I have a Sound Valves monoblock tube amp that uses 12V DC voltage to power the filaments of the two input tubes - ECC801S and 12AU7. The voltage is generated from a LM7812 voltage regulator. When the amp is powered on the filaments on the tubes get brightly lit for a split second and then dim down. I was a bit concerned that the current surge would shorten the life of the tubes (especially the very expensive ECC801S), and I was wondering if I should do something to slow down the initial surge.
The LM7812 has a 1000uf capacitor on its input and a .1uf capacitor on its output. I thought of adding another 1000uf capacitor to the output but wasn't sure if this would be the best way to slow the surge. I'd appreciate any thoughts on simple ways to handle this problem. Thanks.
The LM7812 has a 1000uf capacitor on its input and a .1uf capacitor on its output. I thought of adding another 1000uf capacitor to the output but wasn't sure if this would be the best way to slow the surge. I'd appreciate any thoughts on simple ways to handle this problem. Thanks.
That's normal, it has to do with the bit of uncoated heater wire outside of the cathode. Don't worry about it...John
In my amp I use a 1R 100W resistor in series with the 6,3VAC filament supply for 30 seconds, then it is bypassed with a relay. No more bright glowing filaments as some of my old tubes would glow much more than the others, I feared that eventually it would burn.
Now they all glow dull red before full voltage is applied and go up to bright, looks much more gentle on them.
Now they all glow dull red before full voltage is applied and go up to bright, looks much more gentle on them.
Putting a big cap on the regulator output will not help. Probably no help is needed. Some valves do this heater flash on start-up. Running heaters directly from a stabilised supply will make it slightly worse. Normal AC heaters can get the benefit of the transformer resistance to limit the initial surge.
If it worries you, increase the voltage slightly by adding a couple of diodes in the 7812 ground connection. Then add a WW resistor to the 7812 output to drop the voltage down again - this will limit the surge. Don't be tempted to run heaters too cool with anode current flowing as this can damage the cathode. It looks gentle, but can be harmful.
If it worries you, increase the voltage slightly by adding a couple of diodes in the 7812 ground connection. Then add a WW resistor to the 7812 output to drop the voltage down again - this will limit the surge. Don't be tempted to run heaters too cool with anode current flowing as this can damage the cathode. It looks gentle, but can be harmful.
B+ voltages are delayed too, so there is first anode current flowing when heaters are prewarmed.
Incandescent heaters of most receiving type indirectly heated cathodes are mostly tungsten and exhibit the typical 1:10 resistance ratio from cold to hot. Limiting the current of the supply to the running value will absolutely prevent the tubes from doing more than a graceful warm up, from no incandescence thru gentle glow to normal full brilliance, over a few seconds. The stiffness of a regulated DC supply, as noted elsewhere in this thread, will make the "surge" brilliance more pronounced. Perhaps the simple expedient, rather than an active current limiter, of a resistor that will drop the heater voltage by perhaps 5-8% after full warm up will do it, as it will cut the inrush current by close to .5 the value without the resistor. Might be enough.
Rene
Rene
You can also use a separate transformer, only dedicated to filaments, which is just a little bit more powerful than needed. It's own internal resistance will protect your tubes from being overloaded at start-up, while avoiding extra circuits to regulate inrush current.
The LM7812 has a current limiter which limits at a bit over 1 A. That's far easier on the tubes than a filament transformer would be. With a transformer, the in-rush current is only limited by the internal resistance of the transformer and the cold resistance of the tube heaters. So I wouldn't worry about it.
As others have pointed out, the bright light could also be caused by a bit of the heater wire being exposed rather than covered by the heater/cathode isolation material.
If you're truly worried about it, you could implement a CCS using an LM317. Of course, this requires that you have the necessary headroom on the heater supply.
Finally, you could use a voltage regulator with a current limiter set for a lower current. The current limiter on my Universal Filament Regulator (see my website) is programmable by changing a resistor.
~Tom
As others have pointed out, the bright light could also be caused by a bit of the heater wire being exposed rather than covered by the heater/cathode isolation material.
If you're truly worried about it, you could implement a CCS using an LM317. Of course, this requires that you have the necessary headroom on the heater supply.
Finally, you could use a voltage regulator with a current limiter set for a lower current. The current limiter on my Universal Filament Regulator (see my website) is programmable by changing a resistor.
~Tom
My current amp project needs 7 amps of heater current so I stuck to AC feed to keep it simple and not use huge heatsinks on a regulator.
The transformer will supply 16 amps but the heaters take a while to warm up due to the transformer limiting the current.
So it takes about 30 seconds to get audio up and running but I don't mind.
At least it is gentler on the valves.
The transformer will supply 16 amps but the heaters take a while to warm up due to the transformer limiting the current.
So it takes about 30 seconds to get audio up and running but I don't mind.
At least it is gentler on the valves.
- Status
- Not open for further replies.