Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Problems with power supply in HC11 circuit

Status
Not open for further replies.

ZeleC

Full Member level 5
Joined
Dec 18, 2002
Messages
261
Helped
3
Reputation
6
Reaction score
1
Trophy points
1,298
Activity points
2,173
power supply problems

hello ,help needed guys :(
i recently made a hc11 , expanded mode, circuit that controls a AC motor and a couple of valves commanded by 24V relays.thats good till now .
The problem is when the command of the relay on the motor starts or syops the program goes crazy and the Static ram is erased ( i have the power supply and the motor on the same 220 V line.)
i wachted the 5V of the that supplies the uC on an oscilloscope , when the motor starts or stops it is triggered and then chaos...
so what is a problem ,is it from noise caused by the motor and what is the best solution?
thx in advance
 

Hi,

One of the most important aspects to consider is the snubber circuit, Which is a capacitor (Usually 2uF/400V) series with a resistor (100R/0.5W) , this combination must be connected in parallel with the driven motor to supress the transient spikes reflected to the line, this will prevent the power supply unit from going mad and thus, the uC won't suffer an unstable power lines. This is an extremely important issue.
 

if a huge capacitor 100uf or grater will not work you can use
a pi filter that build with capacitor, coil, capacitor in fron of
the 5v regulator.

kobik
 

Hi,

although i was into programming C51 the reasons for these crashes are basicly the same.

If you have a printed circuit board you should use groundplanes to reduce the impedance of ground connections. Make all connections on the board and all external connections as short as possible. Introduce ferrit beads to external connections and capacitors to ground. So energetic pulses as they appear when switching on a AC motor or any other device will not be introduced to your circuit. If you have not put your circuit put into a metal box, the whole is even more sensitive to electric pulses.

And very important, (you wrote you have measured on the circuit with a oscilloscope) use as few cable connections to the other devices as possible. The are antennas for high frequencies and will introduce additional spikes to your circuit.
If you use a reset circuit with a pin for external reset, make shure no disturbances are introduces via this pin.

In the end, here is a technical Info of Mitsubishi Electric called "Noise suppression technology for microcomputers", seems to be older but never the less quite actuel in tips of how to keep interferences away from µC.

hope it helps,
aoxomox
 

I know such problems for a long time and always have to fight against them.

Most important things:

When switching 230V with relais or whatever a snubber circuit as described above is very important, nowadays i use VDR resistors with very good success for this.

Connect all unused MCU pins to VCC or GND (you must of course programm these pins to be inputs).

All signals that come from the outside of your board should be connected via small ferrite beads.

All used inputs of the MCU should be bypassed with small capacitors (1 to 100 nF does not do any harm in most situations) as near as possible to the input against GND to capture spikes inducted into the connected cables.

Last but not least bypass the MCU power supply with capacitors (100 nF SMD) as close as possible near the MCU.

In most cases a watchdog timer can at least reset the MCU in this situations.

This is a long learning process, hope this helps.

best regards
 

thx guys for ur help . im sure that im learning a lot from all of u
but today i noticed another problem ,although i know that the snubber circuit and the pi filter will help(havnt tried them yet ),
i m having problems even if i dont connect the motor, when the 24v relays turn on it causes some problem on a 4 digit lcd sceen that im using ,it makes numbers dissappear or change ,but does not make the uC go crazy . i noticed that the 24V connected to the relays is passing near the databus . My question is : does that cause a problem , or the problem is caused by the power lines.
thx a lot
 

Should not be a problem ...

Another question do you use reverse polarised diodes on each of your relais to kill the voltage that is inducted whenever the relais are switched on or off ??

If not this could also cause your problem.
 

thx c-man for ur help
well at first a was not using reverse diode but yesterday i added them add it did helped a lot, and i added a filter on the power lines .now the system is much better , the uC is not going crazy anymore , but still there is little noise on it , because when the motor stops or start some didgits on the 7 segment lcd dissappears, so i thing that i still need a pi filter before the 5v
regulator as kobik suggested.
and is there a kind of filter that can be made with op amps that cut the noise and replace the RL pi filter(because i don t want to use a coil )
thx guys .
 

I think the remaining problem of digits disappearing return to your ground scheme. I had a similar problem but with 7-seg one.When I invesyigat it I found that the ground of 7-seg section lay at the end of a ground line that hold high current in some cases of operation, so the ground of the circuit is higher than other circuit ground so it see the control signals '0' instead of '1'.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top