CALCULATING a choke value

[Colas LeGrippa]

Hi, I plan building an external tube spring reverb unit and the schem doesn't show the choke value. I'd put a 2 to 5H in there but how can I figure out it's ''right'' value? ( though the margin is wide, I think ). Here is the schematic.

Regards

Colas

P.S.: kiss your wife NOW before I do it.

[jim]

6K6 screen doesn't need much and the margin is real wide but choices are limited.  Use the "small" Fender choke like for a DR.    Maybe I'm lazy but I only know two sizes-- Fender small and Fender big.

[HotBluePlates]

Well, the power supply is not choke-input, so figuring these things will be easier.

How much current will the choke handle? Class A 6K6 plus 4 triodes. The 6K6 is lower plate dissipation than a 6V6, but you might only have 6V6's handy, so lets figure based on the 14w rating of later 6V6's. Round up B+ number to 300v.

14w/300v = ~47mA

Add 4-5mA for the small tubes (which is a conservative estimate, guessing high), and the choke will need to support ~52mA.

The "small" Fender choke is a 4H, 50mA unit. The "large" choke is a 4H, 90mA unit. Since this isn't critical, and we rounded up in a few places, the small 50mA unit is sufficient.

Given the half wave rectifier, let's say the ripple at the first cap is 12vac. The series reactance of the choke works against the shunt reactance of the cap to form a voltage divider to this a.c. ripple. Because this is a half-wave rectifier, the ripple frequency is 60Hz. The capvaule listed on the schematic is 40uF.

X_L = 2πfL = 2*3.14*4*60 = 1507 ohms (inductive reactance)
X_c = 1/(2πfC) = 1/(2*3.14*60*0.000040) = 66 ohms

The d.c. drops slightly due to the choke's d.c. resistance, but the ripple sees a voltage divider made of 1507 ohms in series and 66 ohms to ground. Ripple at the second cap drops to:

12vac * [66/(1507+66)] = 0.5vac

You could get a choke rated for more Henries, to reduce the ripple further, but it's already quite low. It will be lower still if you use a full-wave rectifier because the ripple frequency will be 120 Hz; reactance of the choke will double while the reactance of the cap will halve, and ripple at the second cap will be 13mV ac.  

[tubeswell]

The "large" Fender choke is a 4H, 50mA unit.

That would be the small choke of course :-)

[Colas LeGrippa]

of course.

[HotBluePlates]

Thanks. Thought one thing and typed the other. The post is edited above to correct that.

[tubeswell]

Happens to me all the time

[PRR]

> figure based on the 14w rating of later 6V6's ... ~47mA

Note there is 24V across a 1,000 ohm cathode resistor. The power tube current is more like 24mA than 47mA.

[HotBluePlates]

Good point. I figured on a non-exact, but worst-case scenario to be sure the choke's current rating was adequate.

But it seems like we might be able to use a smaller, cheaper choke, assuming someone has a source for a ~30mA choke somewhere.

[DummyLoad]

http://www.unclespot.com/FenderXFMRchart.html

deluxe and deluxe reverb used same choke P/N 125C3A. use hoffman P/N 125C3A. 18 bux.  it's 4H @ 50mA.

[Colas LeGrippa]

Thank you all for tutorial, infos and part number !

Best regards

Colas

[Paul_Fawcett]

Just a little side note.  The original rectifier on these units was indeed a half-wave jobby.  Fender was evidently trying to do stuff on the cheap, back when solid state germanium rectifier diodes were still expensive and newfangled.  But with modern and very inexpensive silicon diodes there is absolutely no good reason why a full wave rectifier shouldn't be used, and lots of good reasons why it should.  I didn't even consider using a half-wave setup when I built a 6G15-type unit a few years back, and you shouldn't either.

[Colas LeGrippa]

of course not. half wave is not even in my vocabulary.

[Paul_Fawcett]

For my project I used the transformer bundle from triode. It was reasonably priced and worked well: http://triodeelectronics.com/reverbbundle.html

The choke is 4H 90mADC 110R DCR
Note that while the PT says something about being rated for 27mA, this is actually just the approximate total power draw for the whole unit...about 24mA for the power tube as PRR mentioned, and a couple more for the triodes.