Paralleling two transformers

hot_rod

I have put two 24V 40 Va transformers together in parallel, when phased properly it works fine. I've been told they should be identical transformers, same brand, same voltage output. Spark test the secondary to assure they are phased.

What about a 12Va with a 40Va, in parallel?

EdTheHeaterMan

Why?

hot_rod

Looking for a simple way to add more Va to switching relay boxes.
Caleffi zone relay boxes allow you to plug in two 40Va transformers for 80VA.

But the switching relays have a 12Va populated on the board. Since we use the same enclosure, extra transformers could snap in, but would need to be wired properly.

A bit of trial and error and it seems to be working.

Mainly to add enough Va to power smart/ wi fi stats.

I've heard multiple reasons why some brands of those stats pull high current. One thought is when they go into battery recharge a year or so down the road they overload the transformers. Another thought is the wi fi function pulls the current? Do you know?

Here are a couple of my trainer boards I'm experiment with. Luckily they have circuit protection on the transformers

HomerJSmith

I'm not an electrical engineer, but I play one on TV. I have never read anything about transformers in parallel having to be the same except for voltages on the primary and secondary, but whata I know.

I've talked about the spark test on this site before and receiving a lot of derision. Using a volt meter works, too, and I posted that procedure as well.

The spark test consists of tying the primaries of the two transformers together and tying one secondary from each transformer together, and then striking the other two secondaries, quickly, against each other and observing the resulting spark and then reversing the secondary of one transformer and then striking the two untied secondaries against each other, again, and observing the spark. The spark that is the smallest is the connection that is phased correctly.

I don't see a reason why the transformers have to be matched as to VA or brand. Maybe, inductive reactance?

hot_rod

HomerJSmith said:

I'm not an electrical engineer, but I play one on TV. I have never read anything about transformers in parallel having to be the same except for voltages on the primary and secondary, but whata I know.

I've talked about the spark test on this site before and receiving a lot of derision. Using a volt meter works, too, and I posted that procedure as well.

The spark test consists of tying the primaries of the two transformers together and tying one secondary from each transformer together, and then striking the other two secondaries, quickly, against each other and observing the resulting spark and then reversing the secondary of one transformer and then striking the two untied secondaries against each other, again, and observing the spark. The spark that is the smallest is the connection that is phased correctly.

I don't see a reason why the transformers have to be match as to VA or brand. Maybe, inductive reactance?

I've watched a handful of You Tube, from HVAC guys to engineers, they all claim the Va need to match. One even suggested un-wraping some of the secondary windings to match same Va transformers exactly!

I did the spark test, got them wired properly after one fuse pop. It seems to be working, transformers are not heating up. So, essentially I should have 52Va available?

Any thoughts on why some stats draw high current?

Not a factory approved hack, by the way.

Pughie1

Hot Rod, used to use this method when adding additional VA to Carriers early heat pumps.
1. Connect 115 volt primary leads from additional transformer to same points as primary leads of existing
transformer/
2. Attach one 24 volt secondary lead from new transformer to terminal "C" of existing transformer.
3. Apply power & measure voltage between unconnected 24 volt lead on new transformer and terminal "R"
on existing transformer.
If voltage is "0" transformers are in phase.
If 50 volts (approx) is measured reverse primary leads of one transfer & recheck for " 0" volts.

Perhaps this will help.

John Pughe

HomerJSmith

"why some stats draw high current?" The design of the circuit, I would think. I don't know whether the stat has a charging capacitor or a battery. They don't publish their designs. I would think a charging capacitor as batteries wear out (limited charging cycles) and are larger. My supposition.

As far as matching transformers parameters, I'm not as much interested in the consensus as I am interested in knowing why.

Ironman

Two transformers of different VA rating can be paralleled if three things are observed.

https://pressbooks.bccampus.ca/singlephasetransformers/chapter/paralleling-transformers/

hot_rod

Pughie1 said:

Hot Rod, used to use this method when adding additional VA to Carriers early heat pumps.
1. Connect 115 volt primary
transformer/
2. Attach one 24 volt secondary lead from new transformer to terminal "C" of existing transformer.
3. Apply power & measure voltage between unconnected 24 volt lead on new transformer and terminal "R"
on existing transformer.
If voltage is "0" transformers are in phase.
If 50 volts (approx) is measured reverse primary leads of one transfer & recheck for " 0" volts.

Perhaps this will help.

John Pughe

Yes, thanks that method seem a bit more professional. Although R&C are not always labeled.

mattmia2

the resulting power source may no longer be class 2. technically the low energy section of the code says you can't tie 2 class 2 power supplies together.

Jamie Hall

The article @Ironman provided covers it quite nicely. Itis quite reasonable to assume that two transformers of the same rating from the same manufacturer will match well. It is quite unreasonable to assume that two transformers of different ratings will match, and if they even if do, they they won't share the load in accordance to their rating. The best case will be that they will share the output equally -- so that in your case, 40 VA and 12 VA -- you will almost certainly overload the 12 VA transformer if you draw more than 24 VA from the circuit. It may not fry immediately, but it will fry. In most cases, the phase angle will be slightly off, resulting in those circulating currents mentioned, which will overheat both transformers until the smaller one goes poof.

Steamhead

@hot_rod , I have never liked or recommended paralleling transformers. On a late-night no-heat call, it's way too easy to get the wires reversed and blow them.

With that said, in the second pic you posted, the existing transformer plugs into a jack on the board. There is second, identical jack right above. If that jack is configured to accommodate a second factory-supplied transformer, it would overcome this problem.

HomerJSmith

Re-hash. Should work for parallel transformer.

ChrisJ

Jamie Hall said:

The article @Ironman provided covers it quite nicely. Itis quite reasonable to assume that two transformers of the same rating from the same manufacturer will match well. It is quite unreasonable to assume that two transformers of different ratings will match, and if they even if do, they they won't share the load in accordance to their rating. The best case will be that they will share the output equally -- so that in your case, 40 VA and 12 VA -- you will almost certainly overload the 12 VA transformer if you draw more than 24 VA from the circuit. It may not fry immediately, but it will fry. In most cases, the phase angle will be slightly off, resulting in those circulating currents mentioned, which will overheat both transformers until the smaller one goes poof.

I've thought about it and I'm thinking the 12va wouldn't be overloaded.

My reasoning is this.
Let's say you have a 40va and a 12va xformer and they both run around 27v unloaded.   As you load them down the 12va is going to drop faster than the 40 so the 40 will pick up the slack.

Just because the output of the smaller one drops more than the bigger doesn't necessarily mean it's overloaded. 

@hot_rod unfortunately people say a lot of things and they often aren't true.  I'm talking about the YouTube people not Jamie.

 Unwrapping a winding seems like it would drop voltage not current carrying capacity.  That would go by wire size not the turns.

I don't know though, I've always matched them and never really put thought into it.  A simple bench test would prove it though, just make sure you go extreme. Meaning a small and a much bigger one and load the circuit down and see what happens.  If the smaller one gets too hot you know it'll have an issue.

mattmia2

MikeAmann said:

Ok, with that solved, can you power an LED with AC?

yes. with the appropriate sized resistor and a diode in parallel biased the other way or a second led biased the other way so it conducts both ways and doesn't exceed the piv of the led.

hot_rod

Steamhead said:

@hot_rod , I have never liked or recommended paralleling transformers. On a late-night no-heat call, it's way too easy to get the wires reversed and blow them. With that said, in the second pic you posted, the existing transformer plugs into a jack on the board. There is second, identical jack right above. If that jack is configured to accommodate a second factory-supplied transformer, it would overcome this problem.

Yes the Caleffi zone valve relay boxes all come with two transformer molecules plug in ports
So the 3,4,5 or 6 zone can all have 80Va
The 5 and 6 version include both transformers in the box, or add a replacement transformer to the 3 &4, that we have with molex connectors on them. So a safe UL listed option for the zone valve box.

The common installer question that comes in is how to use 4 nest thermostats on a pump relay box that has a 12Va transformer built on the board? Or any brand of power sterling stat.

My though was to find a way to do this within the relay boxes that have only 12 Va in them.

Although any modification does negate the UL listing, even if done safely and correctly, I’m told.

My prototype shown is working, but I need to load it up as Jamie mentioned

The post that @Ironman attached claims different Va transformers can be matched successfully.I think the are talking large power line transformers.

mattmia2

hot_rod said:

The post that @Ironman attached claims different Va transformers can be matched successfully.I think the are talking large power line transformers.

Measuring the current in the various branches unloaded and under load will quickly answer that question. I think it has a little to do with the exact design of the transformer and how the terminal voltage relates to the current of both transformers.

EdTheHeaterMan

So @hot_rod Bob is now in the R&D Department... NICE!

Jamie Hall

Truth to tell with the little guys we play with -- so long as they are approximately in phase (see the tips above) it probably doesn't make much difference. Where it starts to get fun is with bigger transformers. Never mind paralleling bigger generators.

hot_rod

Used to used to be a fellow from Honeywell that hung here on the wall, 2003-2008 years. He had good answers to all control questions and Honeywell specific questions. BillW I think was his handle?

savy hydronic and HVAC troubleshooters always kept a 75va transformer on the truck. As @Steamhead mentioned, sometimes you had to parallel a couple 40’s to get the heat back on. I think it was Bill that was one of the first to confirm it could be done and how to do to properly, here at HH.

hot_rod

Now you made me go digging!  I printed off the post from May 2000👍🏻

Found  some post even older from Paul with an interesting email address. S8610U@aol.com. He knew a lot about controls also, maybe a HW guy?

mattmia2

hot_rod said:

Used to used to be a fellow from Honeywell that hung here on the wall, 2003-2008 years. He had good answers to all control questions and Honeywell specific questions. BillW I think was his handle?

I remember billw. I hadn't been here for a while and he was gone when I came back. I've wondered what happened to him a few times.

Solid_Fuel_Man

Just a side note, those automotive fuses are only good for 32 volts I think....not 120V. Use a glass fues with the little plastic twist lock fues holder. 

I think the real answer for Caleffi and Taco is that they should be making relay boxes with a 80va transformer and the "C" terminal at every zone. 

For what it's worth, I always just take out the small transformer and put in a 75 when I have more than 1 smart stat. 

ChrisJ

Solid_Fuel_Man said:

Just a side note, those automotive fuses are only good for 32 volts I think....not 120V. Use a glass fues with the little plastic twist lock fues holder. 

I think the real answer for Caleffi and Taco is that they should be making relay boxes with a 80va transformer and the "C" terminal at every zone. 

For what it's worth, I always just take out the small transformer and put in a 75 when I have more than 1 smart stat. 

Yeah I didn't notice that before.

Plastic blade fuses are ok on 24vac but they have no business being used on 120vac stuff.

hot_rod

Solid_Fuel_Man said:

Just a side note, those automotive fuses are only good for 32 volts I think....not 120V. Use a glass fues with the little plastic twist lock fues holder. 

I think the real answer for Caleffi and Taco is that they should be making relay boxes with a 80va transformer and the "C" terminal at every zone. 

For what it's worth, I always just take out the small transformer and put in a 75 when I have more than 1 smart stat. 

Good catch on the fuse, thanks.

The Caleffi ZVR relay boxes for zone valves come with two 40Va in the 5 and 6 zone version, the extra transformer can be added to the 3&4. The 120V side has a self resetting fuse. All our relays have R,W and C connections for stats, FYI.

Trying to decide if adding a 40Va or more to the pump relay box is worth the cost and having it run through UL again. We see more and more smart stats being used and the 12Va currently on the board doesn't provide adequate Va.
I was just looking at work around options, with the board we have now that has the 12Va included.

Requests like the one attached:

Dave Carpentier

How sensitive are those Smart stats to voltage drop ? Voltage drop (loss) on a wire is related to the current flow and the resistance (gauge) of the wire. Probably not applicable for a regular home, but a mansion or commercial building maybe ?

mattmia2

Dave Carpentier said:

How sensitive are those Smart stats to voltage drop ? Voltage drop (loss) on a wire is related to the current flow and the resistance (gauge) of the wire. Probably not applicable for a regular home, but a mansion or commercial building maybe ?

It probably gets regulated or converted to 3.3 v or 5v so it will never be an issue.

mattmia2

@hot_rod a single 280 watt source would not be class 2. there would be too much energy available to a fault that it could start a fire. It would have to be divided in to multiple smaller sources or at least each circuit would have to have individual overcurrent protection that was smaller than the output of the source(i don't know the exact rules, i only know the intent). Cascadable zone controllers would make the most sense here where they have inputs and outputs to connect things like the end switches/heat call together and some way to make a priority call global over all the controllers. It is also of note that each individual connection to a t-stat only carries 1a so using #16 control wiring would not be necessary.

hot_rod

mattmia2 said:

@hot_rod a single 280 watt source would not be class 2. there would be too much energy available to a fault that it could start a fire. It would have to be divided in to multiple smaller sources or at least each circuit would have to have individual overcurrent protection that was smaller than the output of the source(i don't know the exact rules, i only know the intent). Cascadable zone controllers would make the most sense here where they have inputs and outputs to connect things like the end switches/heat call together and some way to make a priority call global over all the controllers. It is also of note that each individual connection to a t-stat only carries 1a so using #16 control wiring would not be necessary.

I agree, just daisy chain the relay boxes with 40 or 80 Va onboard each box, would cover large zoned options. And you get a factory built, UL listed solution.

That was a typical call or question tech support gets often that I attached above.