Controlling on/off of a Ulka pump with 5V

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
I am working on a espresso machine project where I want to control a Ulka vibratory pump using 5V. Once I get this working I plan to use an Arduino to turn the pump on and off.
My current understanding of these pumps is that they have a internal diode blocking half of the AC power, thereby throwing the solenoid with each positive pulse. Since they are solenoid based, I have reasoned that they are inductive loads.

I found a few threads here about the Ulka pump, but no-one was trying to just switch it on and off

So far I have tried:
  • Using a zero-crossing SSR to switch the hot wire to the pump that it works as expected, however when in the off state the pump hums.
  • Using a random fire SSR to switch the hot wire. According to what I have read, these should be the SSR to use. However, it did not work and thepump just hummed like before.
  • Using a electromechanical relay to switch the hot wire. The relay didn't work and the pump hummed as before.
I have attached a schematic of how I wired up each of the relays.
IMG_20170102_145209.v01.jpg
I have been reading up on SCR's and Triacs and think that maybe I need to using a SCR, but I'm still a bit confused.
Can anyone point me in the right direction?
 

ronv

Joined Nov 12, 2008
3,770
I am working on a espresso machine project where I want to control a Ulka vibratory pump using 5V. Once I get this working I plan to use an Arduino to turn the pump on and off.
My current understanding of these pumps is that they have a internal diode blocking half of the AC power, thereby throwing the solenoid with each positive pulse. Since they are solenoid based, I have reasoned that they are inductive loads.

I found a few threads here about the Ulka pump, but no-one was trying to just switch it on and off

So far I have tried:
  • Using a zero-crossing SSR to switch the hot wire to the pump that it works as expected, however when in the off state the pump hums.
  • Using a random fire SSR to switch the hot wire. According to what I have read, these should be the SSR to use. However, it did not work and thepump just hummed like before.
  • Using a electromechanical relay to switch the hot wire. The relay didn't work and the pump hummed as before.
I have attached a schematic of how I wired up each of the relays.
View attachment 117934
I have been reading up on SCR's and Triacs and think that maybe I need to using a SCR, but I'm still a bit confused.
Can anyone point me in the right direction?
I'm assuming this is a 110 volt AC pump. If it is what I think it is your right and 1/2 the ac cycle pulls in the plunger and a spring returns it on the off cycle.
So it should work if you just plug it into the wall. I'd try that first. It might be seized up.
Having said that I think all you will hear is a hum.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
I'm assuming this is a 110 volt AC pump. If it is what I think it is your right and 1/2 the ac cycle pulls in the plunger and a spring returns it on the off cycle.
So it should work if you just plug it into the wall. I'd try that first. It might be seized up.
Having said that I think all you will hear is a hum.
Thanks for your reply ronv. Yes, the pump is a 110V AC type.
Sorry, I should have been clearer, the pump has been tested to be working. The hum I am describing is not the sound of the pump running, instead it is an electrical hum, but only when I am trying to switch the hot wire.
This makes me wonder if I should be switching both wires??
 

ronv

Joined Nov 12, 2008
3,770
Thanks for your reply ronv. Yes, the pump is a 110V AC type.
Sorry, I should have been clearer, the pump has been tested to be working. The hum I am describing is not the sound of the pump running, instead it is an electrical hum, but only when I am trying to switch the hot wire.
This makes me wonder if I should be switching both wires??
Hmm, One should be enough. Assuming there are only 2. Is the case grounded?
 

#12

Joined Nov 30, 2010
18,210
Hmm, One should be enough. Assuming there are only 2. Is the case grounded?
The lack of details is eating us up. That said, if it works on 120 VAC just plugged into the wall, it should run with a triac.
Look at page 4 and see a representation of how to use an MCU to drive a triac for a motor.
 

Attachments

#12

Joined Nov 30, 2010
18,210
If your circuit in post #6 didn't work the page 4 circuit isn't going to fix it. It's the same circuit with a snubber added.:(
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
I think I must have made a mistake in wiring last night because I tried the circuit from the datasheet in Post #5 and it worked this morning.

the only problem is that I still get an electrical hum when the pump is turned off and the pump leaks a little.

Perhaps should I be switching both hot and neutral wires?
 

#12

Joined Nov 30, 2010
18,210
How few microamps does this motor use that the current leakage through a triac will make it hum?
Actually, I don't think current leakage is the real problem. There are millions of triac driven motors that don't hum.
It's time to get serious about the measurements of voltage and current to find where this amount of power is getting into the motor.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
The pump is rated at 41W, which at 120VAC corresponds to 342mA, although since it is basically a solenoid the inrush current is likely to be much higher. I just measured 0.530A through the pump when running for about 5 sec.
I just measured the current through the pump on my 20mA AC range and it showed 0.0mA
The voltage across the pump when it is humming is 112VAC, so since its supposed to be switched off there shouldn't be any voltage across it.
I'm guessing you're correct that there is a few microamps leaking and causing the hum.

I tested my electromechanical relay and discovered that it is dead, so I'll go get a new one today. Any suggestions on the current rating? I was figuring 15A should be sufficient, but I don't know how much the inrush current is.
 

#12

Joined Nov 30, 2010
18,210
7 to 10 x the run current for the relay contacts. about 3 amps.
You don't want to put a 30 amp contactor on a 1/2 amp load. That little current might not be enough to reliably clean the oxide layer off the contacts.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
So I went & got a new relay Sanyou SMI-S-212D
https://www.radioshack.com/products/radioshack-dpdt-5a-250v-relay

The package it came in listed it as a 12VDC DPDT, the datasheet is in Chinese.
https://www.dz863.com/datasheet-8222779663-SMI-S-212D_Compact-Pc-Board-Power-Relay-Solid-State-Relays/

I also found listed here as 12VDC input DPDT:
http://at.ur.ru/comp/relays/index_3.shtml

I put 12VDC across the low voltage side and could hear the relay clicking, however when I connected 120VAC across the high voltage side nothing happened.
Also I could still hear the electrical hum when the relay did not have any voltage across the input side.
Out of curiousity, I metered out the relay while it had 12V across the input side, it was open circuit across the other side, I'm not sure if that was because my meter didn't pass enough voltage across it, or if I have a faulty relay?

Interestingly, this is the same result as the first relay I used yesterday, so maybe that relay wasn't faulty and I'm doing something wrong?
 

#12

Joined Nov 30, 2010
18,210
I connected 120VAC across the high voltage side nothing happened.
Do you mean you connected the relay contacts directly to a power cord and activated the coil?:eek:

This is getting nowhere fast. I think you should post some drawings of how you are wiring things together and we can quit assuming you know how.
 

#12

Joined Nov 30, 2010
18,210
That's a perfectly good drawing if you got the insides of the relay correct. And there is no need for a diode to protect a push button. It seems you words have confused me, not the drawing.

One thing that is bothering me is that relays are hard to kill. You say you have a bad old relay and a bad new relay.
I think you should whip out your ohm meter and find out which pins are connected to what. You seem to have found the coil. Now you have 6 pins left. Connect to one pin and find out if it connects to anything else. If it doesn't, map it as a normally open contact. If it does, it's either a wiper or a receiving contact. Activate the coil and see if it then connects to anything. Repeat until you know what every pin is connected to and make a drawing. When you get done, that will eliminate some of the guessing as to why you have 2 relays that don't work.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
You were right, I had not correctly metered out the relay.
So I have now metered out the relay and used it to switch the hot wire. The pump works when I press the 12V push button, but I still get the electrical hum and the slow leak from the pump.
So I tried switching both hot and neutral wires, with the same result.

I metered across the pump when I can hear the hum and I get 0VAC.

This doesn't make very much sense...:confused:
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
After chatting to another coffee machine person, I just discovered that it is not the pump that is humming, it is the 3 way valve!
I'm going to try switching the hot wire on the pump and the hot wire on the 3 way valve.
 

Thread Starter

Ross Satchell

Joined Jan 2, 2017
42
Yup, that was it!
I had been wrongly assuming the pump was the culprit.
Instead it was the 3 way valve.

Just goes to show that assumptions are the mother of all screw ups!
 
Top