70W HPS bulb on 230V mains with No Ballast - Just trips the MCB :(

eveares

eveares

Joined
28 Jul 2012
Messages
1,351
Reaction score
55
Location
Surrey
Country
United Kingdom
So today I did a little experiment and thought I would see what would happen if one powered up a 70W HPS bulb with an internal ignitor directly on the 230v mains without using a ballast.

Sadly not much happened other than a brief flash for a fraction of a second before the MCB for the circuit tripped. No catastrophic failure or bright luminous prolonged overload. :((n)

Below is a picture of the bulb lit up taken from a frame of the video I filmed. Frame represents 1/50th second and was the only frame where the bulb was noticeably lit up.

[GALLERY=media, 97835]70w HPS Bulb Lit Up by eveares posted 3 Jun 2016 at 4:34 PM[/GALLERY]

At least one now knows what happens in such a scenario.

Also, I don't recommend one tries what I tried in case such a bulb explodes - I turned it on remotely using X10 as a safety precaution.

Regards: Elliott.
 
Hardly surprising. They need ballasts for a reason, to provide the correct striking voltage. Do it at 230 and you've seen what happens. They're not going to explode at 230. They'll strike for as long as the MCB holds in. You'd need to get to photonicinduction levels of silliness to blow them up

I knew someone who did the same thing (by mistake) they put a frosted MH lamp in place of a standard ES lamp. Took out a 16A cartridge fuse on the busbar feeding it
 
I guess the next question would be

"what happens if you don't limit the current with a MCB?"
 
Iggifer said:
Hardly surprising. They need ballasts for a reason, to provide the correct striking voltage. Do it at 230 and you've seen what happens. They're not going to explode at 230.
Click to expand...
would have thought the ballast limited the current, the ballast output voltage would also be around 230 volt to neutral.
It was quite common for lamps to explode and shower molten glass due to major failure of the filament from over current
 
You're clearly right. The ballast obviously limits the current, I've clearly confused my ballasts with my ignitors in that post. I'm pretty sure I even typed that post correctly wrt current not voltage, and then changed it for some reason
 
Well, actually, 333rocky333 & Iggifer ... you are both right, well, sort of!

Once the lamp has struck, the ballast acts as current limiting device to control the lamp current.

In combination with the ignitor (internal or external) the ballast provides a high voltage 'kick' to get the discharge started when the lamp is first powered on... much the same as it does with a fluorescent tube. When the lamp is cold a discharge happens across the internal ignitor (it contains a neon gas) this heats the bimetallic strip which moves and touches a contact, thus making the ignitor (and hence the lamp) look like a short circuit. A current now flows thru the ballast & builds up the magnetic field within the ballast. Since there is no longer a discharge within the ignitor, it cools down & the bi-metallic strip bends away from the contact - it opens. This interrupts the flow of current through the ballast - this sudden interruption causes an inductive high voltage 'kick' which starts the discharge in the lamp. The ballast now commences its secondary function of current limiting.

When you put the lamp direct on the mains, what I think you are seeing 'light' briefly is the internal ignitor. Once the ignitor heats up & closes, it presents a hard short to the mains which opens the MCB. If you tried the same experiment with an external ignitor lamp, I don't think anything will happen - it wont strike because the voltage isn't high enough.
 
The correct operation is to start the lamp with normal control gear then short the ballast when it's running. I did that with a 5' tube a few years ago and it popped, glass all over the bench...
 
Iggifer said:
They need ballasts for a reason, to provide the correct striking voltage.
Click to expand...

But is that not what the ignitor is for, which in this case is inside the bulb. Hence it is a SON-I bulb and not a SON-E.

securespark said:
Nice workshoppy.
Click to expand...

Thanks, it's nice when it is clean. Please excuse the fire alarm panel monstrosity, it's a old side hobby of mine. The single gang box next to operates at ELV and is to do with the panel, so no to worry.

Iggifer said:
Just a shame about the fatmax drivers :eek:
Click to expand...

They were bought along time ago, there not bad as non VDE screwdrivers for occasional use.

plugwash said:
"what happens if you don't limit the current with a MCB?"
Click to expand...

I wonder :D, It did trip both the 20A MCB in the shed and the 32A one down in the house. Both Type B of course.

Maybe my next port of call is to start it normally using the ballast and when it is fully warmed up, to then bypass the ballast in the circuit.

I do have a 80W MV lamp as well, what work rather nicely in a 70W HPS ballast.
 
My wife did the same experiment with a SON bulb, she routed in my garage when the E27 spot lamp failed and found the SON bulb, she screwed it in switched on and the fuse and breaker both went open circuit. Never asked her if it flashed, they were not my choose words!

I did however hide all the 110 bulbs better.
 
No wife is not an electrician, I had an assortment of bulbs in the garage which included some 110 volt E27 bulbs, I had thought all lights in house were BA22d but it seems she had bought some Ikea standard lamp which was the odd E27 which I was unaware was E27 at the time.

Not sure if discharge bulbs are marked with a voltage? Seem to remember as an apprentice measuring the voltage across a fluorescent tube, around 200 volt at start but dropped to around 60 volts at run, this would vary with length of tube, I would assume other discharge lamps are the same, never measured. It is the control gear which is matched to voltage not the bulb/tube.

So other than blowing the heaters, what happens if you put a fluorescent tube into a fitting converted to use LED tubes? Assuming no ballast. Clearly three ways it could be wired, A) line neutral just one end, will just blow heater but not recommended wiring as if LED tube fitted wrong way around it is a dead short. B) both ends is series with a fuse where starter is normally, I expect that fuse would blow. C) both ends is series without starter now not sure what will happen, will it strike as the filaments blow? and if it strikes what current will it take? And will the tube break?
 
Don't be surprised if the ignitor has welded it's contacts together.
 
You must log in or register to reply here.
Back
Top