20W LED tube interchangeable with 20W fluorescent tube. 11

[grigson]

Hello,

We want to put a 20W LED tube into a socket for a fluorescent tube with a high frequency, switch mode fluorescent ballast driver.

...we dont want to remove the fluorescent ballast as it would be inconvenient.

Are there any standard solutions for this?

-we still want it to be quite efficient, so dont want a linear driver to be added in for the LED current control...we want efficient switch mode current control of the LED current.


...Ultimately, we would like to design a 20W LED tube (containing auxiliary circuitry and converter , etc) which could run off ANY fluorescent ballast.
..that is, it should be able to run off either a 50Hz magnetic ballast, or any high frequency switch mode fluorescent ballast.

....do you know if standard solutions exist for this?

 

[VEBill]

The little detail that this project is perhaps intended for passenger aircraft was not revealed until just a few posts back. Of course it changes everything, making most of the discussion above a bit of a waste of time.



Way deep down inside each million dollar aircraft avionics modification you'll find several hours of 'actual' engineering work (no offense intended to all the *.*ilities folks that also perform extremely important work).

I suspect that the aircraft owners would prefer to have the ballasts removed to save weight, power, and improve reliability. Perhaps they can be replaced with dummies or PS with a "simple" (LOL) box swap - you won't find wire nuts on aircraft, the ballasts should be plugged in.

 

[grigson]

I must admit, i thought, for example, removing fluorescent ballasts from the isle of a passenger aircraft and replacing them with LED drivers is just a short, cheap job for a guy with a screw driver and a few other simple tools?

 

[MacGyverS2000]

Is it just me, or do you keep taking on projects where there's no firm grasp on actual requirements, actual markets, etc.?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[VEBill]

The approach would be to create a driver (if the LED tube requires an external driver) that was "Form, Fit" compatible with the ballast. The third F (Function) would obviously differ.

The Airworthiness / Safety / Human Factors folks should rightfully object if the LED tube fits into the same sockets as the Flourescent tube....unless you followed the original proposal (that brings the other problems listed above).

The same consideration applies to your new Driver box; people will ask how you ensure that they won't get mixed up and what happens if they do get mixed up. This can require very careful analysis and may demand an extremely clever solution to solve the quandary.

Dan mentioned "requirements" (post just above). Oh. My. Gawd. This sort of project could easily call up a dozen thick specification documents, each with hundreds and hundreds of applicable requirements. Done correctly, it might require a dedicated DOORS SME working for hundreds and hundreds of hours to support PoC and formal Airworthiness. A full suite of qual tests can eat $100k.

PS: are you adhering to the NDA?

 

[grigson]

Well, somebody's doing it already.............

http://www.everled.com/led-lighting-products/everled-tr/

.....I cannot understand how they are managing to sell these.

-if they can sell these then they can sell ice to Eskimo's.

Now matter what the EverLED-TR is like, it can never be anywhere near as good as a LED tube with matching LED Driver.

...i notice in the instruction manual pdf doc, they ask customers to report "unusual operation" to them.

...if the overall product catches fire....who is liable i wonder?.....is it the fluorescent ballast manufacturer, or EverLED?

....I think i'm now going to go away and find a machine that converts beer into brandy....somebody must be doing that somewhere too.

 

[itsmoked]

LED drivers are switching current drivers. It's not hard to build one that takes in all the typical ballast outputs including striking spikes, rectifies it, and uses the result in the controller.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[grigson]

I agree that the ballast output has to be rectified.

However, we cannot put a smoothing capacitor after the rectifier.....because.....

There is inrush into such a storage capacitor at start-up...
..and we dont know what kind of current limiting the fluorescent ballast features.....in fact, we have no circuit details of the fluorescent ballast at all, because the fluorescent ballast companies dont actively want us to put LED tubes in their fluorescent tube sockets.

..also, the nature of current draw into a rectifier/smoothing capacitor is in the form of spikes of current occurring at the peaks of the input voltage.
...this is not a viable load to put on the output of a fluorescent ballast...

...the load has to be pretty much the same as a fluorescent tube.....or basically look like a resistor.....

...so what is actually needed is a Power Factor Corrector type stage.....that is, a stage which will monitor the voltage at the ballast ouput and make the current proportional to it...

....this is not so easy as for a normal mains PFC controller, because the frequency at the output of the ballast is anywhere between 50Hz and 70KHz...depending on the ballast.

So you are looking at the first stage of your interface circuitry being a highly customised Power Factor Correction stage, almost certainly controlled by a DSP or high-end microcontroller.

The current feedback loop in a mains PFC stage is a couple of KHz.......it has to be much more than the line frequency.
-With a 70KHz fluorescent ballast, the current feedback loop bandwidth will have to be much much higher than 70KHz.....that in itself makes the solution expensive.

We have already clocked up quite a few dollars for the solution so far, and we're nowhere near finished.....

....the PFC stage will just be able to shovel the energy to its output......and it will have to be a flyback PFC....because boost type PFC's don't isolate the output capacitor from the input...and the inrush problem rears its ugly head again.

After our flyback PFC stage, which is software controlled, we will then need another switching stage to provide the specific current regulation to the LEDs.

As you know, this is a lot of engineering effort for very little in return.
....the resulting "hybrid" ballast is going to be much poorer in performance than even the simplest and cheapest LED tube/LED Driver combination.

...What's more, the fluorescent ballast will, in time, fail.....and the customer will try and replace it......but they will find that its probably gone obselete......and then they'll be left with this highly expensive LED tube, with all sorts of electronics in it, which is now useless to them.

It'd be much cheaper to simply unscrew the fluorescent ballast in the first place and put in a LED Driver.

After all, as mentioned, the fluorescent ballast is going to fail anyway and need replacing, so why not simply replace it right away and put in a cheap simple LED Driver for the LED tube.

I think the EverLED product is for individual householders who want to dabble in a LED product because it happens to tickle their fancy.

....i cannot see how this solution can provide any kind of serious marketing solution for a corporation wanting to replace thousands of lighting fixtures....its not financially viable.

 

[MacGyverS2000]

It'd be much cheaper to simply unscrew the fluorescent ballast in the first place and put in a LED Driver.

Didn't we say that already?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[VEBill]

Hints for plug-compatible avionics:

1) The LED Driver module (assuming it's plug compatible with the existing aircraft wiring, replacing the ballast) has to be certified safe in case it's accidentally connected to a legacy fluorescent tube. This should be trivially easy.

2) The LED tube (assuming it's plug compatible with the existing sockets) has to be certified safe in case it's accidentally connected to the legacy ballast. It doesn't have to work, it just has to be safe in all respects. (It can even be damaged in these circumstances, but that's a very poor selling feature. Better if it's not damaged.)

3) One should also write up a 'white paper' to make sure that the aircraft wiring can handle the lower voltage and higher current. 20W is so low that this is not likely to be a problem, but it has to be proven and certified.

 

[ScottyUK]

"....i cannot see how this solution can provide any kind of serious marketing solution for a corporation wanting to replace thousands of lighting fixtures....its not financially viable."

While I agree that it is a horrible solution, don't dismiss the costs of re-lamping as a justification for replacement. A fixture with an iron ballast will last 40 years - my facility has hundreds of this vintage - but they are re-lamped every couple of years. Access is expensive because of the nature of the facility, so if that could be pushed out to an interval of, say, ten years then there would be a very worthwhile saving in labour costs. Breakdown replacement is even more expensive than planned re-lamping because of the one-off nature of the work.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[grigson]

I accept that your magnetic ballasts are going to last a long time..................

.....I am not really too convinced though, even in your case.

The fact is, a LED tube which runs off a LED driver, will be the longest lasting of all....and ultimately need least re-lamping......Supposing the LED ballast is inside the tube, then it will run cooler than the interface circuitry that gets used in EverLED-VE type products.....cooler running means longer lasting.

So taking a long term viewpoint....LED tubes with LED Drivers is the cheapest and best way......It ultimately means less re-lamping than for the "Fluorescent Ballast - LED tube" option.

I accept your point if you are saying that your lamps are difficult to get to....but then again...in the long term...that just means its best for you to use "LED Tube - LED Driver"

The "Fluorescent Ballast - LED tube" option *appears* to be a short-termist option.....

...but then again....."Fluorescent Ballast - Fluorescent tube" is the ultimate short term option because fluorescent tubes are much cheaper.

So i am afraid i can see only one reason for the EverLED-VE LED tube light...........

-the sales people dont know how to properly explain the options to the customer, and the customer doesnt know what they're doing.......and its the company's money theyre spending, (not their own) so what the heck, just go out and buy EverLED-VE.


This doesnt denegrate the quality of the EverLED product as a product in its own right.....i am sure its a very good product at doing what it does (running leds off fluorescent ballasts)...
I am sure it has great market potential for individual homeowners who want to experiment.

 

[grigson]

Hello,

I am still trying to discover if this product, the EverLED-VE LED light tube has passed UL or CE approval, or in fact, has it passed any approvals tests such as Safety and EMC related?

http://www.everled.com/led-lighting-products/everled-ve/

(as mentioned in previous threads, its a LED tube which runs off literally any kind of fluorescent ballast)

I have written to EverLED about this but get no reply.
There is no evidence of any approvals on the website.

 

[MacGyverS2000]

What does it matter if it has UL/CE certification? What are you trying to determine?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[LionelHutz]

In my experience, there is no approval when the company makes no mention of any kind of approval. Companies generally want everyone to know when they spent the cash and got a listing for a product.

There likely is no suitable UL standard to approve the product under. And yes, I know there are UL standards for lighting but there is likely nothing in any UL standard about a LED lamp running off a fluorescent ballast meaning UL can refuse to list it.

 

[VEBill]

Companies will also proudly list all the approvals...

Even if they're still pending.
Even if they've simply "planned".
Even if they're infested with deviations.

 

[grigson]

OK Thanks, I am trying to determine if its actually likely to be possible to get such a product through any kind of EMC or Safety test.......and thus trying to work out if it is worthwhile for a company i worked for to actually consider designing one of these type of products.