larryllix Posted January 16, 2014 Posted January 16, 2014 I was thinking more of the wrapped core kind that are placed inline rather than just the beads where wires run though and confused the two in my head. I probably could harvest a bead that encases the wires by cannibalizing one from a keyboard or probably ten other things I have lying around the house. The first thing I'm going to try tonight is using an external 24VAC power supply. If the PS works then focus on the PS wires for the beads. Putting both wires through one bead will knock out common mode noise but that doesn't sound like your problem with an almost isolated unit. If it had wires into a computer or another system that would be more likely. If you can only muster two cores put them on lead of the 24v supply at first. My second thought on this (editted) is it would be a nasty relay in your furnace on the heat, A/C, or fan control line, whose triac in the thermostat is breaking absorbing the counterEMF spike from the relay coil. Electronics don't like that and it is being done right on the thermostat's board. (thinking of the circuit board pictures nicely pointed out by Brian H) That would be the wires to try the toroid cores or steel nuts, next. I had to put a MOV on a relay that my TM751 (X10 RF receiver) to stop it from zapping a second unit on me. Absorbing the spike right at the source is usually best but we can get to that later. Or do both if the parts are in your hand. You could possibly wrap them a few times if the hole size and wire length permits. IOW pass them through the same hole core repeatedly the same direction (wind a coil). Hopefully you will get somewhere with this experiment. If this works for you I am looking at getting the wireless unit and that would give me some hope for it. If this works partially we are going to get some small despiking capacitors across those terminals too and possibly a low-voltage MOV if you have an electronics supplier near you. Who said you weren't an EE type? They are cheap enough these days but you would want something closer to the 24vac rating. http://www.digikey.com/product-search/e ... -07D820KTR I really want something to get analogue temperatures, as well as other values, into the ISY but I haven't seen any way to do that. It may take some A/D conversion using four or five bits and D/A math in the ISY to get values. The night is still young! We always want more!
shannong Posted January 16, 2014 Posted January 16, 2014 I've been kicking this around in my head. During lunch today I came up with another idea. The board is supplied by 120V AC from the house. Obviously things like the blower and other components use the 120VAC. There is a 120VAC lead that comes off the board to a separate 24VAC transformer that then connects back to the board to supply 24VAC to the thermostat and any other components needing the stepped down power. I was thinking I could supply that 24VAC transformer with a separate external 120VAC rather than from the control board. This might isolate it from the voltage spikes occurring when the electric igniter is arcing.
larryllix Posted January 16, 2014 Posted January 16, 2014 I've been kicking this around in my head. During lunch today I came up with another idea. The board is supplied by 120V AC from the house. Obviously things like the blower and other components use the 120VAC. There is a 120VAC lead that comes off the board to a separate 24VAC transformer that then connects back to the board to supply 24VAC to the thermostat and any other components needing the stepped down power. I was thinking I could supply that 24VAC transformer with a separate external 120VAC rather than from the control board. This might isolate it from the voltage spikes occurring when the electric igniter is arcing. For a trial run I would get something and plug it into a AC wall socket completely separate from the furnace and prove the point first. I don't know how much electronic gadgetry you can borrow to try these things but before you go drilling and mounting things I would make sure your plan works first. If you can get it working reliably then start moving things in permanently. To try a doorbell transformer comes to mind. There are cheap 24v units but a 16v unit may work, also. (Edit: I can't find anything to support anything other than 24vac might work) Don't forget, you are going to have to tie the commons of the two transformers together to make your controls work. Only one 24vac side though, not both (live and common.! After doing some thinking I am becoming convinced the signal lines are the ones bringing the noise into the unit as per my second thoughts above. I am not sure, exactly what he components are on the 2441TH circuit board but they sure look like plain old triac output devices. This probably means no isolation or noise suppression on the lines that run the relays that run the fan motor, heating contactor, or A/C contactor. Coils are real nasty for putting out huge spikes that destroy electronics when interrupted (turned off). If you interrupt them at the peak of the AC waveform you can easily get 10-20 times the supply voltage in a short voltage spike! Interrupting them right on the sensitive circuit board is asking for trouble without special opto-isolating output parts. I doubt they have used that much protection technology there judging by the picture of the board. Another thing I just thought of: Do you have one of those modern DC fan motors with multiple (like 10) speeds? I have a GE one with programmable logic inside and one 24v AC input was phase sensitive. That means I was supplying 24vac from another transformer source and the fan motor wouldn't start off at it's slow speed and ramp up the way it was supposed to. I went through some voltage measurement with the factory guy on the phone and he freaked when I told him one voltage was 24.6v and one was 26.4v. When I retorted "What's the difference?" He admitted they used polarity sensing to reduce the number of logic inputs required. I temporarily changed the supply breaker over to the other 120 volt leg, matching the furnace supply, and the ramped up fan speeds worked again. I couldn't believe it... 24 volt A.C. system and it was polarity sensitive on a furnace fan motor!
larryllix Posted January 17, 2014 Posted January 17, 2014 I don't know what happened to this thread but here was my last response. ----------------------------------------- This occurred to me but how do I identify the common/neutral of the 24VAC side of the transformers to connect them together? I already happen to have a 24VAC transformer on hand. That's what triggered the idea. Connect one wire of your transformer to the wire of the furnace transformer that is connected to the metal frame of the furnace (ground). If you can't locate a jumper to the metal case then it won't matter. It will probably have a black wire going to your stat.(24COM) Now take your voltmeter on AC volts and measure across from the free wire on your transformer to the "other" wire of the furnace transformer (live=24Rh&24RC). This terminal will probably have a red wire in it going to the stat (24Rh and 24RC). Make sure you stay on the secondary sides so you are only dealing with 24 volts, if you're are poking around inside the furnace. Now if you get less than 5 volts the secondary windings are in phase with each other. (small difference) If the reading will be about 50 volts then the secondary windings are out of phase. The voltages add up. Turn off power and change your transformer wire, connected to the furnace common, to the other wire on your transformer.(or reverse your plug) Repeat voltage test to prove them in phase. It may not make any difference but why have the higher voltage around? If you are doing this at the stat you will need to shove both established common wires into one terminal (24COM). Pull the red wire out of the R terminal(s) and replace it with your transformer live wire. Tape up the furnace feed R wire so it doesn't bark to anything. Good luck!
shannong Posted January 17, 2014 Posted January 17, 2014 I don't know what happened to this thread but here was my last response.----------------------------------------- Good luck! Oops. I deleted it because I figured out what I was doing wrong. Sorry you saw it and typed out a response in the interim. Connecting the on board 24VAC transformer of the furnace to an external 120VAC power source didn't resolve the issue. All wiring and connections are being down in the mechanical closet with the furnace. When attempting to connect a separate 24VAC transformer I was putting one of the leads of my multimeter on the neutral on the 120VAC side and looking for a zero reading on one of the 24VAC terminals. However, I didn't have the transformer grounded so I wasn't getting a zero reading through any combination. Once I grounded it to the metal frame of the furnace then... Viola! I now have a separate 24VAC transformer powered external of the furnace that is also happens to be connected to a surge supressor, although same branch circuit. The common/neutral wires of the two power systems are connected together. All is working as expected. So far I've cycled it about 10 times without the damned Energy mode engaging but that isn't enough to verify that the problem is eradicated. I'll have to monitor over the next 24 hours. I've got an ISY program setup to detect set point changes that would be indicative of Energy mode being enabled locally at the thermostat. Thanks for your continued help.
shannong Posted January 17, 2014 Posted January 17, 2014 So far I've cycled it about 10 times without the damned Energy mode engaging but that isn't enough to verify that the problem is eradicated. I'll have to monitor over the next 24 hours. I've got an ISY program setup to detect set point changes that would be indicative of Energy mode being enabled locally at the thermostat. Thanks for your continued help. Arrrgh! Energy mode just turned itself on. Looks like the only answer is a inline solution of some fashion to neutralized the spikes/noise.
larryllix Posted January 17, 2014 Posted January 17, 2014 Arrrgh! Energy mode just turned itself on. Looks like the only answer is a inline solution of some fashion to neutralized the spikes/noise. Can you identify the timing of the mode change with a cycle turning off? This would zoom in on the relays dropping out causing spikes doing the damage.
shannong Posted January 17, 2014 Posted January 17, 2014 Can you identify the timing of the mode change with a cycle turning off? This would zoom in on the relays dropping out causing spikes doing the damage. The timing is easily identified after watching it about a hundred times. It occurs when the electric igniters begin sparking to ignite the gas. So I assume there are voltage spikes on the lines 3-4/sec. Oddly the two igniters are fed from two different spots on the control board. One is a simple one-wire bladed connection and the other also a one-wire entering into a black cylinder on the board.
larryllix Posted January 17, 2014 Posted January 17, 2014 Can you identify the timing of the mode change with a cycle turning off? This would zoom in on the relays dropping out causing spikes doing the damage. The timing is easily identified after watching it about a hundred times. It occurs when the electric igniters begin sparking to ignite the gas. So I assume there are voltage spikes on the lines 3-4/sec. Oddly the two igniters are fed from two different spots on the control board. One is a simple one-wire bladed connection and the other also a one-wire entering into a black cylinder on the board. Can you shield the thermostat wiring from the high-voltage ignition wiring? Wrap them in aluminum foil and ground it with a clip lead? Careful with the contact spots! Try to keep loops and antenna formations away from the ignition wires? Bundled together. This sounds like RF through the air pickup. Is the furnace transformer (24v) secondary grounded to the furnace case? If it not visible a simple voltage check from both 24vac wires to ground should tell.
shannong Posted January 17, 2014 Posted January 17, 2014 Can you shield the thermostat wiring from the high-voltage ignition wiring? Wrap them in aluminum foil and ground it with a clip lead? Careful with the contact spots! Try to keep loops and antenna formations away from the ignition wires? Bundled together. Maybe. The difficult part will be where the wires fan out to each of the contact screws. Is the furnace transformer (24v) secondary grounded to the furnace case? If it not visible a simple voltage check from both 24vac wires to ground should tell. It is. This sounds like RF through the air pickup. Why do you say that? I'm not disagreeing just looking for details.
larryllix Posted January 17, 2014 Posted January 17, 2014 Can you shield the thermostat wiring from the high-voltage ignition wiring? Wrap them in aluminum foil and ground it with a clip lead? Careful with the contact spots! Try to keep loops and antenna formations away from the ignition wires? Bundled together. Maybe. The difficult part will be where the wires fan out to each of the contact screws. Yeah, do the best you can even if the foil splits and you get some individual wrapping of wires. Make sure of no contact or proximity to the bare wires and terminals. This sounds like RF through the air pickup. Why do you say that? I'm not disagreeing just looking for details. I am guessing but... the ignition will be a high-voltage discharge like a spark plug on a car, right? This creates all kinds of radio waves from the little antenna going from the transformer to the spark plug discharges the high voltage as a spark. Sparks and lightning over-discharge and then some ricochets back to the source in an attempt to equalise charges. This creates a high frequency ringing back and forth trying to find equilibrium. (think electron inertia). High frequency doesn't like to follow insulation properties and it bleeds to everything. Some wires going to your thermostat get this RF induced into them and attempt to find their way to ground, getting back to their source, through your thermostat. It's all theory, at this point, until it works, then it was the facts! Don't give up! It sounds like you are a stubborn SOB and that is good here.
shannong Posted January 18, 2014 Posted January 18, 2014 I used aluminum duct tape. It is has better workability than aluminum foil. I left the paper on and put black tape on the edge facing the board. That allowed me to create a collar that went completely over the screw terminals and touching the board. I then covered the entire length of wires to the entrance hole of the housing. Since the aluminum tape was actually in this port it was touching the metal frame thus grounding it. Sadly, it made no difference.
shannong Posted January 18, 2014 Posted January 18, 2014 I spent an hour in chat with Smarthome yesterday. They sent a 24VAC power supply to provide external power. Their first wiring instructions were "just connect it to the C & R". I made them be specific about how they wanted it connected so that when the issue wasn't resolved it would show their work around does not work. They wanted me to connect it such that the C and R terminals of the tstat were connected to the power supply only so the noise would be separated. I argued with him for about 15 min explaining that wouldn't work because there would be no return path for the W,G, and Y signals back to the furnace. He specifically said the C could not be connected to the furnace because that would allow the noise back in. That's the only thing we agreed upon. After he sent me a few drawings that weren't legible I finally relented and tried it his way to prove him wrong so we could move on. As you already know it didn't work for the stated reason. I explained to him there needed to be a common/neutral wire to provide a return path. So he now relented and said connect the C from the furnace and the new PS. I asked him which terminal on the PS I should use and he said the "black one". There isn't one and their both just labeled "AC". He didn't initially get that it was important to identify which was the common/neutral vs hot/line. Luckily, I had a multimeter handy to identify it. So I connected it and it of course now worked but again the noise/spikes were back in the system thus no resolution to the Energy mode issue. He then really pissed me off and said I should get an HVAC technician to look at the furnace to see what's wrong with it. I quickly replied it's the tstat allowing noise in and nothing is wrong with the furnace. He also suggested replacing the 2441THs with the Venstar+2441V. I asked him how that would help if the problem is my furnace and not the tstats as he stated. He replied it was just something to try. I suggested the engineering team instead should identify an inline filter that would work. He then suggested then ACT AF300 20amp inline filter. I asked him how he would suggest I wire that in. I didn't get a response. I asked for an RMA on the damn things and he readily agreed. On the forms he put the reason as "Not working as expected". I forced him to change it and he then put "defective firmware or software". I pointed out that also isn't true since they could then fix it with a software update, which of course they can't but I was done with him at that point. I closed with reminding him it's a known issue and complained about by many people and they were unethically making more and selling them with the known issue. I suggested they at least should list compatibility issues if they weren't going to fix it. No reply to that. I'm going to file a complaint with the BBB they're a member of.
larryllix Posted January 18, 2014 Posted January 18, 2014 I spent an hour in chat with Smarthome yesterday. They sent a 24VAC power supply to provide external power. Their first wiring instructions were "just connect it to the C & R". I made them be specific about how they wanted it connected so that when the issue wasn't resolved it would show their work around does not work. He also suggested replacing the 2441THs with the Venstar+2441V. I asked him how that would help if the problem is my furnace and not the tstats as he stated. He replied it was just something to try. I suggested the engineering team instead should identify an inline filter that would work. He then suggested then ACT AF300 20amp inline filter. I asked him how he would suggest I wire that in. I didn't get a response. I asked for an RMA on the damn things and he readily agreed. I'm going to file a complaint with the BBB they're a member of. Too bad on all counts there. My next and last try was the ferrite beads on all the lines but easy access to them would be a necessity and lowest risk. Sounds like the support tech was another "know a bit" type. At least they didn't get you to unplug your fridge and spin around three times like some read off their scripted "fix" software. The Venstar units are a big jump in price from what I see. I just want something as a temperature sensor, so maybe you can tell me this since I have searched for several days , without success, repeatedly on this point: (I am going to spell funny ) Can you read the analogue value of the temperature inside the ISY? Not just the binary status points but the real actual analogue value in some form of degrees.
shannong Posted January 18, 2014 Posted January 18, 2014 My next and last try was the ferrite beads on all the lines but easy access to them would be a necessity and lowest risk. Access is easy on both ends. I just made the whole in the wall a little bigger to able to slip a bead inside the wall. I added a single ferrite bead on the tstat side but could only loop the wires through once (thus two passes of the wires) and the problem persisted. I didn't know the ISY had a temp sensor on board.
shannong Posted January 18, 2014 Posted January 18, 2014 !!! One more update.... And finally some good news! Earlier I had posted about using a surge suppressor with protected "network" ports to filter out the noise/voltage spikes of the electric igniters on my gas furnace that was causing the Energy mode to turn itself on. It did prevent the Energy mode from coming on but only worked for Heat. For some reason I was unable to turn on Fan or Cool. I tried removing just Cool (Y) and Fan(G) from the suppressor but then the noise was back and tripping the Energy mode again. Considering it a failure, I deleted the post. I was thinking about this method again and thought it strange that Heat would work and not Cool or Fan. They all use the same voltage. I decided to try this workaround with another surge suppressor I have on hand that also has network ports. I connected it the same as before and the Energy mode has not yet turned on after 20 cycles of the Heat. (the most I've gone without Energy mode being enabled is 10 times. Usually it's about one out of 5) And this time I CAN turn on Cool and Fan!!! It seems there was something funky about the way the first suppressor was handling the filtering. Ethernet calls for a 4-pair RJ-45 (8 wires) but in reality only uses 4 of the wires. 1,2,3 and 6. Perhaps it was handling the others differently. Not sure..... The surge suppressor is simply connected inline between the tstat and the furnace using the network ports. I created two pigtails so that one is used to connect the wire in from the tstat to the suppressor and another to go from the suppressor to my furnace control board. I have my ISY monitoring it looking to see if Energy mode is turned on. However, I'm in TX and it's 70F today so there won't be any heat cycles until sometime later tonight when it gets "cold". I'll update if anything changes.
larryllix Posted January 18, 2014 Posted January 18, 2014 I didn't know the ISY had a temp sensor on board. I didn't state my question clearly enough and subject to misinterpretation or maybe it was my Canuck accent, eh? "Can you read the analogue value of the temperature inside the ISY? Not just the binary status points but the real actual analogue value in some form of degrees." To clarify my Q: Can you read the actual temperature reading from inside the ISY that was transmitted from the 2441TH? Is it only whole number degrees or are there decimal places implied?
larryllix Posted January 18, 2014 Posted January 18, 2014 !!! One more update.... And finally some good news! Earlier I had posted about using a surge suppressor with protected "network" ports to filter out the noise/voltage spikes of the electric igniters on my gas furnace that was causing the Energy mode to turn itself on. It did prevent the Energy mode from coming on but only worked for Heat. For some reason I was unable to turn on Fan or Cool. I tried removing just Cool (Y) and Fan(G) from the suppressor but then the noise was back and tripping the Energy mode again. Considering it a failure, I deleted the post. I was thinking about this method again and thought it strange that Heat would work and not Cool or Fan. They all use the same voltage. I decided to try this workaround with another surge suppressor I have on hand that also has network ports. I connected it the same as before and the Energy mode has not yet turned on after 20 cycles of the Heat. (the most I've gone without Energy mode being enabled is 10 times. Usually it's about one out of 5) And this time I CAN turn on Cool and Fan!!! It seems there was something funky about the way the first suppressor was handling the filtering. Ethernet calls for a 4-pair RJ-45 (8 wires) but in reality only uses 4 of the wires. 1,2,3 and 6. Perhaps it was handling the others differently. Not sure..... The surge suppressor is simply connected inline between the tstat and the furnace using the network ports. I created two pigtails so that one is used to connect the wire in from the tstat to the suppressor and another to go from the suppressor to my furnace control board. I have my ISY monitoring it looking to see if Energy mode is turned on. However, I'm in TX and it's 70F today so there won't be any heat cycles until sometime later tonight when it gets "cold". I'll update if anything changes. I would be paralleling every line signal line and 24vac power (R,G,W1-2,O etc..) to ground at the furnace terminals where the ©ommon is tied to the case of the furnace with a 24vac MOV. The DC rating would be about a 48 vdc (28vac x 1.414 x 120%) to absorb any spikes on the waveforms. Do not extend the lead lengths! May have to put threaded grounding screws/terminals all along for connection. A/C hasn't been involved yet so it could be done later, if it works. OK, with them and a ferrite bead on each wire and it still does it, where is the trash can?
LeeG Posted January 19, 2014 Posted January 19, 2014 "Q: Can you read the actual temperature reading from inside the ISY that was transmitted from the 2441TH? Is it only whole number degrees or are there decimal places implied?" Yes. 1D 6C D6 1 2441TH - Main 5.11.10.0 true 0 0 0 1D 6C D6 1 D06
larryllix Posted January 19, 2014 Posted January 19, 2014 "Q: Can you read the actual temperature reading from inside the ISY that was transmitted from the 2441TH? Is it only whole number degrees or are there decimal places implied?" Yes. Thank you LeeG !! Qs. - Does the 241ZTH (I see you have one) report the same (used just as a temp and humidity sensor)? - Can programmes do conditionals based on these analogue values similar to a variable? - How are fractional degree values handled? - Where is the setup parameters code being found?
LeeG Posted January 19, 2014 Posted January 19, 2014 "Does the 241ZTH (I see you have one) report the same (used just as a temp and humidity sensor)?" Yes. 1F F B4 1 2441ZTH - Main 5.10.11.0 true 0 0 0 1F F B4 1 "Can programmes do conditionals based on these analogue values similar to a variable?" "How are fractional degree values handled?" Yes. In full degree only, no xx.5 "Where is the setup parameters code being found?" Sorry, don't understand question.
larryllix Posted January 19, 2014 Posted January 19, 2014 " 1F F B4 1 2441ZTH - Main 5.10.11.0 true 0 0 0 1F F B4 1 Thanks again! Where is the list (above) of setup parameters being found? In short: is some other hardware or software needed to read temperature out of the 2441(Z)TH units or is this straight Insteon to the ISY?
LeeG Posted January 19, 2014 Posted January 19, 2014 The information is sent to the ISY using normal Insteon messages. I used a REST call to retrieve that information. http://192.168.2.3/REST/nodes/xx yy zz 1/ - xx yy zz is Insteon address If looking at how to access the information with a client look at the WSDK forum category.
larryllix Posted January 19, 2014 Posted January 19, 2014 The information is sent to the ISY using normal Insteon messages. I used a REST call to retrieve that information. http://192.168.2.3/REST/nodes/xx yy zz 1/ - xx yy zz is Insteon address If looking at how to access the information with a client look at the WSDK forum category. Thanks! and Thanks!
shannong Posted January 19, 2014 Posted January 19, 2014 I would be paralleling every line signal line and 24vac power (R,G,W1-2,O etc..) to ground at the furnace terminals where the ©ommon is tied to the case of the furnace with a 24vac MOV. The DC rating would be about a 48 vdc (28vac x 1.414 x 120%) to absorb any spikes on the waveforms. Do not extend the lead lengths! May have to put threaded grounding screws/terminals all along for connection. A/C hasn't been involved yet so it could be done later, if it works. OK, with them and a ferrite bead on each wire and it still does it, where is the trash can? All connected and working. Heat, AC, and Fan control. The Energy mode hasn't magically turned itself on, yet. It's probably cycled on about 40 times. The surge suppressor is grounded on the same circuit as the furnace and I verified the frame of the furnace is grounded. I'm not sure why I'd add the MOV considering all leads are connected to the surge suppressor which I would assume has an MOV of some flavor.
Recommended Posts