A/C kicks on... whole house "dims"

[brian john]

Problems with attachments duplicate post SORRY

 

[brian john]

Did a job with light flicker yesterday, I have attached the disturbances as well as a text print out of the voltage and current levels.

This house had an unusual number of incadescent light firtures 150+, and 7 HVAC units. This house had a 200 amp service that was upgraded to 600 amp, appears the utility left the 200 amp triplex in use, about 150" changed to 250 AL undereground 150, plus what appears tro be a 30 KVA pole mounted transformer (300 feet away).

Flicker is not just load, conductor lenght or transformer sizing dependent, it is also end user sensitivite. I have flicker but never really notice it, unless the wife says something. I have been to customers houses where it PLAIN WAS DRIVING the homeowner crazy and max VD recorded was 3 VAC. Others are just worried they have connection issues (a legimate concern).

 

[mhulbert]

Regulating Dimmer

Regulating Dimmer

Winnie,
You need not look further than Lutron for a regulating dimmer. Of course, you need to get a grafik eye or dimming panel to have it! They call it RTISS, and I see they now make one for electronic transformers, called the RTISS-TE. I have installed units that have this technology, but have not had the equipment to test it and see if their claims are true.

Check out these white papers (PDF's)...
http://www.lutron.com/technical_info/pdf/Power_quality.pdf
http://www.lutron.com/technical_info/pdf/RTISS-TE.pdf

 

[winnie]

Thanks for the links to the Lutron info.

I was finally able to read the Lutron RTISS paper. It does a good job of describing why dimmer based systems may be _much_ more sensitive to power quality issues than a non-dimmed incandescent lamp, and it crudely describes what they do to deal with these issues.

In a nutshell, ordinary 'phase control' dimmers use a solid state switching element to permit current to flow only for a portion of the AC cycle, effectively lowering the RMS voltage presented to the lamp. In order to chop up the supply voltage, this solid state switching element needs to be triggered in precise synchronism with the AC supply voltage; for example if you want 50% output you need to trigger the switching element half of the way into each 'half cycle' of the AC supply voltage.

Apparently line disturbances (voltage variation, noise, distortion, etc) will cause the dimmer to improperly time the 'turn on' pulses. The relevance to the original poster on this thread is that if there are lights on dimmers, this will amplify any problems with the electrical system.

The RTISS system uses filtering techniques to precisely synchronize to the line and properly time the triggering of the semiconductor switch, even in the presence of power quality problems. By doing this, when the dimmer is set for 50% output, it will remain at 50% output.

As far as I can tell, there is nothing in the document which describes an attempt to _compensate_ for supply voltage fluctuations. I take this to mean that if the supply voltage drops from say 120V to 110V, and the dimmer is set for 50% output, then the output will drop from 60V to 55V, which is quite possibly better performance than most other dimmers.

It seems to me that there is a further stage that one could go, where the output actually gets regulated to compensate for supply voltage changes, such that if the dimmer is set to 50% with 120V supply, then should the supply fall to 110V, the dimmer would automatically adjust to 55% output to maintain a constant 60V on the output. I don't think that the RTISS system does this, but perhaps this is a feature that they didn't trumpet in their ad copy.

-Jon

 

[mhulbert]

Jon,
I was under the impression that the dimmers accounted for voltage variations by "boosting" the dimmer setting on low voltage. I re-read the paper, and I'd agree that it is not clear on whether or not they actually do this. I need to talk to Lutron and see...

Mike

 

[dduffee260]

and unfortunately this particular customer doesn't care about cost- they want the problem fixed.

Then fly me out there and I bet we can figure something out. What is it that member on here has on their signature? " I can fix any problem as long as you draw it on the back of a big enough check"?

 

[allenwayne]

I like the quote...... How much ?????? Don`t worry it is only money and the government prints it ever day

 

[chevyx92]

Hey all, thanks for taking the time to help. Here's the scenario; we have a large residential project we completed some time ago. The homeowner is complaining that when the (4) A/Cs kick on, the lights in the entire house go dim. There is a 400A 120/240V service on the house, and a total of 3 subpanels. THe ACs are run from a remote subpanel on a 2nd floor. This has gone as far as a) the utility company has come and set a new transformer outside, b) the breaker feeding the sub panel has been relocated from the lighting breakers, c) ampereage and voltage have been measured and found to be within limits..

We're out of options. Would a voltage regulator/stabilizer, and/or a power conditioner help? If so, where would i find one that can effectively monitor an entire house? I imagine this would be a common problem in factories, hospitals, etc... how do they deal???

Any help would be appreciated!

I would like to know what the tonnage on the AC's are and what size wire is feeding each? Also I can't imagine that all 4 are ALWAYS turning on at the same time. Unless they're on one zone which I wouldn't think they would be. I would think they would each have a stat.

 

[brian john]

Chevy In the job I noted above a single unit resulted in the VD noted, I have completed numerous of these residential "FLICKER" investigations and as noted I have seen as little as 3 volts result in flicker, this was from a single phase load running current was 7.some amps, I forget what the captured inrush current was, but I will try and located the report and post.

 

[dbuckley]

The RTISS system uses filtering techniques to precisely synchronize to the line and properly time the triggering of the semiconductor switch, even in the presence of power quality problems. By doing this, when the dimmer is set for 50% output, it will remain at 50% output.

As far as I can tell, there is nothing in the document which describes an attempt to _compensate_ for supply voltage fluctuations.

For most dimmers, that is correct. To get the timing right requires just enough electronics to determine if there is voltage present on the mains, or not, so as to detect the zero crossing point. The timing calculations and zero cross detection filtering is done in software, as these dimmers universally use a small microcontroller. To compensate for mains voltage requires measuring the mains voltage, which requires a few more components, but cruicially it also requires a bigger, more expensive microcontroller with an A/D converter.

Excellent dimmers, such as good stage lighting dimmers, track both voltage and frequency of the incoming mains, and the perceived output of these dimmers is rock solid despite really naff things happening on the input, such as bad gensets with poor voltage and frequency regulation.

It can all be done, it's just a matter of price. A $2 dimmer is not a very sophisticated thing.

(All this is really important in countries that use "ripple control", which is data carried across the power wiring for controlling streetlights and switching meter tariffs. Bad dimmers can get awfully confused and flicker during ripple transmission.)