,causing a reset.
Just a thought.

Best regards

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Thanks, Parallax!

Tom M

Thanks for all the quick responses!

Mr. Chan: NO. the SX Power Ground is isolated from the Enclosure.
All the various Enclosures are grounded together and connected to AC Power Ground.

SamTheMan: Would I Parallel all my 0.1uf Bypass Caps with the 0.01uf or Replace them?

Mr. Bernard: Using Diodes across the Relay Coils cause a Long Dropout Delay, so I can't use them.
The relays are located in a separate enclosure about 3' away.
Additionally I'm using Optically Isolated Relay Drivers between my SX Board Outputs and the Relays.
The Relay Driver and Relays, have their own power source, isolated from everything else.

Mr. Richey: That's making the most sense. I'll try the batteries on Monday.
If that is the case, I'm tempted to Replace the Wall Wart with a Switching Regulated PS.
Think that would help with a the voltage drop.
I currently using a 4800uf filter cap.

I have another Question.... Searching the archives I found where someone posted a PDF file on Designing for
this type of problems. I mentions using "Suppressors" on the Relay Contacts to prevent Arching.

What are Suppressors? Are they a single component or a circuit of some sort?

Tom M.

SamTheMan

just add the .01uf cap in parallel . there is no need to replace any caps

william chan

Tom,

Have you tried running with a battery?
I tried last time, but it didn't work for me. The SX still resets or go crazy.

See http://forums.parallax.com/showthread.php?p=567019

Did you enable the watchdog? Are you using hs1, hs2, hs3 or xt2 for your resonator?

Guenther,

Tom's experience is proof that having sufficient bypass caps still does not solve the problems with sparks nearby SXes.
A few other boards I designed with very close bypass caps also has the same problem with sparks.

My experience now tells me that the further the SX is away from the sparks, the more immune it is to them.
I would recommend 2 meters as a starting point and slowly move further away if it still resets.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my

ellizard

HI All,

Just a stupid suggestion,

It happened to me, the unused I/O pins caused any kind of strange behaviour, until I setted them output low and shunted to gnd with a 470 ohm resistor.

In the environment there was tree actuators for the building elevator that emitted a lot of electrical noise, both trough the line and trough the air (RF).
the shielding and grounding of the case also contributed to reduce the interferences.


regards
Stefano

SamTheMan

Was your problem fixed?
what was the correct solution?
I'm going to build a project which will work in a noisy environment, and I want to learn from yours

Thank you


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔