Valves not actually being stopped by the controller

[Vlad]

I have v2.3AC OpenSprinkler, an extension board and 11 valves from here. All of them are in sequential mode. For some reason, sometimes, when the device stops a valve, it doesn’t actually stop. It shows as stopped in the interface, but, in reality, it’s still opened (measured ~24V on controller valve output). Can be one or more that don’t get stopped. I can only stop them by unplugging the controller and plugging it back in after a few seconds. Happens the same if I plug the valves in the extension board. To be even more confusing, it doesn’t do it all the time. Sometimes it does, sometimes it works just fine. Lately, this happens quite often. It seems to happen to the same 2 groups of valves (most of them are grouped in 2, plugged in the same valve output port in the controller or expansion board).
Can those cheap Chinese selenoids create this type of behaviour and f*ckup the device? I can change the valves, but I’d rather not spend money on other valves (and time to set everything up again) if there’s a possibility that this is not the valves to blame.

Thanks,
Vlad

[Ray]

Are you using a 24VAC transformer? If your transformer outputs DC voltage, that would result in the symptom you observed — basically, OpenSprinkler AC must be powered by a AC transformer (which means the output voltage is AC, not DC). If you power it with a DC transformer, it will turn on the valves but not turn them off. This is just due to the way triacs work.

[Vlad]

No, I’m using an AC transformer, a toroidal transformer plugged directly into the controller.

[Vlad]

Hmm, could the valves needing too much current explain this?

[Ray]

Do you have a picture of your AC transformer label? Could you post it here? Also, what type of valve do you have? It helps to know the brand name / model number of the valve.

[Vlad]

this is the transformer i have: http://www.tme.eu/ro/details/58-0032-024-s/transformatoare-toroidale/talema/
this is the valve: http://www.aliexpress.com/item/1-2-inch-stainless-solenoid-valve-2W160-15-normally-clsoed-G1-2-DC12V-DC24-AC12V-AC24V/2054934823.html

[ShawnHarte]

Those valves can be operated as normally open or normally closed, do you have them set as normally closed? They are also 2way valves, have you tried reversing the valve? Some will operate properly in one direction but not the other, personal experience there, I do not have a good explanation as to why though, maybe Ray would know.

Toroidal transformers are not to fond of being fed DC, make sure you do not have a DC offset in your supply power, and no backfed DC to your secondary winding.

Also what is the full part number for the parts you have in your possession, valves and transformer, the numbers can be ‘decoded’ to see what you actually received. The reason I ask is because both sites use a generalized product listing for several products. For instance you could have a 24V*DC* transformer (on board bridge), or a 110VAC valve, either of which wouldn’t work, or would give weird outcomes, such as the one you are seeing being unable to close a valve.

[Vlad]

Update: I’ve now connected them one at each port, and only one valve seems to be behaving like before. Also, I’ve been searching to find how much current the valve needs and it appears to be 1A. Isn’t that too much for the MAC97? I have another 2 more valves on the way here, I’m going to measure exactly how much current they need to open and how much to keep open.

The valves are normally closed. But even if they are not in the right direction, that’s a mechanical problem. My problem is electric. The controller still outputs 24V on that port, even if it says all valves are closed. Even if a start again and close the valve. And, like I said before, the only way to fix the problem is unplugging the power for the controller for a few good seconds. Turning if off from the button doesn’t seem to fix it. It does close the valve, but as soon as I turn it on again the valve turns on.

The DC offset in the supply power.. you kinda lost me 😀

I don’t have a picture with the transformer at this point (have to take apart the box to get access to it), but it’s definitely 24VAC. I’ve measured it a few times and it provides 24VAC. There’s no bridge to transform AC to DC.

I’ve attached a picture of the valve label. As you can see, it’s 24VAC.

Thanks,
Vlad

Attachments:

1. 

   UT8MzD2XqtaXXagOFbXL.jpg
   

[ShawnHarte]

The reason I asked about the direction of the valve was concerning diaphragm valves. They will use water flow to help hold them open. In your case you are correct this is not the issue you are describing.

With a measured 24VAC output on your transformer you do not have any DC issue. If you did it would be very apparent in the output measurements.

The only thing I can think of at this point is that perhaps you are running into the power output limit of the OS. Though this should keep the valve from opening, not closing…

Next question would be why are you able to measure 24 VAC when the valve is supposed to be closed? This is probably best answered by Ray but I will look into the hardware for 2.3 to see if I can find something. Also check your common wire vs the ground of the power supply. It should measure 0V, if not you may have a grounding issue, but again this is unlikely as the cause.

[Vlad]

I’ve measured the voltage between the power input bottom pin and the RJ45 metal shield: 24V. Isn’t the shield supposed to be connected to GND? If so, it seems I might have that grounding issue you were telling me.
I forgot something that might be relevant: a few weeks ago the fuse F0 burned out and I’ve replaced it with http://www.ebay.co.uk/itm/390813577146.

P.S.
This is starting to be pretty frustrating. I have to unplug and plug back the controller every morning….

Thanks,
Vlad

[Ray]

1A is indeed beyond the limit of each channel of OpenSprinkler (800mA max), although I suspect this is not the source of the issue.

Here is my suggestion: in order to isolate potential problems with the valve vs. controller, just unplug the valve wire (let’s say it’s valve 6), and run a quick program, then after the program (or that valve) stops, measure the voltage between COM (or the power input bottom pin, which is internally wired to the COM pin) and station 6 port pin. Is it still 24V? If so, that’s a problem with the controller. If not, that’s most likely caused by the valve.

Technically, the way triac works is that once it’s triggered, it will continue to conduct until the zero crossing in the current (i.e. the current drops close to 0). When running on AC, there should be a zero crossing once every 1/120 second (assuming 60Hz input power). However, if due to some reason there is no zero crossing in the current, the triac will continue to conduct and will cause the valve to not close. One potential reason is if you feed DC voltage to the controller, resulting in no zero crossing in the current. But there might be other reasons.

Other than these, there are also mechanical reasons for valves to not close : for example, a lot of valves rely on sufficient water pressure to build up internally to properly close. So if the water pressure is low, or there is a mechanical issue with the valve, it won’t close. But mechanical issues are easier to tell apart from electrical issues.

[ShawnHarte]

See Vlad, I knew Ray would be able to explain it a little better. The “DC offset,” can occur if you have 2 ground points at differing potentials or more basically ‘voltages.’ At any ground point in the system you can use a basic multimeter or voltmeter to measure across to another ground point and you should read 0 volts, if not this can ‘latch’ the triac and hold the valve open. Like Ray said, divide and conquer… isolate the issue, find a potential cause, fix it, then reintegrate the portion you isolated and test again.

[Vlad]

Thank you guys. I’ll try that and post here with the results.

[Author]

Posts