Heater power supply help

[BillyF]

I need some help from you guys who are more familiar into DIY stuffs.
My pre-amp PS has gone wild. The pre-amp tubes 2x6922 were very bright and when I did the measurement, heater supply went up to 12V. I did some measurement on components and found this regulator GL317 shorted. I replaced with LM317 and tuned the desired voltage 6.3V for the filament but, it can only run for less than a minute then it shut-off. The regulator turns very hot when loaded with the tubes.

My x'former secondary output 10.3Vac to the bridge rectifier 100V/6A and measured 12.96Vdc after filtering section and serves to be my input to LM317.

Any idea how can I fix this???

Thanks a lot for your halp.

 

[TG]

Throw away the regulator, connect tube filaments in series and heat 'em with your 12.96 V...

6922's filament current is 0.3 A.
You have now (12.96-6.3)*0.3*2 = 4 W dissipation on LM317.
If there's no heatsink, LM will die very soon...

 

[Burnedfingers]

I don't know the current limits of the LM317 without looking it up. It most certainly is figured with a proper heatsink. Could it be the GL317 was a heftier part? If so cross reference the GL317 to a part of equal quality.

Joe

 

[JOE DIRT®]

The Lm 317 is rated for 1.5 amps with a modest heatsink

I would measure what each heater is drawing there could be a chance you have a bad tube

also is there a current limiting resistor going to each heater>???...if so that could be the culprit


DIRT®

 

[Brett]

With 6.3V across the filaments and about 2.5V across the 317, the simple solution is to make a pi filter by adding a series resistor in the PS and a second filter cap to bring the voltage at the input to the 317 to about 9V. Calc's out to 6.6R and about 10W.

I don't like series stringing tubes, esp when the voltage is a bit high. You can never tell exactly how they will line up in terms of voltage across each individual filament, so one may end up > 6.7V and shorten it's life.

Heatsink the 317 really well.

 

[BillyF]

The LM317 has a trim pot connected to the common pin to (B-). This where I tune the ouptut voltage to 6.3V and a 150ohms resistor connected between commom to ouput pins.
It will only runs hot if the load is present. I did tried to load only 1 tube since my pre has 2 6922 tubes, but still hot. The regulator shuts-off after reaching a certain temp.
how????

 

[jan.didden]

The Lm 317 is rated for 1.5 amps with a modest heatsink

I would measure what each heater is drawing there could be a chance you have a bad tube

also is there a current limiting resistor going to each heater>???...if so that could be the culprit


DIRT®

Exactly. He had a system that worked, no it doesn't anymore, blows the reg. Logically, that points to a (partially) shorted heater wire (or series R, if there is one). So, advise on a complete redesign is not what he needs, because it wouldn't work anyway without getting rid of the cause.

jan Didden

 

[fdegrove]

Hi,

I don't like series stringing tubes, esp when the voltage is a bit high.

Me neither...unless we're talking a single block using //ed valves you still have a real risk of signal breakthrough between cathode and heater.

Anyone ever capcoupled a heater supply and listened to it?

Powersupplies?

Cheers,

 

[BillyF]

Thanks guys for all your suggestions. But, I am a bit lost where to start. Roughly, I may need ~8Vdc without the load. It will be just nice maybe for 6.3V.
What is the best step now? Bring down the input voltage to ~9Vdc?
Any other design I can DIY with my existing 10.33Vac for the filament?
thanks again...
Billy