12.6v+ vs 6.3v+/- heater supply

[mountainhick]

I can go with either 12.6 or 6.3 +/- with center tap.

The first pic below shows a circuit with 7812 regulator and a diode on the regulator ground to bump the voltage, but my understanding is the 7812 regulator needs a few extra volts on the supply side, hence the specified 15v transformer. Is this correct? (I already have a 12.6v transformer and would prefer not having to buy a different one).

The second pic shows a 12v +/- center tapped PS. I am wondering about adapting this to 6.3v by using my 12.6v center tapped transformer and 7806 and 7906 regulators. Certainly the 12.6v transformer will work. But, how can I bump the voltage from 6 to 6.3v? If diodes are added to bump the output of the regulators, won't the voltage be a bit too high? Or, can 12ax7 type tubes run well from the slightly lower 6v anyway?

 

[Tom Bavis]

78 series regulators need about 2.5V headroom to regulate. There are "low dropout" regulators that only need 0.6V or so - probably what you need. In any case, use Schottky rectifiers for their low forward drop (1N5822) to get the most from your transformer. 12AX7s CAN run well below rated heater voltage, but transconductance drops. OK if you don't have to deal with a lot of voltage swing as in a phono preamp.

 

[Svein_B]

6V for heater is within the specified +/- 5% and should be quite OK.

If you need to bump 0.3V you could try a shotkey diode. They have a low voltage drop of about 0.2-4V.

7812 is specified to require 14.6V input + 0.6V for your bumper diode = 15V. So you should be OK with the 12.6V PT.

Svein.

 

[mountainhick]

7812 is specified to require 14.6V input + 0.6V for your bumper diode = 15V. So you should be OK with the 12.6V PT.

Svein.

I don't understand this. How is the diode used to bump 12.6v to 15v before the regulator?

 

[Zen Mod]

use LM317 as CCS for heaters

elevate (connect) heater supply to some positive voltage ( say 40V above cathode potential) , made with resistive divider from high voltage supply to gnd

give us more information about exact project , and I 'll tell you exactly how to make that

 

[Zen Mod]

I am adding a separate PS to power one 12ax7 tube heater. I have a 12.6v transformer to use. It is that simple.

nope ;

difference between "it is that simple " and "nope " is what differentiate mediocre and good stage ; heating is important , not just helper circuit .

you see some tricks here : http://choky.diyaudio.rs/WOT_preamp_and_battery_charger.html

 

[mountainhick]

nope ;

Yep, 12.6v+ or 6.3v +/- regulated heater supply for a single tube is all I need. I do not need to complicate this issue.

 

[Zen Mod]

whatever ;

I was willing to draw exact schematic , for your exact project ;

 

[woody]

What is the voltage and current rating of your transformer?

 

[mountainhick]

whatever ;

I was willing to draw exact schematic , for your exact project ;

Please do. Really, the exact project is to add a transformer and regulation to supply one 12ax7 tube with regulated heater voltage. That's it.

 

[mountainhick]

What is the voltage and current rating of your transformer?

12.6-0-12.6
0.45 amp

 

[tomchr]

Note that the 12.6 V on the transformer is RMS. The rectified DC will be about 1.3x higher. So your 12.6 V turns into 16.4 V DC before the regulator. With 2.5 V head room across the regulator, that leaves (16.4-12.6-2.5) = 1.3 V for ripple. That's a reasonable design but does require you to use a fairly sizable reservoir cap. You can calculate the cap as:

C = i*t/V

where C is the capacitance in Farad, i is the DC current drawn from the supply, t is the discharge time (set this to 1/(2*f) where f is the mains frequency - 60 Hz in US, Canada; 50 Hz most other places), and V is the ripple voltage. The factor of 2 in the t = 1/(2*f) arises because I assume you are using double-rectification, i.e. using a diode bridge or a diode to each end point of the 12-0-12 winding.

I haven't looked at your schematic, but I assume you're inserting a diode in series with the ground lead of the 7812 to get to 12.6 V. This works by using the forward voltage drop of the diode (0.6~0.7 V for a silicon junction diode) to raise the ground voltage of the regulator. The regulator will maintain 12 V between GND and OUT, hence, the resulting voltage will be 12.6~12.7 V. But note that this output voltage will vary with temperature. It's probably not a big deal in your circuit, but the output voltage will have a tempco of about -2 mV/deg C.

A better solution would be to use an adjustable regulator such as the LM317. By setting two resistors, you can program any output voltage in the range of 1.2~37 V.

For any regulator, do use the bypass caps on input and output. On many of the legacy products you might be able to scrape by without (but why bother) but on the newer products, they are required for stability.

Hope this helps...

~Tom

 

[mountainhick]

Thanks Tom!

So please check me, is this correct?:

-12ax7 has 150ma current draw at 12.6v,
-60 hz
-1.3v ripple

(.15/120)/1.3= 0.00096 farad or 960uF


I have a 4700uF 25v cap that I don't need for anything else. Any reason not to use this higher value?

Is it necessary to heat sink the regulator?

I already have the 7812 regulator so I'll give that a try first. If it ends up over 12.6v under load with the diode, I'll switch to the LM317

 

[Eli Duttman]

Thanks Tom!

So please check me, is this correct?:

-12ax7 has 150ma current draw at 12.6v,
-60 hz
-1.3v ripple

(.15/120)/1.3= 0.00096 farad or 960uF


I have a 4700uF 25v cap that I don't need for anything else. Any reason not to use this higher value?

Is it necessary to heat sink the regulator?

I already have the 7812 regulator so I'll give that a try first. If it ends up over 12.6v under load with the diode, I'll switch to the LM317

12.6-0-12.6 FWCT rectified by 2X Schottky diodes will be fine, when combined with 7812 regulation. Forward drop in Schottkys is roughly 0.5 V. Subtracting that 0.5 from 12.6 and then multiplying by 2^1/2 yields 17.1, which is quite sufficient in the headroom dept. for a 7812.

As long as ventilation is decent, you probably don't need a heatsink for the regulator IC, with only a 150 mA. draw.

12 V., on the dot, is within 5% of the nominal 12.6 required by the 'X7 heater. Stop fretting! If anything, the tube will last a bit longer. 😉

A 25 WVDC/1000 μF. 'lytic is excellent as the filter. However, full wave rectification implies a 120 Hz. ripple freq., which works to your advantage. The 25 WVDC/4700 μF. part you have on hand is overkill, but go ahead and use it.

Improve 7812 action by installing a 1 μF. Panasonic ECQ-V stacked film part (DigiKey stock # P4537TB-ND) right at the IC's I/P terminals. Further improve 7812 action by installing a 15 WVDC/10 μF. 'lytic right at the IC's O/P terminals. Finally, install a parallel combination of a 15 WVDC/10 μF. 'lytic and a 10 nF. ceramic part directly across pins 4 and 5 of the 12AX7's socket.

BTW, if you don't already have suitable diodes on hand, the TO220 case MBR20100CT common cathode twin Schottky is "perfect" for the job.

 

[Minion]

12.6v AC will give you about 17v DC so there is plenty of headroom for the regulator ......

 

[Poindexter]

You guyz are flogging yourselves. For a 12AX7, just use AC heaters. Pad down to 6.2V or so, and in front of the pad, put a 2.2 - 5.6μF NP electrolytic cap to keep all the hi-frequency nasties out of the heater line. Observe the Vh-k rating. You are going to get no heater leakthrough. Give yourselves a break, spend your energies on something real.

Aloha,

Poinz
AudioTropic

 

[mountainhick]

Thanks everyone, I think I have plenty of info to make this work.