If your looking at a Trane controller they also do low side switching. In other words they switch the grounded side of the 24v power supply. Many including myself often read voltage to the unit casing. In this application you can't do that because there is always 24v at the actuator so you will read 24v even if the controller isn't calling when you read to the cabinet case. You need to actually measure voltage accross the wires with the actuator conected. You may still read the bleed voltage same way you do on a SCR controled heater.
Originally Posted by Wj Stevens
On some cotrollers it is also quite possible that the code is bad or written incorrectly. Bad code could energize both outputs at once, hence 24 v to both open and close at the same time.
Edited by powerhead on more than one occasion
The reason your seeing the "bleed through" is most likely because the controller has Triac outputs, totally different than a set of contacts. A cheap and quick way to test an output of a Triac is to simply use, in this case, a 24vAC relay as a meter, if the relay pulls in, the Triac is "closed". If the relay does not pull in, the Triac is "open".
Here is a decent place to start if you want more understanding:
Or do a google search for more.
It is impossible for one to learn what one thinks they already know"
codewriter...... yes it does have triac outputs....... I like the 24v relay idea.
Its pretty hokie I guess... but most people do have an extra 24vAC relay banging around somewhere, so its a quick and easy tester.
I can hear people laughing at me now... O'well
It is impossible for one to learn what one thinks they already know"
That's not hokie, it's exactly why I have a 24vac test light in my bag.
Originally Posted by codewriter
I've used a relay for testing these outputs as well. Hokey or not it works better than a meter in this case and is reliable. Most small unitary controllers I've come accross use triacs regaurdless of brand. Cheaper/smaller would be my assumtion but you know what they say about that. If you have enough of a coil/actuator motor hooked to the output a meter can be used but you will still see a couple volts of bleed through just not the full 24v.
Last edited by willf650; 09-07-2007 at 05:45 PM.
You guys are awsome..... I am nominating you all for control guy of the month..... lol
These kind of answers are sometimes a challenge to get in the field.
I wound up using a 24v relay with an indicator light as a test light......... it will live in the toolbag now .....my permanent 24v tester, until I need it to make a boiler run.
The timing from the controller looks to be out of wack now that I can see it, the stroke.. time on both of those actuators is 80..110............ on recycling the controller, the hw actuator runs one direction only for 130 sec. then deenergizes. The CW runs in one direction only for only 30 seconds and deenergizes.
Then with the stat all the way down I didn't get a signal to open CW......
I think I have exhausted all the simple stuff, My next step is to interface with it with the software and see whats going on.... I haven't done that before but my boss should be able to help me out on that.
I have similair type actuators I work with to control evaporator pressures for temperatures. In supermarket work, I like to see everything proofed. Current sensing relays for comp run. Fan proofing with duct static sensors. Ect Ect.
I can not understand why the valves can not be built with a simple end switch of sorts proving either closed 100 percent or open. So then during a daily synch it can position exactly.
I agree......... Then again If it could sence the resistance when it reaches a stop, the resistance would give the needed feedback.
Originally Posted by Dowadudda
Most actuators I work with either have end switches or feedback wires....these dont.
I see that others have explained to you how the floating point motors work.
They're pretty simple devices to understand and work with.
When troubleshooting, I generally try to be systematic about it, and try to test the easiest and most obvious thing first. That is ... does the floating point motor work at all, and will it do a full stroke?
There are two schemes commonly used to stroke them electrically. One is to connect the actuators black wire to common, and to alternately apply 24 VAC to either the red or the white wires to stroke the actuator this way or that, using relays or triacs. The other scheme is to connect 24 VAC to the black wire, and to alternately switch the red or white wire to common, to stroke the actuator.
I'll usually simply determine which scheme is being used, then disconnect the red and white wires from the controller. Then, one at a time, tap the red or white wire, by hand, to a common or to a 24 VAC source (depending on which scheme was used) and watch to see if the actuator moves thru full stroke. Timing it while I do so. Then I'll manually stroke it the other way.
I'm, at this point, simply trying to see if the actuator is functional and whether or not the damper or valve being moved (plus any mechanical linkages involved) is capable of full movement.
If I get full movement both ways, problem is NOT with the actuator or the mechanics of the device it is supposed to move.
If I don't get full movement both ways, I disconnect the actuator mechanically from the driven device, and try again. To determine if the actuator is at fault ... or that the valve or damper is binding, running into mechanical stops, OR perhaps the actuator is undersized/underpowered for the task at hand (which I've occasionally found to be the case).
But if everything works via by method of hand switching to stroke the mechanism, then the problem is likely to be with the controller.
The other guys' tip about cycling power on the controller is a good one. It -SHOULD- be programmed to do a recalibrate/repositioning upon a power fail and restoration. This usually involves some control scheme that causes the controller to try to stroke the actuator 100% closed for some period of time, ranging from 100% of that full stroke time period the actuator requires, to 200%. After which it should start to reposition the valve or damper to the required position.
Be generous, give it lots of time if you don't know the settings and program scheme. Reconnect the red and white wires, cycle power to the controller then wait and watch for several minutes. What you're looking for is that the controller successfully cycles the actuator first one way, then the other. Any amount. If it does that, then ALMOST certainly the triacs or relays of the controller are working fine.
If they are, then the problem is likely elsewhere. It's time to start looking at inside the controller at it's setup and programming. Or, maybe some sensor used as a MV in the actuator control loop is at fault. Etc.
Commonly, as a programming error, I have found that programmers not infrequently forget to include the code or setting for some built in firmware, that will instruct the controller to do a periodic "calibration check". (Where controller cycles actuator all the way closed for some time period equal to or greater than the cycle time of the installed actuator, then repositions it to the needed position.) Or, the programmer set up the control loop with an assumption that cycle time for installed actuator was, for instance, 90 seconds ... where what was actually needed was 60, or 95, or 180 seconds.
Just some thoughts, good luck.
Very good post osiyo......... I appreciate it a great deal........ You guys have really simplified this for me. I did not understand the idea of the calibration sequence or for that matter how the thing knew where it was, and it really confused me.
When it calibrates ... does it indeed sense the resistance when it runs into a mechanical stop, or does it simply run it in one direction long enough that it knows it should be at full stroke? This part doesn't matter that much to me at this point just curious.
I guess I also have to assume if someone walks by and moves the operator position manually its gonna be screwed until the next calibration.