VFD as a LED light dimmer 4

[goldeneagle272]

Hello All,

Does anyone has any experience using a VFD as a light dimmer? I am a recent graduate and I have just started working at this company not too long ago, my boss wants me to take a ATV12H018M2 (240V Single Phase) and use it as a dimmer. I know there are triac, sine and IGBT dimmers out there in the market but they seem to be either incompatible or expensive.
Anyhow, I have read the manual and since it is meant to be for a motor I don't understand what configuration menu to use for my purpose. I have so many questions. Firstly, I would like to know about the drive outputs u,v,w and GND. How do I connect a 3 wire lamp to these outputs? U-Hot, V or W-Neutral and GND - Ground wire?
Secondly, if I connect drive output to a oscilloscope would I be able to see a PWM signal? I have so many more but I want to start with these two first. I think what he would like me to do is control when it fires or crosses zero. I'm so clueless and the manual seems to be not so helpful either. I would really appreciate if someone could guide me through this. Attached please find the user manual for the VFD I have been using.

Thanks,
GoldenEagle

 

[Skogsgurra]

I was assigned a similar task on my first "real" job. At that time, there were no such things as VFDs. I remember starting out with the assumption that I should find a good solution but I soon realized that this was a test to see how independent and analytical I was. Or just clueless.

I think this is a similar situation. One thing you shall NEVER try is to connect GND to the VFD output. N is OK, but not the GND.

Make sure you understand the specs for the LED first. Is it one single led? Or a string of LEDs? Is there any driver involved? Or series resistors?

Then read up on VFDs. They are all PWM these days. Connect between U and V (or W) you will probably fry something when doing so. Then find out why it didn't work. After adding a full bridge rectifier, some resistors and capacitors, you may have something that blinks - or burns.

You should chose scalar control, aka U/f.

Varying the frequency and thus the output RMS voltage, you will notice that the LED's output varies between dark and bright at double the set frequency. If that is what your boss wants, then show him that result. If it isn't (which I think) then tell him that it will not work. And explain why.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[jraef]

Neutral here in North America is always Ground potential, so if you are here, do NOT connect the output of the VFD to anything that needs the Neutral to work. Instant VFD death.

So here's what you are left with. You CAN use it, but only if the lighting is Line-to-Line powered. I would also be VERY hesitant to connect it to anything other than incandescent lighting, or rather nothing that has a ballast. Unlike a PWM UPS, the output of a VFD is not at all filtered (for the most part) so it may damage another SMPS or magnetic ballast by over heating it with the harmonic rich output. By the time you got done buying the appropriate filters, you would likely have spent less using a device designed for this purpose.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[goldeneagle272]

Thank you guys for your response.
Skogsgurra, I thought I would give it a try and tell him that it won't work too. But, when I talked to my colleagues(30+ years experience) they said they did this last spring and were able to control the brightness of the lamp using the same VFD I have with me now. They did by using the controls and codes from the user manual but they have not recorded the procedure anywhere. So, I have no choice but to say him that I don't understand how to use the terms, codes and configurations on the manual for our setup. We do use full bridge rectifier and current sink (transistors and resistors) as the lamp driver.

jraef, yes I live in NA and I'll keep that in mind. Thank you. As in the harmonic output, I was under impression that there is stage (DC bus/link) just for filtering in a VFD.

 

[sibeen]

oldeneagle272, the DC is filtered, but the output of the VFD is a PWM waveform - without any filtering. Lots and lots of harmonics in the waveform, as jraef has stated.

 

[jraef]

By the way, we used to test VFDs (and soft starters) by hooking them up to incandescent heat lamps on the test bench. You just have to be cognizant of the voltage ratings. I can attest to the fact that a 120V rated lamp glows brilliantly at 480V, albeit very briefly.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[LionelHutz]

In North America the neutral is grounded at the source. You should take a look at the circuit, but I would expect you can put power on the neutral without any issues. The incandescent base (E26 base?) lights use the outside screw shell as the neutral to reduce shock hazard.

Now, it sounds like this lamp is 120VAC, so you could use a 240:120 transformer and this would filter most of the PWM signal out of the waveform on the secondary side. You can also ground one of the transformer secondary terminals and call it the new neutral.

If you factory reset the drive it would likely work as programmed. Set the motor voltage to 120V if you direct connect the lamp. Give a start command and see what happens. Wear the proper PPE.

 

[jraef]

For some reason, up until now I had not noticed the "LED" reference in the title. Not going to be feasible. Here in the US anything you install (that will be subject to National Electric Code authority inspection) must be NRTL listed, by someone like UL. UL has a rule on dimmers that a dimmer MUST be tested and listed SPECIFICALLY with any non-incandescent bulb. When you buy a wall dimmer, it will come with a list of specific bulbs it has been tested with, technically you cannot use it with untested bulbs. So in effect, by doing this YOU are taking on that responibility and unless you have an extra $50k floating around for UL listing of that VFD as a dimmer for non-incandescent lamps, you will be in code violation. Maybe that's OK for your boss because he is not planning on pulling a permit and getting inspected, but he should also check with his insurance carrier. The underwriters often throw in clauses which stipulate that any damage caused by a fire determined to have been started in equipment that is not used and listed for the intended purpose, will not be covered by insurance. That's usually an idea killer on things like this.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[LionelHutz]

Jeff - The big question - Is this for an installation or just to experiment?

 

[jraef]

I am a recent graduate and I have just started working at this company not too long ago, my boss wants me to take a ATV12H018M2 (240V Single Phase) and use it as a dimmer.

Jeff - The big question - Is this for an installation or just to experiment?

Given tha above statement, I see your point. His boss may just be giving him a "what if" sceanrio.

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[goldeneagle272]

Jeff and LionelHutz, It is just an experiment for now but we might potentially use them with our lamps if we can get them to work like we want.

 

[jraef]

Good luck with that...
Let us know how it turns out, preferably with pictures of the fireworks.

Take precautions...

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[goldeneagle272]

So far this is what I have got.

Around 200V AC i/p was applied to a VFD to get 120V o/p(across a L and GND).
The moment I apply voltage to a non-running VFD it generates a surge output and this surge is enough to a flick a lamp.
When the lamp was powered through a VFD, I have tried changing the frequency using a jog dial on a VFD. I was able to see change in pulse width on oscilloscope but I have not seen any change in the brightness of the lamp. And the last but most annoying part was when the VFD was running, a buzz sound was heard from a lamp, the buzzing sound changes with the change in frequency and stops when I stop the VFD. I think the buzzing has to do with the 8KHz output of a VFD.

Sorry Jeff, no fireworks hence no pictures.

 

[Skogsgurra]

You don't get much variation on voltage between output and GND. You need to connect between two output lines.

Told you so.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[itsmoked]

Which may mean you need two bulbs in series to meet the voltage.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

...or produce Jeff's fireworks...

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[oldfieldguy]

Or if you're planning on using an 'off the shelf' LED lamp assembly, be cognizant of the the fact that the input leads are rated for a specific range of input voltages. Internally, the fixture's power supply rectifies and regulates the actual output to the LED array.

I strongly suspect that if you vary the input voltage, specifically lowering it to dim the array, the poor power supply will do its very best to maintain the nominal output of the LED's until it either shuts down or dies trying.

It might be interesting to try. I'd think I want to get a lot of instrumentation around to see what happens to voltages and currents and input and output waveforms.

old field guy

 

[goldeneagle272]

Skogsgurra, I think the voltage across lines (U, V and W) within themselves was same to what I saw between a line and GND. I'll double check on that tomorrow.
I don't get why changing the frequency is not changing the RMS output am I doing something wrong?
May I know what is a scalar control U/f you have mentioned above? Is it a kind of VFD with that control?

Thanks

 

[itsmoked]

Why would it change the RMS anything? The VFD is going to deliver the same RMS at all frequencies because that is what determines the delivered hp. Just because the user selects a different frequency doesn't me they want only a 1/4hp instead of 1/2hp.

If you want to effect the RMS you need to adjust the programmed power/hp that the drive is set for.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

Not quite so. Voltage and frequency are proportional to each other up to base speed (in scalar mode). It is more about grid voltage and PWM voltage being superimposed when you connect between output line and ground. I shall measure on a VFD on my bench. Stand by!

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.