Taco Switching Relay Fried Twice--Tstats at fault??

HomerJSmith

jac34, You can read that on the label. On the metal covered SRs they are 15 VA . You can call Taco tech support for a definitive answer.

jac34

HomerJSmith said:

I did ask you what that black (blue actually) wire was connected to the Rc connection on the base and I didn't get an answer. If the latest group of pics is correct and the blue wire is connected to the C connection on the Taco, there's your problem. The Rc is like the Rh connections, they are both the 24 Vac power connections not the C common connection. The C (the green wire) connection on the Ecobee should be connected to the common where the blue wire is connected. Disconnect the blue wire on the Ecobee and from the Taco common as it isn't needed and connect the green wire to the common. Read my third post. You have a direct short on the secondaries of the transformer. That's all the news fit to print! Of course, "I'm often wrong, but never in doubt."

Heck, what's the green wire to the 24 Vac connection on that Taco terminal? You could have drilled out the pop rivets on the transformer and used a nut and bolt to put in another transformer, easy peezy.

That blue wire was going to the com. I believe you are right and this is prob what blew the 2nd transformer. I know for sure the green was going to COM before they came out to check on the failure and the blue was in the 24VAC terminal all along. The tech put the fuse on but had the blue (Rc) going into the COM rather than the green. He must of switched the wires by mistake when he installed the fuse. After he left and determined the Tstat was the problem, I then put it back on the base and it popped, making the fuse useless since the blue was going to com rather than green. Would that do it?

wmgeorge

Yes and remember my advice above? Mark the wires, the next guy will assume Green goes to fan and so on. Install the fuse, 2 amp on 40 VA.

jac34

Thanks, I went ahead and took the blue wire out of the equation all together, I understand what you mean. I don't know why it was ever included unless they were just tryin to isolate it? I know this is prob a stupid question, but is the COM supplying the power to the Tstat or is the red wire supplying the power? And what purpose does the terminal "24VAC" where the blue wire was connected serve?

Here are the updated pics on the wiring for zone 1

Jamie Hall

Com is returning power from the thermostat at all times -- if it's hooked up properly. R can be viewed as the power supply. In a two wire arrangement, the idea is that the return from the thermostat is through W, but at too low a current to switch the controlled device on but enough (across the open switch between R and W in the thermostat) to charge the battery. This works, provided that there are long enough off intervals, close enough together, to keep the battery charged. This is true on most forced air systems -- which is what they are designed for. It also tends to be true in the shoulder seasons for most other systems -- but not in heavy demand times, which is why a Nest or what have you will work fine most of the time, but not when it's cold out.

tim smith

Have you got it fixed? I would check all stat wires between C and R then W for resistance. I just had one where client used a 4 connection push in connector to extend a wire and found r, w, g all shouted to C internally in that push in connector. All made up connections looked good too. Also if you have 4 eco bees drawing on the one Taco transformer, that’s too much in my book.

jac34

Well, it’s working with the temp 40va transformer. I now have to figure out the long term fix. I could either get a new board that has a better transformer that can handle everything. 

Where im still confused is understanding how much power is being pulled from this transformer and then deciding on what size transformer is needed. The board, Taco SR 503 comes with a 15va 24v transformer that failed after 3.5 yrs. The other 2 ecobees come to this board for a total of 3, but just for heat on the red/white zone connections going to the boiler but are not powered from this board. When the transformer failed only the zone 1 heat only ecobee lost its power. The other 2 still had their screen displays. I’m assuming b/c they are getting their power from 2 different units, an air handler and a high velocity a/c unit. 

So I’m assuming the original transformer failed b/c it was overloaded and the second one failed immediately b/c the tech wired the common incorrectly causing it to short?? 

Is there a way to estimate the power draw on the transformer so you can fit it with the right specs? 

I believe this is why ecobee recommends an external plug in transformer if the internal transformer is running other devices like my 3 circulators?

Zman

You could put an amp meter on it and see what it is drawing.

MaxMercy

Quick general comment on transformers. Transformers that are fused *internally* are fused only for heat, not current (although a severe short could blow the internal fuse I suppose). These thermal fuses are buried deep in the windings are not replaceable without unwinding the transformer. The idea is that the primary will fail open long before the transformer reaches an exterior temperature high enough to cause it to catch fire or damage anything close to it, particularly if an *exterior* current fuse is bypassed or raised in rating.

It would be interesting to try a 15VA transformer back in and monitor the temperature with all devices calling. If that transformer runs, but runs hot, it might have just been a cumulative heat failure over time - the first failure anyway. The second could be a wiring issue as others have supposed.

John

Daveinscranton

Perhaps you can use one transformer, appropriately sized, to power everything.  You could clip the leads from any internal board transformers and wire your 24 volts to the boards.  One transformer.  One connection to 120 volts.  Less chance of getting something out of phase.  Transformers don’t like being paralleled unless they are exactly the same make and model.  Not even a great  idea to parallel them even with that caveat.  Maybe fuse each board with 1 amp fuses.  You put the ammeter in series with the load to determine the load.   Not parallel as you would a volt meter.  Give yourself some head room on transformer sizing.  A bit bigger is ok.  Doesn’t have to be grossly oversized.  In the end, you want it to all be trouble free.  For decades.  And keep good notes.  Labels are your friend.  Being consistent with wire colors is your friend.  Meticulous wire termination is your friend.

best wishes 

wmgeorge

Com or C wire normally Black in standard color code for Tstats is the "other side" of the 24 vac power source Only needed to power the digital Tstat, Red in this case is suppling the power to Rh. Yes it does draw power from the transformer I do not know how much unless you check the amp draw. You can mount the new 40 va transformer in a 1/2 inch KO and leave it connected, especially if your going to run 4 digital Tstats. Please, please mark the wires and fuse the secondary, I am fussy because I was the one going out on someone else's callbacks and have seen Stupid in Action! Yes the way you have shown above is correct.

Jamie Hall

Your best bet is, as has been noted, an external transformer. You might find a board with greater power capacity -- but keep in mind that the transformer on the board is intended to power the board only. The thermostat connections on a board are not intended to power anything, and certainly not a power gobbling wi-fi thermostat.

jac34

I just find this interesting that none of the Taco boards have larger than a 15va internal transformer on their boards; I reached out to their tech support today to see if I could get a different board that's beefier. Yet, ecobee and other smart Tstats call for a 40va transformer? I guess that's why they promote the external plug in transformer separate from the relay board, but it doesn't seem like the HVAC contractors understand this as they just connect them directly to the board? Seems the two aren't aligning with the changing marketplace as the smart Tstats are becoming more commonplace.

My air handler running my other zone for heat/ac has a 40va inside of it that is powering my other ecobee so why do the Taco boards use a 15va? I'm assuming a boiler with circulators/zone valves uses more power than a traditional furnace?

Daveinscranton

I am sure that it has been mentioned but it is worth bearing in mind that the length of the wire run and the wire gauge (size, as in how much resistance per foot) are important.  Folks get used to thin wire for millivolts and micro amps going 12”.  String a long enough run of undersized wire and you have a significant resistor.  You can get away with undersized wire.  Until you can’t.  Ohm’s Law is a physical reality.

Jamie Hall

I think there may be many factors, @jac34 . One of the most prominent is that there is no standardization at all in the "smart" thermostat market. Even various models of single brands -- Nest or Ecobee, for instance -- have different power demands and, perhaps more important different wiring.

Another is that the various control boards -- Taco and other manufacturers -- are designed, and designed very well indeed -- to accept digital inputs (that is, basically closed or open switched inputs) from various other devices. These may be powered thermostats. They may be unpowered or battery powered digital thermostats. They may be on/off aquastats. They may be on/off analogue thermostats. The variety is huge -- but they all have one characteristic: they present a single pole, single throw input. Further, the boards are designed to translate that input into a similar single pole, single throw output to control other devices, which may be pumps, valves, burner ignition controls, etc. which are powered (with a wide range of power demands) externally.

The various boards do this very well indeed, but to do it they don't require much power.

To put enough power on a control board to power all the possible devices which might be connected to it would be silly.

A much more intelligent solution would be for the power hungry thermostats or other devices of that sort to have their own internal power supplies, independent of the switched output. That, however, would severely limit the sales potential to Joe Homeowner who might not be able to figure out how to connect the power supply. There is also an internal problem: most of the newer "smart" devices do not have the switched output isolated from the power supply and computing logic. Usually the switched output (or outputs) will be Triacs or similar devices. From the electrical engineering standpoint, while this is elegant and efficient and cheap in terms of devices, it is actually exceedingly poor practice, as one -- or often both -- of the switched output terminals (think R and W, for instance) are actually internally connected to portions of the computing and power supply circuitry. There is an astonishingly large number of ways in which this type of interconnection between what should be independent circuits can cause malfunctions at best, and fatal damage at worst. Unfortunately, triacs and the like are cheap, and relays, which is what should be used, are not.

Fortunately, the majority of heating systems in the markets at which these devices are aimed have a rather small set of characteristics: that is, both the on and off desired cycles are short, in the range of minutes, and the controlled device has available a power supply and an internal relay, with one pole of the power supply available on one of the two wires (the return being on the other side of the relay coil from the other control wire). So -- when the widget is not calling for heat, it is possible to extract power from the two control wires and use it to charge an internal battery (often a super capacitor, not a real battery). When the device is calling for heat, this source is shorted out, but the internal battery can supply power long enough for the device to continue functioning.

For this type of controlled system these devices work very well indeed.

In my own view, a much better approach would be for these widget manufacturers to offer a simple intermediate board. This would have its own 24 VAC transformer -- perhaps 40 or 80 VA capacity -- and a number of inputs for the widgets, each with three connections. There would be, the two control connections and the required third wire power return. There would be a relay for each input which would be a single pole, double throw relay, thus offering a choice of straight through or inverse logic on the output.

jac34

Jamie Hall said:

I think there may be many factors, @jac34 . One of the most prominent is that there is no standardization at all in the "smart" thermostat market. Even various models of single brands -- Nest or Ecobee, for instance -- have different power demands and, perhaps more important different wiring.

Another is that the various control boards -- Taco and other manufacturers -- are designed, and designed very well indeed -- to accept digital inputs (that is, basically closed or open switched inputs) from various other devices. These may be powered thermostats. They may be unpowered or battery powered digital thermostats. They may be on/off aquastats. They may be on/off analogue thermostats. The variety is huge -- but they all have one characteristic: they present a single pole, single throw input. Further, the boards are designed to translate that input into a similar single pole, single throw output to control other devices, which may be pumps, valves, burner ignition controls, etc. which are powered (with a wide range of power demands) externally.

The various boards do this very well indeed, but to do it they don't require much power.

To put enough power on a control board to power all the possible devices which might be connected to it would be silly.

A much more intelligent solution would be for the power hungry thermostats or other devices of that sort to have their own internal power supplies, independent of the switched output. That, however, would severely limit the sales potential to Joe Homeowner who might not be able to figure out how to connect the power supply. There is also an internal problem: most of the newer "smart" devices do not have the switched output isolated from the power supply and computing logic. Usually the switched output (or outputs) will be Triacs or similar devices. From the electrical engineering standpoint, while this is elegant and efficient and cheap in terms of devices, it is actually exceedingly poor practice, as one -- or often both -- of the switched output terminals (think R and W, for instance) are actually internally connected to portions of the computing and power supply circuitry. There is an astonishingly large number of ways in which this type of interconnection between what should be independent circuits can cause malfunctions at best, and fatal damage at worst. Unfortunately, triacs and the like are cheap, and relays, which is what should be used, are not.

Fortunately, the majority of heating systems in the markets at which these devices are aimed have a rather small set of characteristics: that is, both the on and off desired cycles are short, in the range of minutes, and the controlled device has available a power supply and an internal relay, with one pole of the power supply available on one of the two wires (the return being on the other side of the relay coil from the other control wire). So -- when the widget is not calling for heat, it is possible to extract power from the two control wires and use it to charge an internal battery (often a super capacitor, not a real battery). When the device is calling for heat, this source is shorted out, but the internal battery can supply power long enough for the device to continue functioning.

For this type of controlled system these devices work very well indeed.

In my own view, a much better approach would be for these widget manufacturers to offer a simple intermediate board. This would have its own 24 VAC transformer -- perhaps 40 or 80 VA capacity -- and a number of inputs for the widgets, each with three connections. There would be, the two control connections and the required third wire power return. There would be a relay for each input which would be a single pole, double throw relay, thus offering a choice of straight through or inverse logic on the output.

Some of that was over my head, but I hear what you are saying about the boards. It would probably be overkill in most cases and would hurt their bottom line to make them all this way.

I am going to go ahead and connect the Tstat directly to a plug in 24v 20va transformer that I have from a ring device. I'm assuming 20va would be enough if it's dedicated to just the Tstat? it has to open wires on the end that I can just wire directly to the blue/green wires coming from the Tstat base.

Jamie Hall

Sounds good.

jac34

Zman said:

You could put an amp meter on it and see what it is drawing.

I have a clamp multimeter, what would be the best way to test the draw?

Daveinscranton

Depends on the meter.  Not a fan of clamp on meters for this sort of thing.  That said, I have a relay currently energized.  The draw as per the manufacturer is 200 milliamperes or 0.2 amps.  The panel meter hardwired into the circuit reads 200 milliamperes or 0.2 amps.  I put my Fluke meter 90 degrees to the wire.  If I positioned it just right, it read 0.2 amps.  So, you may well be able to get a ballpark figure from your meter.  

HomerJSmith

MaxMercy, cool name, I have unwound more than one Class 2 transformers to see why they failed. They all had a fuse wire in the center of the winding on the primary side. I always thought that the failure was on the secondary side. A simple ohm meter test would have confirmed which it was. Of course I didn't do that.

HomerJSmith

jac34, you could just clip the primary and secondary wires at the transformer that has failed, drill out the pop rivets holding the transformer to the back plate, mount a 40 VA 24 V transformer with small bolts and nuts to the back plate.

Strip about 3/8" of insulation off the original wires and wire nut them to the new transformer wires. Primary is the black and white 120 Vac and the other two are the 24 Vac secondaries.

Some transformers have multiple wires because they are rated in various voltages in one transformer.

That's what I would do.

https://www.supplyhouse.com/White-Rodgers-90-T40F1-Transformer-40VA-60-Hz-120V-Primary-24V-Secondary-Foot-Mount

Joe Mattiello

Hi
sounds like it’s been working for a while and then crapper out.
What’s the amp load on T stat? The SR504 has a 15VA transformer which is powering the power state control and thermostats? Calculate the load. Transformers are pretty cool, where they will continue to accommodate your needs until they die from over dedicated service.
the good need is the ST control will work fine with external transformer so for all practical purposes your service technician already repaired it.
if you need more help, contact Taco Tech support at 401-942-8000. Ask for tech support.
let know how you made out. You can always reach out to me too. Hopefully this was helpful. 

tim smith

The Ecobee 3 draws less than 3:5 va, so 3 of them + taco Sr relays shows why it was right on the cusp of blow rating of 15va. Would take take all on plus a longer run cycle I think.

MaxMercy

HomerJSmith said:

MaxMercy, cool name,

Thanks, from the movie The Natural - a sportswriter character played by Robert Duval. Great film.

HomerJSmith said:

I have unwound more than one Class 2 transformers to see why they failed. They all had a fuse wire in the center of the winding on the primary side.

Here's a little secret (shhhh). A lot of the primaries of these transformers have three legs, not two. The third is actually a junction for the internal fuse. A little trick is to jump the middle pin to one of the hot legs to restore the transformer. Of course, there will be no more thermal protection, but some installations don't need them.

jac34

Ok, so I decided to just get a larger, 40va transformer, remove the old one from the taco board and install this larger one. Not too bad except getting the darn rivets out to remove the old one!

All seems working fine. It was a little larger than the factory transformer so I had to drill a new hole for the foot bracket to work and it's a little tighter against the cover, but I was able to close it firmly.

Out of curiosity, I ran each zone separately then added one at at time while measure amps with my clamp mm. Interesting as it added about .1 amps per zone (see pics and note the red lights on right indicate the zone is calling for heat). So it reached a total of .32A when all 3 zones were running. If I'm calculating this correctly that is roughly 7.7 VA. Only half of the 15VA of the original transformer. I also measured the temperature of the transformer and it came in around 95 degrees.

Am I missing something? I'm just trying to understand how the original one failed; seems sufficient enough unless I'm missing something? Or is it temperature based rather than current?

Jamie Hall

Are you also powering the thermostats?

jac34

I guess that’s where I’m getting confused. When u say power, do you mean actually lighting up the screen? Only the zone 1 thermostat lost its power when the transformer died. The other 2 are powered by different units (a/c unit and an air handler) but the red/white wires from all 3 come to this taco board to call for boiler heat via circulator pumps. 

So does the red/white only count as powering the other 2 Tstats? 

Daveinscranton

I believe that phasing error will be at the route of your trouble.  I will be surprised if it is not.  I will be surprised if this is a problem fundamental to the actual boards.

Daveinscranton

I think that the easiest way out of this will be to use a single transformer, with appropriately sized wire, to power everything that wants to see 24 volts.  Wiring/phasing error would do it. 

HomerJSmith

If you think that transformer phasing is the problem, then switch the secondaries of one of the interacting transformers.

There is an easy and a harder, longer, way to determine if the phasing is correct between two interacting transformers.

Daveinscranton

It is possible that the easy, and harder, longer way may have taken 3 years.  (Or however long the first one lasted)

Perhaps better electrical practice to not parallel different transformers of different manufacture with different va ratings, even though output voltage is the same stated 24 volts.  They could be inadvertently paralleled and in odd ways too.

Daveinscranton

One way to safely check for phasing error would be a few 12v incandescent light bulbs, in series, connected in series with the suspect transformer leads as a current limiter when you try swapping leads.  One way they may be brighter than the other.  

hot_rod

Some thermostats go into a battery charging mode occasionally, that is when they pull high current and fry transformers. We had one job that ran fine for over a year, then the stats went into recharge and took out the transformer 

Its not easy determining how the various power stealing stats work, and manufacturers are still tweaking them, the Nest for example, different versions behave differently. Not even tech support fully understands how they work😉

STEVEusaPA

...

jac34

hot_rod said:

Some thermostats go into a battery charging mode occasionally, that is when they pull high current and fry transformers. We had one job that ran fine for over a year, then the stats went into recharge and took out the transformer 

Its not easy determining how the various power stealing stats work, and manufacturers are still tweaking them, the Nest for example, different versions behave differently. Not even tech support fully understands how they work😉

I don't know who to believe anymore, but according to ecobee there are no batteries in their units and can only be powered by the 24v. But, based on my current test that I did, I don't see how a 15va transformer couldn't handle the set up I have. They do recommend a separate plug-in 40va transformer for set ups that have circulators/zone valves. I'm not sure that's just to cover their **** or if it truly would pull more too much power in conjunction with the thermostats. Doesn't seem to given the .32 amps when all zones are calling for heat.

Phasing is way over my head, not sure I even understand that.

Daveinscranton

The reason why electrical phasing is important is that you can easily construct a situation where you have transformers at war with one another.  You can sort out the phasing.  Or power your 24v devices from one, appropriately sized, transformer.  That would take electrical phasing out of the equation.

best wishes 

wmgeorge

As long as the T'stats depend on the C wire to operate and you keep the Rh and Rc separate I do not think you will have any issues as you have it now.

MaxMercy

jac34 said:

Out of curiosity, I ran each zone separately then added one at at time while measure amps with my clamp mm. Interesting as it added about .1 amps per zone..

Honestly, I never trusted the accuracy of current clamps at very low currents, and also to be honest, I don't know how accurate that meter is. I would open the line and insert a proper ammeter in the line to get a sure picture of what the draw off the transformer is.

jac34

MaxMercy said:

jac34 said:

Out of curiosity, I ran each zone separately then added one at at time while measure amps with my clamp mm. Interesting as it added about .1 amps per zone..

Honestly, I never trusted the accuracy of current clamps at very low currents, and also to be honest, I don't know how accurate that meter is. I would open the line and insert a proper ammeter in the line to get a sure picture of what the draw off the transformer is.

I hear ya, maybe it's off a bit but it's no where close to reaching 15va. Just trying to understand how the power robbing Tstat could of really done the damage here when I'm getting these readings. Even if it's underestimating it by 10%, (they claim they are within 2% accuracy) it's not even getting close to the .625 amps the 15va can handle. Something else is the culprit in my opinion. Maybe it was just a junk transformer on that original board and died prematurely, but hard to believe that's the case. 40va transformer seems way overkill, but I may be missing something here.

It's working fine and I'm sure it will be, just trying to understand the HVAC contractor's claim that the Tstat was the issue here causing this to fail.

wmgeorge

On my Flukes you can just either put one side in series with the 10 Amp (max) fused plug in jack on the meter or just make a loop of 10 turns of wire connected in series with the load and divide your clamp reading by 10.