Post by thread:mclr on pic16f84

hello I am working on a project that involves controlling electric motors through relays with a PIC16F84. The motors and relays have the appropriate back EMF suppression circuits, but I find that the PIC occasionally resets when the motor and/or relay swithes on/off. I suspect that this has something to do with the MCLR pin on the PIC. Can anybody advise a way of making the pic resistant to this type of resetting, or possibly disabling the mclr pin? Any assistance that you could give would be most appreciated. Mark

> Check that you do not connect common ground of your PIC and motor. If so you need to separate them. If they separated then the problem is in relay's inductance.Check if you have proper decoupling capacitors near to PIC (it was discussed in the List) Tha problem may occure if power rises above 7 V due to any reason. And this can not be disabled. Check your power supply engineering - DOES IT HAVE ENOUGH POWER TO SUPPRESS EMF? The better idea is opto coupler with triac to control a motor that fully decoupled motor circuit from control circuit. Regards, Vassili

Check that you do not connect common ground of your PIC and motor. If so you need to separate them. If they separated then the problem is in relay's inductance.Check if you have proper decoupling capacitors near to PIC (it was discussed in the List) Tha problem may occure if power rises above 7 V due to any reason. And this can not be disabled. Check your power supply engineering - DOES IT HAVE ENOUGH POWER TO SUPPRESS EMF? The better idea is opto coupler with triac to control a motor that fully decoupled motor circuit from control circuit. Regards, Vassili

If it don«t help tying MCLR to Vdd there is probably severe noise in chip supply. 1) Are you certain that ni spikes etv are conming in return fron outputs, or overshooting the inputs? Maybe try inserting resistors close to chip in serial to the suspected pins? 2) Are the PIC supply stable and well decoupled? 3) Or, I dare say, there is a program error. We all do them... Have you made a very simple test progran that make all output operations, and detect reset? PS We are all interested on avoiding problems of this kind, so please tell if You find anything! DS Regards /Morgan At 22:46 1998-01-25 -0800, you wrote: {Quote hidden}>hello > >I am working on a project that involves controlling electric motors >through relays with a PIC16F84. The motors and relays have the >appropriate back EMF suppression circuits, but I find that the PIC >occasionally resets when the motor and/or relay swithes on/off. I suspect >that this has something to do with the MCLR pin on the PIC. > >Can anybody advise a way of making the pic resistant to this type of >resetting, or possibly disabling the mclr pin? > >Any assistance that you could give would be most appreciated. > >Mark > > Morgan Olsson, MORGANS REGLERTEKNIK, Sweden, ph: +46 (0)414 70741; fax 70331 ============================================================================

In a message dated 98-01-25 05:02:18 EST, you write:

<<
hello

I am working on a project that involves controlling electric motors
through relays with a PIC16F84. The motors and relays have the
appropriate back EMF suppression circuits, but I find that the PIC
occasionally resets when the motor and/or relay swithes on/off. I suspect
that this has something to do with the MCLR pin on the PIC.

Can anybody advise a way of making the pic resistant to this type of
resetting, or possibly disabling the mclr pin?

Any assistance that you could give would be most appreciated.

Mark

>>
Hello Mark
In addition to the other reponses you will be getting about capcitance etc, make sure the diodes you are using for the back emf are fast enough. Most common diodes are fairly slow..

Dave

The PIC chip is very sensitive to noise on MCLR. Keep the line to the MCLR pullup very very short and well away from noise sources. Decouple the PIC directly across power and ground with a 0.1 monolithic capacitor. Ensure that power and signal grounds are separate and join them at the edge of the board and go straight from there to ground. {Quote hidden}> ---------- > From: Morgan Olsson[SMTP:spam_OUTmrtTakeThisOuTINAME.COM] > Reply To: pic microcontroller discussion list > Sent: Monday, January 26, 1998 5:09 AM > To: .....PICLISTKILLspam@spam@MITVMA.MIT.EDU > Subject: Re: mclr on pic16f84 > > If it don«t help tying MCLR to Vdd there is probably severe noise in > chip > supply. > > 1) Are you certain that ni spikes etv are conming in return fron > outputs, > or overshooting the inputs? Maybe try inserting resistors close to > chip in > serial to the suspected pins? > > 2) Are the PIC supply stable and well decoupled? > > 3) Or, I dare say, there is a program error. We all do them... Have > you > made a very simple test progran that make all output operations, and > detect > reset? > > PS We are all interested on avoiding problems of this kind, so please > tell > if You find anything! DS > > Regards > /Morgan > > At 22:46 1998-01-25 -0800, you wrote: > >hello > > > >I am working on a project that involves controlling electric motors > >through relays with a PIC16F84. The motors and relays have the > >appropriate back EMF suppression circuits, but I find that the PIC > >occasionally resets when the motor and/or relay swithes on/off. I > suspect > >that this has something to do with the MCLR pin on the PIC. > > > >Can anybody advise a way of making the pic resistant to this type of > >resetting, or possibly disabling the mclr pin? > > > >Any assistance that you could give would be most appreciated. > > > >Mark > > > > > > Morgan Olsson, MORGANS REGLERTEKNIK, Sweden, ph: +46 (0)414 70741; fax > 70331 > ====================================================================== > ====== >

Andrew Mayo said: >The PIC chip is very sensitive to noise on MCLR. Keep the line to the >MCLR pullup very very short and well away from noise sources. Decouple >the PIC directly across power and ground with a 0.1 monolithic >capacitor. Ensure that power and signal grounds are separate and join >them at the edge of the board and go straight from there to ground. > Good advice. Also, it is desirable to use a fairly low pull-up resistor. I certainly wouldn't go any higher than 1K between Vdd and MCLR. You can tell if this is contributing to the problem by trying different pull-up resistors and seeing if higher values increase the frequency of undesired resets. Ferrite beads can be very helpful; wind a few turns around them and put them in series with power and signal lines. If the resets are only occasional, it can be difficult to tell whether a particular modification has improved the situation. Connect a big relay in "buzzer" mode and power it from the same line that powers the PIC circuit. Hold it physically near the chip and the rest of the circuit. This often broadcasts enough wide-band noise to expose marginal conditions. If you can harden it against such "brute-force" attacks, it should improve the situation. Reg Neale

Thanks to all who gave suggestions about how to fix the reset problems i was having with the 16f84. After trying out a number of ideas with mixed success, i ended up fitting a dc-dc converter and optos on the i/o, which appears to have solved the problem by isolating the pic completely. thanks again mark