Initial current surge with "ramp up" dimmer modules

[Willy]

How about an MR16 with a GU10 base? Those are available in 20, 35, and 50 watt varietys.

(remember I missed part of the original thread, so perhaps you said you couldn't use those).

Willy

[LightingMan]

I didn't know you could get a 12 volt MR16 with a GU10 base. Is this possible? That would be the perfect answer.

Any help or input gratefully accepted - Thanks - LightingMan

[TKM]

Hmmm. Check that carefully. GE shows MR16s (20,35 or 50 watt) with the GU10 base, but those lamps are rated for 120 volts.

The 12 volt MR16s have what GE calls a "Turn-and-Lock" base. That base is really a GU7.

I thought the GU10 base was only used for mains voltage lamps.

Terry McGowan

[John Rumm]

The inrush spike you see on a filament lamp will be mostly over by the time the filament is close to glowing - and that is almost your starting point. Remember that with a dimmed filament lamp the reduction in light output is not equally matched by a reduction in current flow of equal proportion.

You are not going to see the instantaneous inrush on any form for meter easily though. You would really need to look at the voltage drop across a known resistance in series with your dimmer supply using a digital storage scope. (alternatively some clamp meters have a "scope out" facility that makes them quite useful for capturing this sort of reading without needing physical contact with the circuit).

This is a reasonable assumption, although note that the MCB may have amplifiers and other circuitry in there to trigger a magnetic trip once a threshold is reached.

Based on experiments I have done with dimmers etc, I have tended to observe the same - I doubt you will have made matters worse in reality.

[meow2222]

Hey - I think you've just hit on the answer. Using 120v bulbs in series pairs would make a big difference to trip on blow issue. The downside is you lose 2 bulbs when one blows, but thats a huge improvement on losing the whole lot!

NT

[meow2222]

Yes I'm sorry if I came across wrong, I guess communication is the solution.

You can test it, then you'll really know. Just apply the prods to a steady voltage, eg mains or a battery, and watch the pointer's response to a step change.

Its at least an order of magnitude too slow to spot switch on current surges.

what ratio are you seeing of cold R to calculated hot R (based on lamp ratings)?

indeed. Avo's too slow, and your experiment IIRC did not start from zero volts. If you've got access to a variac you could try and see - or just measure cold R and calculate the resulting initial current.

both are magnetic mechanisms, but they're significantly different mechas. If you look at trip curves for B and C type MCBs you'll see theyre orders of magnitude faster than any moving coil multimeter.

Well, with or without you've still got the same trip - blow problem.

OK, lets clarify what soft start means. The issue soft start addresses with filament lamps, in particular mains halogens, is temperature overshoot of small portions of the filament during startup. This reduces life of said lamps, and is caused by the low initial on resistance, plus uneven wear (evaporation) on the filament. So, soft start means avoiding that temp overshoot.

Now, there is no need to limit current to do this, which is just as well since triacs are inherently unable to control current flow. Instead temp overshoot is prevented by chopping the waveform, ie reducing waveform duty cycle. Key point here: this does not control the current, it only controls the temperature. The switch on current spike still happens.

this will reduce energy efficiency, and mean you need more total wattage for your desired light level, which doesnt help.

oh, that sounds worse than I expected.

I reckon the 120v lamps mentioned on another post could fix your problem. You may find this kills more lamps, as a 120v lamp isnt going to be amused by being used as a surge limiter on 240v, but it should mean a dead lamp instead of a dark corridor, and 120v

35w lamps are going to perform rather better than 240v in terms of efficacy/life too.

NT

[TKM]

Using two 120 volt lamps in series on a 240 volt circuit is common enough and it used to be routine when we tried to sell North American lamps in Europe and other 220-240 volt areas; but I don't see why both lamps would fail if one blows unless the first one failed via a short or an arc. In either case if the fault was inside the sealed part of the lamp, the fuse in the supply lead might open before the second filament.

Terry McGowan

[Don Klipstein]

I see enough 120V incandescents producing an arc when they fail, especially when they fail during starting. With 240V across a filament break, I would think that the tendency to produce an arc would increase.

If there is a second filament in series with the arcing lamp, I think that the second filament has some chance of protecting the fusible lead in the first lamp by blowing first. :)

- Don Klipstein ( [e-mail])

[Adam Aglionby]

Using PAR 64 4559 aero landing lights as effect lights it`s a known problem, `64 8" lamp but 600W at 28V with near paralell beam, common to use 9 in series in Europe, but if one blows whole string usually goes with it, acompany with rendition of `I`m forever blowing bubbles.`

Adam

[LightingMan]

Dear all - LightingMan back

What a fantastic idea about two lamps in series! It seems like the "blowing" cycle is possibly more of the problem than the "start" cycle. Probably extra bulb life with 120 volts?

It also means if that people in the block ever try to steal the 120 volt "communal bulbs" to use them in their flats it will blow with a great flash and trip their MCB's!! (They wont do that again!!) Even in a prime central London block (full of high earning professional types) people have stolen the existing "low energy" bulbs from the communal areas! What a great "side benefit". I'm also glad Meow-2 and I are friends! .......

Finally I'm not worried about a bit of loss in the dimmer as the lights are only on minutes a day. Let me give you an example. On the top "penthouse" floor there are four flats. Only two are occupied. I am told that people "go to work" at 6am and return late. Basically if there were PIR's like in my test corridor the lights would switch on 4

-6 times a day for 2-3 minutes. At the moment the "low energy" lights are on 24/7 waiting for that "city worker" to walk by!!!

Apart from a "work of art" , that's why I love my solution!!!!..........I hate low energy lights (well CFL's) as they are so white!!.....Now where do I get 120volt GU10's in the UK.......Newey & Eyre??

As usual any suggestions!!! Thanks - LightingMan

[Willy]

Yes, but not 12 volt.

I believe that GU10 is reserved for 120V applications. The GU10 is the more recent base of choice for products seeking an energy efficient or energy star rating. A decorative fixture can have a GU10 base, and lamp manufacturer's offer this lamp in a compact fluorescent ONLY, thereby making it difficult if not impossible for the consumer to change the lamp from an energy efficient product to a standard incandescent.

Not to beat a dead horse, but I'm wildly curious. What is your application where you would need these?

Willy

[Andrew Gabriel]

GU10 (and GZ10) are 10mm mains lamp bases.

I suspect you are thinking of GU24 (24mm diameter) bases. That is US only AFAIK. There has been a proposal in EU for a modified GU10/GZ10 base for the same purpose from one manufacturer, but I can't see it catching on.

[TKM]

It's the GU24 for which Energy Star and others are seeking the "energy efficient" rating. The industry needs a mains-voltage socket which will only accept high efficacy lamps to avoid testing and other problems in those areas which mandate high-efficiacy lighting. Having the socket rating encompass only high-efficacy products means the lamp products in those sockets can be designated high efficacy "by design". There are proposals being worked upon by UL as well as state and federal legislation which will limit the GU24 to (at the moment) CFLs or LEDs.

Terry McGowan

[TKM]

Adam, your example brought back some memories. There is a 6 volt 100 or 120 watt PAR64 lamp called a "transmisometer lamp" -- very narrow beam which generates 1 million + candelas. The original purpose of the lamp was to measure cloud heights at airports. For the lighting of a fountain jet once, I used 20 of them on a rack in series on 120 volts. The lighting effect was astonishing. Lamps did fail; but I can't recall that more than one failed at a time. Finding the burn-outs was easy. You could look into the clear face of the lamp and see the dangling filament in the reflector. A neon tester could be used for an electrical check too since the lamp has bare screw terminals. I can't say the installation was electrically safe; but it was temporary and worked just fine every evening over a summer season.

I agree with Don that either the filament or the fuse will likely blow in the event of a filament arc in the other lamp; but filaments are strong. I've put plenty of 120 volts filaments on 220-240 volts by mistake, turned them off quickly when things got brighter than expected and then had the lamp continue to work for some time.

Terry McGowan

[meow2222]

Either way, series lamps may reduce blow on trip events. May... it might not.

Yes, 240v 35w halogen is a very compromised filament since the filament is so thin. A 120v 35w is a quarter the resistance, so you get a better lamp life / efficacy relationship. IOW you'll see either more life or better efficacy, or a bit of both.

True... but only time will tell whether the 120v ones blow in pairs or not. If they do, you may lose more that way than by theft. It might be worth trialling a floor with 120s to see whether you get all the benefits or not. Or even wire up the fittings in such a way that its easy to change them between 120 and 240.

NT

[Willy]

Darn it, you are right... I am referring to the GU24.

Thanks for correcting me - last thing I want to do is cause confusion. GOOD GRIEF!!!

Willy

[Willy]

Dear LORD I can't believe I made that mistake. I just don't know what I was thinking when I made that statement. And worse yet I was looking at an image of a GU10 MR16 when I typed it, so I should have known better.

Yes, of course you're 100% correct, and everything YOU said was exactly what I was trying to say.

Willy