Inrush current/"Bounce back" voltage 120/12v landscape transformer

Status
Not open for further replies.

grasfulls

Senior Member
A client with a rather long driveway had the landscaper install several embedded in asphalt driveway LED lights, cumulative wattage is around 150 watts. The client wants the lights to come on at night whenever there is motion at either end of the driveway. We made use of some spare #10's at the gate to power a motion sensor/photocell rated at 250watts incandescent, then mounted another at the top of the driveway. Evidently the landscaper used a 600 watt transformer, I need to confirm that it is not 300 watts, but I want to assume 600 for my questions. The transformer is within 8 ft of the top sensor. The sensors are in parallel and their outputs were directly to a plug for the transformer. The top sensor failed within 2 days.

We installed a 30amp rated relay to feed the power to the transformer so we can feed the sensors without a common tap from the sensor to its internal contact, we have ordered higher wattage sensors. We are still using a single source of power, we can segregate and put the sensors on a different circuit from the transformer.

My questions:
1) Is it ok to install transformers much greater than the load they will be driving?
2) What are reasons to stay within, say 25-30% of the load, if any reason at all
3) Is there an current inrush or a "Bounce-back voltage or current" that could damage the feeding components?
a) is our relay going to be safe?

I have attached a base schematic.
Thanks!
gare
 

Attachments

  • transformer ltg Model (1).pdf
    45.9 KB · Views: 2

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
My questions:
1) Is it ok to install transformers much greater than the load they will be driving?
2) What are reasons to stay within, say 25-30% of the load, if any reason at all
A oversized transformer will not be as efficient as one that is sized closer to the actual load, but in this case the difference would be pennies.

3) Is there an current inrush or a "Bounce-back voltage or current" that could damage the feeding components?
a) is our relay going to be safe?
I don't think there should be any problem.
 

grasfulls

Senior Member
Inrush current

Inrush current

A oversized transformer will not be as efficient as one that is sized closer to the actual load, but in this case the difference would be pennies.

I don't think there should be any problem.

I have not found the specifics on the reduction of efficiency, I recall reading someplace about no exceeding 33% oversized. But I have found quite a bit on transformer inrush.
Wikipedia:
"When a transformer is first energized, a transient current up to 10 to 15 times larger than the rated transformer current can flow for several cycles."
It is apparently worse when the transformer is torroidal- "Toroidal transformers, using less copper for the same power handling, can have up to 60 times inrush to running current".
End Wikipedia

So I read that as the "rated transformer current", not the connected load. The 30 amp contact can sustain a continuous 30 amps at 120 volts, so 3600 watts, and I am sure it may accommodate the momentary inrush, I just want to make sure. My remaining issue is still sharing the same common source of voltage, though now not immediately within the sensor housing. The relay is about 24" from the transformer (given the cord connection), so not a lot of wire run to dissipate a spike.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150305-0949 EST

grasfulls:

I need clarification relative to your first post. You refer to the total load as 150 W of LED lights.

1. Does this mean an actual load of 150 W or is this the equivalent of 150 W of incandescent load? If it is equivalent load, then the actual steady-state load is about 50 W.

2. Incandescent load if turned on at a voltage peak may produce a peak current spike of 10 to 20 times the RMS steady-state current.

3. What does the word transformer mean? Is this an ordinary magnetic type transformer, or an electronic transformer that uses a rectifier followed by a capacitor input filter, and then a high frequency chopper to a small trnasformer?

4. Any time you have a rectifier followed by a capacitor input filter with low internal and external series impedance, then you can have very high peak currents when connected at a voltage peak with zero charge on the filter capacitors. Both LEDs and electronic transformers may fall in this category.

My questions:
1) Is it ok to install transformers much greater than the load they will be driving?
No problem. But you get greater inrush current, and there is more residual power loss.

2) What are reasons to stay within, say 25-30% of the load, if any reason at all??
No different than question 1. Transformer life will increase as load on it is reduced.

3) Is there an current inrush or a "Bounce-back voltage or current" that could damage the feeding components??
I don't know. This depends upon many factors relative to inrush current. I have no idea what you mean by "Bounce-back voltage or current". I can imagine some things. There is nothing that would be a bounce back current. By bounce back voltage you might mean the transient voltage produced when the circuit to an inductor is opened while current is flowing thru the inductor. This occurs on turn off, not turn on.

a) is our relay going to be safe?
Your relay may be a good approach. What do you mean by safe? With an electro-mechanical relay you will get some transient voltage from the relay coil on turn-off.

You need to disassemble failed components and determine what failed. This way a more intelligent decision can be made relative to a possible solution.

.
 

grasfulls

Senior Member
Current Inrush MLV transformer

Current Inrush MLV transformer

150305-0949 EST

grasfulls:

I need clarification relative to your first post. You refer to the total load as 150 W of LED lights.

1. Does this mean an actual load of 150 W or is this the equivalent of 150 W of incandescent load? If it is equivalent load, then the actual steady-state load is about 50 W.
.

300watt FX Magnetic Low Voltage 120/12VAC with ten ea 13 watt LED fixtures - i just added 20 for the transformer

150305-0949 EST
2. Incandescent load if turned on at a voltage peak may produce a peak current spike of 10 to 20 times the RMS steady-state current.

3. What does the word transformer mean? Is this an ordinary magnetic type transformer, or an electronic transformer that uses a rectifier followed by a capacitor input filter, and then a high frequency chopper to a small trnasformer?
.

Do you think an off the shelf photo/motion sensor can handle that for a few cycles? - It is an ordinary MLV, maybe it should be electronic for this application? Is there less inrush?


150305-0949 EST
4. Any time you have a rectifier followed by a capacitor input filter with low internal and external series impedance, then you can have very high peak currents when connected at a voltage peak with zero charge on the filter capacitors. Both LEDs and electronic transformers may fall in this category.

No problem. But you get greater inrush current, and there is more residual power loss.

No different than question 1. Transformer life will increase as load on it is reduced.

I don't know. This depends upon many factors relative to inrush current. I have no idea what you mean by "Bounce-back voltage or current". I can imagine some things. There is nothing that would be a bounce back current. By bounce back voltage you might mean the transient voltage produced when the circuit to an inductor is opened while current is flowing thru the inductor. This occurs on turn off, not turn on.
.

I think bounce-back may be a bad term, perhaps current inrush and a return voltage on OFF as the field collapses?

150305-0949 EST
Your relay may be a good approach. What do you mean by safe? With an electro-mechanical relay you will get some transient voltage from the relay coil on turn-off.

You need to disassemble failed components and determine what failed. This way a more intelligent decision can be made relative to a possible solution.

.

That last point is a good one, but I have a feeling it will be board level troubleshooting, completely beyond me. By safe, I mean safe form equipment failure. Even though I have introduced the relay, now I get no operation. When I called the manufacturer to ask about minimum wattage, they did not know. So, just a FYI at this stage, I have ordered 2 ea sms50 RAB MotionSensor/PhotoCell and I called their tech support. He told me that there SHOULD be no problem connecting the sensors to a relay but I may get chatter on the relay due to lack of load. He said to connect a .1 uf 400vdc capacitor across the relay coil, that I will find that in the installation instructions as well.

I cannot find a 400VDC, but I do see a 400V non-polarized....

Anyway, I appreciate the help!!
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150304-1943 EST

grasfulls:

Good information. Look at photos P1 (incandescent peak inrush current is about 20 the RMS steady-state current), and P6 ( a magnetic transformer with no load has a peak inrush current is about 27 times its full load rating of RMS steady-state current) at my website http://beta-a2.com/EE-photos.html .

You have a magnetic transformer so its worst case inrush characteristics will be like photo P6, but greater because of the larger size transformer. You could possibly have more inrush as a result of your LED lamp load on the transformer output.

Do the failed sensors fail closed or open? On opening a failed sensor your will probably find a welded relay contact (this would be a failure to the on state if an electro-mechanical relay exists), or some component that looks burned.

Inrush current is probably your problem.

If the sensors use a solid-state relay (Triac or equivalent), then their peak current capability may not be very good.

If the sensor output switch is a Triac or equivalent, then there is a minimum holding current required, and if your relay does not provide enough load to exceed the holding current, then that could be the chatter mentioned. Add additional load across the relay coil and see if that solves the problem.

.
 

grasfulls

Senior Member
MLV inrush, sensor failure, relay

MLV inrush, sensor failure, relay

150304-1943 EST

grasfulls:

Good information. Look at photos P1 (incandescent peak inrush current is about 20 the RMS steady-state current), and P6 ( a magnetic transformer with no load has a peak inrush current is about 27 times its full load rating of RMS steady-state current) at my website http://beta-a2.com/EE-photos.html .

You have a magnetic transformer so its worst case inrush characteristics will be like photo P6, but greater because of the larger size transformer. You could possibly have more inrush as a result of your LED lamp load on the transformer output.

Do the failed sensors fail closed or open? On opening a failed sensor your will probably find a welded relay contact (this would be a failure to the on state if an electro-mechanical relay exists), or some component that looks burned.
.

They fail in what appears to be a partially closed state, so maybe like a triac? I measure 40volts on the switch leg out with the sensor not engaged.

150304-1943 EST
Inrush current is probably your problem.

If the sensors use a solid-state relay (Triac or equivalent), then their peak current capability may not be very good.

If the sensor output switch is a Triac or equivalent, then there is a minimum holding current required, and if your relay does not provide enough load to exceed the holding current, then that could be the chatter mentioned. Add additional load across the relay coil and see if that solves the problem.

.

Will the .1uf, 400vdc capacitor simulate the added load? The tech said that is what it would do, the image in the instruction manual says to put that across the coil of the relay if you are driving more wattage than the sensor is rated. Plus it states vdc but does not show where the + or - go.
Can I put a power resistor across the coil?

I see in your images what appears to be a single waveform spike? I read on wiki that it could be several waves, but it is so over my head. It would be neat to measure the spike on the trans with the load attached and detached to see the impact.

You have been a tremendous help. thanks
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150305-2320 EST

grasfulls:

Inrush transient current from magnetic core saturation or incandescent lamp resistance change at 60 Hz is mostly a 1/2 cycle phenomenon.

A physical switch has very near infinite resistance when open unless there is non-air conductance around the contacts, and low resistance when closed.

A solid-state relay contact has some samll leakage thru the solid-state device, but more likely there is a snubber across the switch and this probably produces a much greater leakage path than the switch itself.

In your test put a load on the output, a 60 W bulb, and now read the voltage. Just a high impedance meter load is very likely to read voltage when the a solid-state switch is off.

A 0.1 ufd capacitor at 60 Hz is about 26,000 ohms. This might help get or keep a solid-state switch on. The fact that a capacitor is being suggested across the relay coil implies a solid-state switch as a mechanical set of contacts would not require the capacitor. Also you must have an AC relay, and not a DC relay. An AC relay has a copper slug around part of the end of the magnetic core.

Will the .1uf, 400vdc capacitor simulate the added load? The tech said that is what it would do, the image in the instruction manual says to put that across the coil of the relay if you are driving more wattage than the sensor is rated. Plus it states vdc but does not show where the + or - go.
the image in the instruction manual says to put that across the coil of the relay if you are driving more wattage than the sensor is rated
This is a strange statement, not at all clear.

There is no +/- marking on a non-polar capacitor. Electrolytic capacitors are polar and must be connected with + to the more positive DC voltage. There are some electrolytic capacitors that are non-polar, essentially a back-to-back pair of polar capacitors.

Can I put a power resistor across the coil?
Yes. For example a small bulb.

Wiki is not quite giving you a good picture.

.

.
 
Last edited:

grasfulls

Senior Member
Addressing the odd statement - plus polar and non-polar - and other stuff

Addressing the odd statement - plus polar and non-polar - and other stuff

150305-2320 EST

This is a strange statement, not at all clear.

There is no +/- marking on a non-polar capacitor. Electrolytic capacitors are polar and must be connected with + to the more positive DC voltage. There are some electrolytic capacitors that are non-polar, essentially a back-to-back pair of polar capacitors.

.

Sorry about the lack of clarity. The instruction manual (pg 6 of attached pdf) for the installation of the pc/ms has an image of using a capacitor, the title of which is "To switch more than rated load". The tech told me to put it there if the relay chatters. The image shows a .1uf 400vdc capacitor, but it is across a 120vac coil, so no putting the positive toward "the more DC voltage". Do you think they should be specifying a non-polarized capacitor?

150305-2320 EST
you must have an AC relay, and not a DC relay. An AC relay has a copper slug around part of the end of the magnetic core.
.

Yes, but it is completely sealed. I will look at the next one I get my hands on that is clear to see if merely by looking at the coil I can tell if it is AC or not.

150305-2320 EST
In your test put a load on the output, a 60 W bulb, and now read the voltage. Just a high impedance meter load is very likely to read voltage when the a solid-state switch is off.
.

I will measure what I can when all of the components are in and reinstalled. I want to thank you yet again for all of the educational input. One would think with all of the MLV loads out here that someone would make an outdoor rated PC/MS with a set of dry contacts, or at least made to work off the shelf.

Last question, would it be better to use a separate circuit for the actual load so the power to the coil of the relay cannot possibly be impacted by any inrush current?
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150306-1121 EST

What grasfulls did for his first circuit is logically correct and it should have worked, and it did logically work, but there were component failures rather quickly. Why? Unknown at this time.

Next he changed the circuit to include a relay. This circuit is logically correct and should have worked. But a new kind of problem arose.

More components can be tried, but unless the causes of the problems are determined, then the problem solution by this method is not likely efficient.

.
 

grasfulls

Senior Member
Inefficient - simple -

Inefficient - simple -

150306-1121 EST

What grasfulls did for his first circuit is logically correct and it should have worked, and it did logically work, but there were component failures rather quickly. Why? Unknown at this time.

Next he changed the circuit to include a relay. This circuit is logically correct and should have worked. But a new kind of problem arose.

More components can be tried, but unless the causes of the problems are determined, then the problem solution by this method is not likely efficient.

.

Inefficient - I am in complete agreement from both a fiscal and time perspective. However, given the cost of the components involved, it will ultimately aid in making for a better installation with respect to controlling lighting in this fashion. I plan on installing all of the remedial components if ever facing this scenario (an unusual one) in the future. If this had been an installation when the house was built, the sensors would have merely provided an input to the existing lighting system and then the lighting system would have turned the lights on. PLUS, I like to learn and when someone who knows a lot more than I is willing to share, well...who can pass that up?

I am awaiting all of the components and will test the functionality after their installation. If there is an order of install/test that may help to pinpoint the failure path, I am all ears. If there is a means for me as a simple electrician to investigate and find the failure point, I am open to that as well.

Thanks!
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150306-2355 EST

grasfulls:

Sorry about the lack of clarity. The instruction manual (pg 6 of attached pdf) for the installation of the pc/ms has an image of using a capacitor, the title of which is "To switch more than rated load". The tech told me to put it there if the relay chatters. The image shows a .1uf 400vdc capacitor, but it is across a 120vac coil, so no putting the positive toward "the more DC voltage". Do you think they should be specifying a non-polarized capacitor?
That the suggested capacitor is 0.1 ufd at 400 VDC tells me it is non-polar (in otherwords you can apply AC across the capacitor), if it were 10 ufd I could not make that statement.

The 400 V rating is good because the peak voltage of a 120 V RMS sine wave is about 170 V.

The statement "To switch more than rated load" and adding a 0.1 ufd capacitor makes no sense. The 0.1 ufd capacitor across the relay coil can not increase the relay load capability.

.
 

Electric-Light

Senior Member
Electronic photo control based on triac can fail when you directly operate an inductive load. The old thermal bi-metal or the kind that makes clicking sound is a better bet, because they have no load interaction issues at all.

The most damaging load for the relay is a string of CFLs and some line voltage Light Emitting Decorations. They have a different pattern of inrush.

Imagine transformer's inrush like someone shoving you with their hand which is generally harmless. CFLs inrush is more like someone throwing a sharp pebble at you at the same spot every day which tear up your skin pretty rapidly.
 

grasfulls

Senior Member
Final Capacitor issue

Final Capacitor issue

150306-2355 EST

grasfulls:

That the suggested capacitor is 0.1 ufd at 400 VDC tells me it is non-polar (in otherwords you can apply AC across the capacitor), if it were 10 ufd I could not make that statement.

The 400 V rating is good because the peak voltage of a 120 V RMS sine wave is about 170 V.

The statement "To switch more than rated load" and adding a 0.1 ufd capacitor makes no sense. The 0.1 ufd capacitor across the relay coil can not increase the relay load capability.

.

Thank you gar. I will continue with my inefficient replace/add/test method, I do not say this negatively except with respect to my own ignorance on how to do anything else with solid state controls. I will also update this thread with the final outcome, unless the moderator feels it should go to a private message. I do think it has a lot to do with lighting control in today's LED implementation market. Perhaps one of your sites could have a "how an electrician can pinpoint failure" section, followed by "the simple solution"?
 

grasfulls

Senior Member
Defiant/Home Depot.Heath Zenith

Defiant/Home Depot.Heath Zenith

Electronic photo control based on triac can fail when you directly operate an inductive load. The old thermal bi-metal or the kind that makes clicking sound is a better bet, because they have no load interaction issues at all.

The most damaging load for the relay is a string of CFLs and some line voltage Light Emitting Decorations. They have a different pattern of inrush.

Imagine transformer's inrush like someone shoving you with their hand which is generally harmless. CFLs inrush is more like someone throwing a sharp pebble at you at the same spot every day which tear up your skin pretty rapidly.

Slowly this sinks into my brain. I had run to Home Depot and grabbed a couple of off the shelf bronze DEFIANT motion sensors (2.2amp rating) to replace the 4.2 amp rated Heath Zenith. Nothing on the package indicates "do not use with" or "minimum load". I called the tech support line - it is home depot. They put me through to the "electrician". When asked "is there a minimum load"? He said "they do not tell us that". HE did say hang a load across the output, if it comes on for 4 seconds in test mode and then turns off, the sensor is working. Well, I hung a 13 watt fluorescent, it just stayed on. I called back, "oh no, use like a 60watt incandescent", I said "the box and instructions do not say that"....

Anyway, I just tore the skin. I am going to wait for everything, just need the capacitors now.
 

grasfulls

Senior Member
well, that was not the final capacitor post = TWO QUESTIONS please

well, that was not the final capacitor post = TWO QUESTIONS please

150306-2355 EST

grasfulls:

That the suggested capacitor is 0.1 ufd at 400 VDC tells me it is non-polar (in otherwords you can apply AC across the capacitor), if it were 10 ufd I could not make that statement.

.

I will not ask about why 10 is different...
Do you think I may then install non-polarized capacitors?
Can the vdc capacitor be installed either direction across AC?
 

grasfulls

Senior Member
old bi-metal clicking

old bi-metal clicking

Electronic photo control based on triac can fail when you directly operate an inductive load. The old thermal bi-metal or the kind that makes clicking sound is a better bet, because they have no load interaction issues at all.

Where might one find one like this. I have not found anyone with an exterior rated motion sensor with a separate set of dry contacts to which I may run a separate power source. Everything has a common input so I am assuming the coil is being directly impacted when the sensor engages the output. Does this not also mean, even if it were a dry contact, that the inrush still have a damaging relationship with the coil?
 

iwire

Moderator
Staff member
Location
Massachusetts
Electricians installing capacitors in lighting control circuits?

This sounds absolutely nuts to me.
.
If you are worried that transformer inrush will kill the relay my suggestions would be to use a larger relay or simple leave the transformer on 24/7 and switch the secondary side.

In my personal opinion reinventing the wheel is a disservice to the customer because in the future people will not understand the reasons or the methods.

Keep it simple, keep it consistent with the other eight billion lighting control circuits.

To each their own. :)
 
Status
Not open for further replies.
Top