I have been asked to determine why a 12 volt Pizo SonaLert continues to fail. Upon investigation this pizo is wired in parallel with a 1/2 hp 12 volt DC motor. The supply voltage is from a car alternator and it is generating 14.4-14.5 volts. My first concern is the SonaLert maximum is 15 volts. That is very close to the upper limit. Placing a scope on the motor/sonalert terminals showed a 50 volt spike on turn-on. That is probably the cause of the destruction.
I am considering installing a series resistor of 12 ohms and a 12 volt zener (1N5349BG) to limit the spike voltage just to the sonalert. My concern is the power rating of the zener. I calculated the resistor to be 1 watt since the sonalert draws 200 ma. That would drop the supply voltage to 12 volts without any zener current. When the spike occurs I am hoping the 5 watt zener will be able to handle the surge. I measured the duration of the spike and it is about 20 ms. This should not damage the resistor even though it is grossly under rated during the spike.
Any thoughts about this solution or an alternative. I have looked at varistors but a 12 volt varistor has a max clamping voltage of 38 volts. That would not do the job.

Thanks,

 

It is more likely that the 1kV spike generated as the motor turns off is the culprit.


Sonalert only takes a few mA, right?
Just use a series resistor and a shunt 12V Zener.

 

It is more likely that the 1kV spike generated as the motor turns off is the culprit.


Sonalert only takes a few mA, right?
Just use a series resistor and a shunt 12V Zener.

The Sonalert takes 200 ma. I agree, the combination of poor regulation of the alternator and the inductive kick of the motor is producing the spike. The L di/dt was not taken into account when the alert was wired in parallel with the motor.
I did not see a spike on Turn-Off probably due to the fact there is a large flywheel cutter attached to the motor shaft and when power is removed from the motor the Sonalert actually continues to operated because the motor is now a generator. There is a slow decrease in output voltage aas the motor slows down.
Still have the power rating questions of the zener.

 

What Sonalert takes 200mA?

I have several around here. For example, Mallory SC628. Rated 6-28Vdc, 3 to 14mA.

 

I have been asked to determine why a 12 volt Pizo SonaLert continues to fail. Upon investigation this pizo is wired in parallel with a 1/2 hp 12 volt DC motor. The supply voltage is from a car alternator and it is generating 14.4-14.5 volts. My first concern is the SonaLert maximum is 15 volts. That is very close to the upper limit. Placing a scope on the motor/sonalert terminals showed a 50 volt spike on turn-on. That is probably the cause of the destruction.
I am considering installing a series resistor of 12 ohms and a 12 volt zener (1N5349BG) to limit the spike voltage just to the sonalert. My concern is the power rating of the zener. I calculated the resistor to be 1 watt since the sonalert draws 200 ma. That would drop the supply voltage to 12 volts without any zener current. When the spike occurs I am hoping the 5 watt zener will be able to handle the surge. I measured the duration of the spike and it is about 20 ms. This should not damage the resistor even though it is grossly under rated during the spike.
Any thoughts about this solution or an alternative. I have looked at varistors but a 12 volt varistor has a max clamping voltage of 38 volts. That would not do the job.

Thanks,

The TL431 adjustable zener can handle 100mA. Ideally the zener should be capable of absorbing the full load current in case the load is ever open circuit.

You can boost the TL431 current capability by adding a PNP emitter follower in the manner of a Szicklai pair.

 

SonaLert PS-551Q 12 V 200 ma 105 dB

 

SonaLert PS-551Q 12 V 200 ma 105 dB

The calculations I come up with are a 12 ohm 1 watt series resistor using a 1N5349 12 volt 5 watt zener

 

A tranzorb is a zener diode specially constructed to handle large transient currents. For example, A tranzorb in a 5 W diode package is rated for 600 W peak power dissipation. Most zener diodes have a similarly large peak to average power ratio. As a first order approximation, you can calculate in watt-seconds the total energy in the part of the transient that the zener has to handle, multiply by the transient pulse width in seconds (quickie pulse width ratio conversion), and get the average power level over one second. So if the transient area is 50 w s and the width is 20 ms, then the average power is 1 W.

also, the zener diode datasheet might list a max non-repetitive peak current value, and the difference between that and the max continuous zener current is another indicator of a diode's transient capability.

ak

 

Thanks AnalogKid. I've been out of the component selection portion of the industry for so long I've never heard of these devices. I did a search and found one that I think might do the trick. There is a supplier on Long Island that I think I'll call on Monday to verify my selection. There are some specification terms that are unique to this device that I am unsure of. Also the correct selection of the particular device is based on these terms so I think the call is required. It sure looks like this will cover all the problems with the addition of a single resistor and this device.

 

After doing some investigation I selected a 1N6384 transzorb along with a 12 ohm 1 watt resistor. I am hoping the max clamp voltage of 17 VDC will protect the SonaLert since the specs have a 15 VDC maximum. A transzob with a lower maximum voltage has a breakdown voltage of 10 volts which would mean it is always conducting. I don't think this is the way it should operate.
I have seen applications and products that are designed for the automotive industry but they are not as critical of the upper voltage as I require.
Any thoughts on the device selection?

Thanks,

 

Digging a little further the 1N6376 has a lower breakdown voltage. Also, an app note from Vishay stated a protection device such as a fuse is a good secondary protection in case the transzorb shorts and there is no positive fail safe mode.

Ill wait for all the parts and assemble the components. I will need to measure the motor supply stud distance so I can use these to mount the component board. That's after getting all the parts.

 

Digging a little further the 1N6376 has a lower breakdown voltage. Also, an app note from Vishay stated a protection device such as a fuse is a good secondary protection in case the transzorb shorts and there is no positive fail safe mode.

Ill wait for all the parts and assemble the components. I will need to measure the motor supply stud distance so I can use these to mount the component board. That's after getting all the parts.

Digging a little further the 1N6376 has a lower breakdown voltage. Also, an app note from Vishay stated a protection device such as a fuse is a good secondary protection in case the transzorb shorts and there is no positive fail safe mode.

Ill wait for all the parts and assemble the components. I will need to measure the motor supply stud distance so I can use these to mount the component board. That's after getting all the parts.

 

I assembled the little circuit and installed it on the motor studs. I checked the operation and then measured the maximum voltage of 13.5 V. I told the machine owner that I would check the performance after a month of operation. After the month, I'll suggest the change to the machine manufactuerr and see if he thinks it is a viable upgrade.
Thanks for the suggestion of the Transzorb device.