Controlling on/off of a Ulka pump with 5V

Thread Starter

Ross Satchell

Joined Jan 2, 2017
45
So now that I've got that sorted out, I want to use the triac circuit instead of a electromechanical relay, because I figure solid state will have a longer lifetime than electromechanical parts.
So I am now using an PS2505 and a Schmidtt trigger with hardware debouncing which is connected to an Arduino input pin.

When I activate the brew switch on the machine, I have written a little Arduino sketch to pulse the pump on and off for 1 sec each, 3 times via the triac circuit from the datasheet on Post#5.
I've connected a LED to the output pin that is driving the pump to confirm that the code is working as intended.

When the pump comes on it doesn't pulse, but instead just stays on.
I read somewhere that there is a latching effect on triacs. Does this sound right?
 

#12

Joined Nov 30, 2010
18,223
I read somewhere that there is a latching effect on triacs. Does this sound right?
Yes. It's the same latching effect as a relay contact. Once you turn it (a relay or a triac) on, it stays on until you stop applying the signal to turn it on.
Then a relay takes about 0.01 second to release and a triac takes until the current drops to nearly zero because the voltage wave of the power supply crosses zero. If you are in a 60Hz country, that could take as much as 0.008 seconds. In a 50Hz country, it could take 0.01 second, very similar to a relay.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
45
Yes. It's the same latching effect as a relay contact. Once you turn it (a relay or a triac) on, it stays on until you stop applying the signal to turn it on.
Then a relay takes about 0.01 second to release and a triac takes until the current drops to nearly zero because the voltage wave of the power supply crosses zero. If you are in a 60Hz country, that could take as much as 0.008 seconds. In a 50Hz country, it could take 0.01 second, very similar to a relay.
So if I understand you correctly, when the arduino output pin goes high, which is connected to pin 1 of the MOC3010 the triac latches.
Then when the arduino output pin goes low, the triac should release.

However when the arduino output pin goes low, the pump keeps going.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
45
So the arduino comes on and the bat16 never shuts off?
Correct, the arduino output goes on for 1 sec, off for 1 sec as long as the brew switch is closed.
The BTA16 never shuts off, so the pump keeps going.

Just to be sure I disassembled the section of breadboard with the MOC3010 and the BTA16 and connected it all up again as per the datasheet, albeit a few resistors with slightly different values.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
45
I tried swapping the wiring between A1 and A2 pins on the BTA16 and it worked!

But I don't understand why it worked. Since the triac basically has 2 SCRs inside it, it should conduct both ways.
Any ideas why this would be the case?
 

#12

Joined Nov 30, 2010
18,223
Any ideas why this would be the case?
The gate only has a relationship with A2. It doesn't speak to A1. You reverse A1 and A2 and the gate is trying to talk to the wrong end. Meanwhile the circuitry that is designed to connect to the power voltage end is communicating with the gate, and it shouldn't be!
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
45
The gate only has a relationship with A2. It doesn't speak to A1.
So in the schematic in the datasheet, which one is A2? The top or the bottom?

Also, can I use this to switch a heating element? From what I've read, I would need some circuitry to do zero crossing.
Is this correct?
 

#12

Joined Nov 30, 2010
18,223
which one is A2? The top or the bottom?
A2 is the terminal drawn next to the gate terminal. If you can't find the terminal drawing on the datasheet, gate and A2 are the only two terminals that will pass current to each other as in a diode test setup.
can I use this to switch a heating element?
yes.
I would need some circuitry to do zero crossing.
Is this correct?
no, but it doesn't hurt anything.
The MOC3042 which I posted a datasheet for already has a zero crossing circuit in it.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
45
A2 is the terminal drawn next to the gate terminal. If you can't find the terminal drawing on the datasheet, gate and A2 are the only two terminals that will pass current to each other as in a diode test setup.

yes.

no, but it doesn't hurt anything.
The MOC3042 which I posted a datasheet for already has a zero crossing circuit in it.
Fantastic!

Thank you very much for your help, its greatly appreciated :)
 
Top