G9 bright tube glow on power up

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Smoke

Well-known member
Joined
Nov 16, 2004
Messages
63
Location
Melbourne
Hey guys,

im just at the completion of my G9, it seems to work fine on first powerups.

I have noticed however, that every now and then a RANDOM tube glows really bright for a second or so when I power up the unit. Its always 1 tube and always a random one.it flashes/sparks and calms down to normal. The preamp seems to work fine.

My thoughts are that this would probably wear out my tubes really fast, I could only think they wont like it.

Whats causing it and how should I stop it if its bad. Bad capacitors? what information do you need to know.

Thanks
 
If you've got vintage Amperex Bugle Boys in there, the filaments kind of "flare" when they first get hit with voltage. That's the only tube I've ever seen do this, but there may be others that do the same thing.
 
tube socket??
hows that influencing the thing?
i have used tube sockets.

umm no they are not amperex, I dont even know what they are. lol
 
I have something like that on one of my G9's, with unbranded tubes. I has assumed that the regulator takes a second to get working whilst the caps charge up, and the tubes see over-voltage for a moment. All 4 of the tubes glow bright on power up - but nothing I would call a 'spark'.

I wasn't going to worry about it!

I don't recall seeing it on the 1st G9 I built, which has Mullards in (and different brands & values of electro caps probably - I tend to use whatever I have near by so long as it is similar or bigger). However, that's all boxed up at the studio so I rarely look inside on power-up.
 
[quote author="Smoke"]Whats causing it ...[/quote]

This is normal because the cold resistance of a filament is quite a bit lower then its hot value - resulting in a high initial filament current - yielding extra photons from the filament in the process. It's not dangerous ... :wink:
 
Same thing happens to the 6AL5 in one of my Audimaxes, it means its the tube with the lowest heater resistance and its not a problem, or so Im told.

Dont know why it would change every time, but you might want to check yr tube sockets and see you have good contact on all pins and give them a clean.


M
 
I has assumed that the regulator takes a second to get working whilst the caps charge up, and the tubes see over-voltage for a moment. All 4 of the tubes glow bright on power up - but nothing I would call a 'spark'.

I don`t think this has anything to do with the regulator since I`ve had this happen on several things I`ve built with AC heaters.

I`m with Analogique on this one. The cold filament has a lower resistance than a hot one . You can test this by measuring the resistance of the cold filament with a meter & then calculating it with ohms law from the current flow & voltage when its hot.
 
i have some of the same problem when i was testing some tubes (ecc82)
it flashed and then glows normally. the tubes are from an old used broken device. (transformator was broken) they flashed only once. and ten every time i set them at glow they do normally.
 
obviously im not the only one, and i guess it doesn't sound dangerous to everyone.

i guess tube sockets may add to resistance difference on the heater.

I also observe the first time only glow thing, if i turn it off and on none of them would glow. however the next day one would, which proves that it has something to do with the heat.

I always thought it was the voltage regulator not regulating untill current is drawn - less than a second, causing the glow.

I still think that my tubes wont be happy if it goes on. oh well. Thanks to everyone for the replies.
 
[quote author="Rob Flinn"]
I don`t think this has anything to do with the regulator since I`ve had this happen on several things I`ve built with AC heaters.

I`m with Analogique on this one. The cold filament has a lower resistance than a hot one . You can test this by measuring the resistance of the cold filament with a meter & then calculating it with ohms law from the current flow & voltage when its hot.[/quote]

Makes sense - I was thinking about it wrong! Thanks.
 
IC voltage regulators are fast - there will not be any overshoot if you have the recommended output capacitance on them. If you omit that capacitor they will oscillate, but they never should 'overshoot', and certainly not for a second or so. I've seen lots of tubes do this, the effect of the hot spot in the filament would be to control the inrush current of the heaters, this is especially important on series heater strings (in television sets and radios) where they are connected directly across the 120 volt line. If there are any imbalances during the warmup phase, some tubes will burn out early.
 
Back
Top