Flash seen from bottom of 12AT7 PI socket

[TUBEDUDE]

A bit extreme possibly. But you might get a few more hours from them before the heater boils away.

[maxkracht]

An ntc thermistor would probably do the trick, but it won't have the satisfying relay clicks and a cool box.

[pjd3]

Wow, that some ambitious current limiting. I don't even have a lightbulb fixture. Thank you for posting those pictures.

Makes me think I've been lucky or watched over by electron angels over the course of my three tube amp builds. Of course, I may make up for that by way of over zealous ohming out of especially power paths and any manner of potential shorts. But, that doesn't let me look inside a bad tube.

I first took that tube out and the amp's B+ looked fine/usual 467vdc. I put the tube back in, powered the amp up.. It did flash, but also I saw the same B+ of 467 volts as before. make me thing I was getting some kind of visual skew last time I briefly powered it up.

So I guess I will let that flashing 12AT7 live out its natural life and see about getting on with things. The next step will likely be to install a B+ MOSFET reducer circuit I made with a 20 volt dropping zener in it. Just feel a little better for those old and expensive 6L6WGB's and having around 445 on the plates of those may give some longer life. Plus, with the current state of local venues insisting on reasonable sound level intensities, I will never reach power tube break up. Never. Also thinking of putting a bypass switch in there for standard 6L6GC's, which could very well find their way in their one day. Hell, its only a hole, a switch a minimal wiring reroute.

Thank you everyone for educating me on something I may have never known about. I would have never figured out the "flashing filament" syndrome, alone.
Best,
Phil D.

[pjd3]

Just thought of something ridiculous,

For those who are intent on filling their preamp tube sockets with their favorite "flashing filament" NOS tubes, is there anything stopping someone from implementing an RC circuit from one (or both) of the 6.3vac filament wires to ground as a form of "current trickle". Set the time constant to say, 5 seconds or so, then flip a bypass switch after 5 seconds? That way the current would build slowly to avoid the quick flash and perhaps give longer tube life, should the flash in fact be a symptom of life shortening filament stress.

I really don't know it this works with AC the same way with DC but, in my 4th coming amp build I placed around 50K ohms across the standby switch to create an RC for the rectifier section. It is a diode rectifier circuit and now, it takes about 1 sec to reach 35vdc, 2 seconds to reach 75vdc, on and on, to avoid the certain terrible inrush current. So I lost the purity of a true standby switch, but, I gained sleep knowing that the horrible onslaught of current was now eliminated upon turn on.

Can a similar thing be some with the 6.3vac filament circuit?

Just thinking out loud.
Best,
Phil D

[sluckey]

Can a similar thing be some with the 6.3vac filament circuit?

Use a thermistor in series with the primary winding to prevent high inrush current at turn-on.

[pjd3]

Thanks Sluckey,

And I really should have thought of this, considering I fairly recently spec'd out one of these for my bands Presonus digital mixer. I recall it being around 50 ohms cold, and then decreasing to around an ohm once it heated up from the steady state current. Being just a tech where I work, I have the nice advantage of being surrounded by good engineers that generously come to my rescue from time to time and help spec components like this.

But just to mention, if anyone should ever have an older Presonus digital mixer that starts popping and losing its output, replace the four under-spec'd 5 volt voltage regulators on the smaller signal boards with 78xx types!
Saved about 500 bucks for spending 5 bucks!

I see that this thermistor uses the same model # nomenclature as the ones I bought. Should check the extra ones have.

Thanks again,
Best,
Phil D

[Reeltarded]

Because of the flash I power up my amps through a current limiter on the primary side, between the wall socket and the amplifier power cord.
That also helps the input fuse and the input rectifiers and powers up everything including the anode voltages in a smooth manner…

I don‘t know if all values of the components are the end-status…
IMG_0788.jpegIMG_0784.jpegIMG_0783.jpeg
the power resistors are still missing in the 2nd picture…

Pro tip. Pro tool.

[martin manning]

...I witnessed a whiteish flash through the base of the Phase inverter socket that had a NOS 12AT7 in it. I have not seen this happen before and it alarmed me.

Heater flash from video linked by Mark in Knight emissions tester post: https://youtu.be/SMHS1DNS8FA?t=169