Stray current causing IC chip crashes detection techniques?

takao21203

Joined Apr 28, 2012
3,702
Correct reply Mr Chips but out there its not always handy to carry an equipment park, and for this system, certain things are lacking obviously.

If you dont know about AC motors problems then certainly its a shoot in the dark.

As for heavy industrial, you almost always have transients present, not just local but also on the grid, as heavy loads are switched all the time, compensation as well filtering maybe isnt done properly, surge protectors may be present or not.

You need to check with the person responsible for the whole installation.

Installing a small SMPS (eletronic transformer) as well a filtering choke, good new caps and a choke before the controller also cant be wrong and obviously debug the local motors.

If you have some AC offline chips or transformers, order motor filters, something can be done within a few days.
 

MrChips

Joined Oct 2, 2009
34,807
I will point out one thing I would have rejected as a supervisor:



Notice this xtal and two capacitors. Presumably these are components for the oscillator circuit on the large 40-pin DIP chip which I assume is a microcontroller. The xtal and loading capacitors are located too far away from the pins of the mcu.

How many other similar mistakes are on that board? Who knows?
 

MrChips

Joined Oct 2, 2009
34,807
When professionals and people with experience tell you to do something, you ought to go ahead and do it.

Examples:
  1. ESD prevention procedures
  2. Power supply decoupling capacitors
  3. Proper decoupling capacitors on 3-terminal voltage regulators
  4. Minimum trace lengths at xtal components
  5. Guard ring around high impedance op-amp inputs
  6. Stay within device specs
  7. Heavier traces for larger currents
  8. Ground plane
  9. Sufficient heat sink
For example, do you see any power supply decoupling capacitors on that board? I don't.
 
Last edited:

Thread Starter

steward

Joined Jan 21, 2011
53
well, large motors switched by relay, cables to them, there you have it.
Still you can get pretty awful EMI and transients from a motor (which basically is a large inductor), connected with cables of some length. The cable and the motor coils can pick up transients, and the motor generates EMI and harmonics. You need a filter before the motor.
Thanks for all the replies & the critique of the PCB. From my standpoint, I want to first go after the least expensive external ways of correcting the filtering or shielding issue. To answer the Czech person ... I am just looking to improve on the failure rate of the controls of 16 commercial dryers in our laundromat.

Step one: I can have the contacts just switch a to be added higher rated external relay that will go to the motor that is located about 4 feet away from the controls. Steps following: Route motor power wires farther away from the PCB & possibly twist or use a proper grounded shielded cable for the motor itself. If I used the filter? on the way to powering the motor like Takao suggests ... what filter would some of you recommend for a 1/2 HP single phase 120vac motor.

Like I said, the following cut down on the frequency of the problem: http://www.newark.com/te-connectivity-corcom/6eeb1/rfi-power-line-filter-6a-380ua/dp/70K9990 That Corcom 120vac filter was put inline to all the control circuit including the PCB shown ... was this my best choice? ... or could I have chosen a better filter at that strategic location. Since I got the sixteen brand new filters off of Ebay for a couple of dollars a piece ... I figured it was worth the try.

Mike

P.S. An added higher amp rated external relay to the motor will also probably solve the reality of the occasional sticking of the contacts on the relay.
 
Last edited:

MrChips

Joined Oct 2, 2009
34,807
Now that you have provided more information, we now know the board is a timer controller from a commercial drier in a laundromat.
If you are looking for the easy solutions, do the following.

1) Install Corcom power line filters at the AC mains input of controller board, something like this:



http://www.newark.com/te-connectivi...a/dp/52K3297?MER=PPSO_N_P_EverywhereElse_None

2) Install a few 0.1μF capacitors across the power supply line on the microcontroller.
Show us a photo of the other side of the board and we will be able to make suggestions as to where on the board to place the capacitors.
 

takao21203

Joined Apr 28, 2012
3,702
you can try a 22uH to 100 uH choke just after the 7805, disconnect the 3rd leg and insert it.

A laundromat facility? There you have heaters switched while other motors tumble.
You probably only see lights go a little dim for a moment, but some motor can be in an actual state where it has high momentary reactance and will produce a nasty spike if there is a fluctation.

The whole installation as such is a machine which with some frequency, produces transients the chip/board cant deal with. If the relays are on the margin they will also produce erratic waveforms.

you can connect 0.22uF across the relay contacts, see if that improves the problem.

There are many such small inexpensive measures you can try.

Having controllers on a supply line with heavy or multiple loads always requires filtering, as I said, I often get rid of problems with chokes (inlined after the voltage regulator).

The problem isnt just the controller supply, I found some interfering waveforms can influence the oscillator so it behaves erratic at the wrong time- the controller will essentially hang. So even with the best decoupling capacitors, you still get this interference- it simply ignores capacitors.

controllers have multiple instruction phases and between of them sometimes require exact timing.

you can install filters anywhere and still get some weird EMI travelling across them, and disturbing the oscillator inside the chip.

All kinds of cables work as a good antenna to pick them up- you need to shield cables going to controller boards if there is noise on the supply grid. Not all the way, just a little, twisting a wire around it and earth it.

Harmonics and transients from motors usually are just in the KHz range- not the same as HF noise from somewhere, but often its a combination of unwanted waveforms which, as you described, just sometimes adds up in a way that has an effect. Noise from bad contacts can be high frequency for a short moment- you'd be quite busy trying to measure it down to details.
 
Top