Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Choke Me!

2/21/1999 5:48 AM
Choke Me!
Well, I know nothing about chokes....  
I can't find anything usefull written about chokes in O'Connors TuT or any other books I have.  
I have a 5e3 circuit. The PT I'm using is a direct replacement for a Blackface deluxe, which has more power voltage and current-wise that the tweed circuit needs.  
I measure 360V on the secondary before rectification, and that is too high. With tubes installed, I am getting 410V on the plates of the 6V6s (and by the way, they are idling at 49.2mA).  
Things are too high, as I might expect for using an overpowered PT. So, I need a choke.  
How do I find a choke with the correct specifications? Heck, what are the specifications? Do I just purchase any old choke?  
And now, a word from our sponsors:

2/21/1999 8:09 AM

Go find an old ARRL Handbook from the 50's or 60's, at a hamfest or used book store.  
There will be an excellent description of power supplies, chokes etc.  
If memory serves me, B+ to the plates on a 5E3 is tapped off BEFORE the choke (ie. the filter is a capacitive input type), so just adding or changing value of the choke following the standard schematic will not cut it.  
Chokes are measured by their inductance value, specified in Henries (H). Power supply filters come in two main flavors: Capacitive input and choke input. With cap input, you usually end up with a B+ about 120 to 140 percent higher than your AC input. With choke input, you usually end up with a B+ of around 90 to 95 percent of AC input. The type of filter is determined by that which directly follows the rectifier, choke or cap.  
There's a cool trick to dial in B+ I'll cover shortly, but you may be looking at a bit of a redesign of the PS to use that PT, or for a more simple solution, use the old Zener on the center tap trick to drop the B+ down into "6V6 happy" range. (although Leo Fender managed to run PP 6v6 circuits in the +400 vdc range, and you could too. According to O'Conner, it's screen current that kills the tubes, so upping the screen resistors to 1K 3 or 5 watters might be all you need with your existing B+, and NOS 6V6's of course)  
If you really want to switch to choke input, then there's a formula for critical inductance based on voltage and current. I can't remember it right now, but a 10H choke would do fine. You also need to consider that all current of the amp will be going thru the choke, not just the screens and preamp as in the 5E3. So you'll need a 10H 150ma choke to be safe. (Maybe less H is ok, but that goes back to the formula I can't remember right now, and it's too cold up here to run out to the shed :-)  
Redesigning the PS into choke input will result in a B+ around 330-340 with your current iron. That might be cool for a "brown sound". The trick I referred to above is to design a hybrid between choke and cap input. Putting a small cap before the choke will allow you to select your B+ in the range of %90 to 130% of VAC. Values between about .5uf and 2uf are what we're talking about, and they usually need to be experimentally determined.  
I just tailored a hifi amp this way, raising B+ from 330 to 390 by adding a .68 cap before the choke.  
And of course, there's a whole lot of other stuff going on too, like fluctuating current draws in the amp, mainly based on the power amp class (A, AB etc.), and choke input might alter the classic 5E3 sound too.  
The "trick", basically is a balancing act between choke and cap input. Bruce Collins deserves full credit for clueing me in to this one.  
Sounds pretty complicated, don't it? But do-able, especially if you have an ARRL book handy.  
Now after all that rambling, I'd probably abvise you to just drop the B+ with a zener on the center tap, and protect the screens with higher value resistors.  
Have fun...  
... Whit
2/21/1999 3:04 PM

Whit the 5E3 has no choke.  
I think they run at around 360vdc.  
But I've built class A 6V6 amps using 400vdc with good results.  
The little cap in front of the choke is OK to do as long as you have some DC resistance in the choke too.  
Here's another "old time" ham radio trick I mentioned once before but nobody took me up on it.  
Install the 3Hy to 7Hy choke (rated at full current) after the recitifier, but bypass (or shunt) across it with a GOOD .22uF/1000v to .33uF/1000v plastic cap.  
This forms a resonant LC tank circuit (close to 60Hz)who's Zed is very high at that frequency and that will effect the max ripple voltage to the filter cap.  
We use to use this little trick in "Big Butt" 2.5Kv to 4Kv 1amp power supplys where the darn bleeder resistors alone would be forced to dissipate huge amounts of heat when not under a load while the caps charged way up.  
Imagine: 3Kv and a 360K bleeder resistor.  
At about 8.5ma of current, it would have to sink 25 watts!! That's some heat man!  
Mission Amps
2/22/1999 7:18 AM

I like this tip, and thanks for bringing it up. If I'm reading you correctly, then it's almost a tuned notch filter (around 60HZ) which attenuates the ripple freq. , right?  
Does one have to experimentally tune it, or are the values you mentioned close enough?  
... Whit
2/22/1999 9:55 PM

Yeah, that is basicly it. It is not a super high Q filter, as it is only one pole, but it does make a difference.  
You might have to tune it in a bit with the cap value because it is supposed to be a 60Hz resonant tank circuit.  
At resonance there is maximum Zed.  
Not knowing the real inductance of the choke, when loaded and drawing current, is a bit of a problem but those figures I gave were fairly close to get started.  
2/22/1999 11:20 AM
Jim S.

"But I've built class A 6V6 amps using 400vdc with good results."  
What value cathode resistor do you use on these amps? At 400V plate voltage, you probably don't want each 6V6 to draw much over 30ma at idle or you'll exceed plate dissipation ratings (12 or 14 watts).
2/22/1999 10:15 PM

I'm using a 10 watt 340ohm cement resistor in that design right now but have run it down to 300ohms with only slight difference in tone.  
Actually I've even used as low as 270ohms and run it with very high idle current at close to 46ma with 350vdc, but,this was a little hard on the 6V6GTA tube and PT, so I've bounced around the Rk values from 300 to 340.  
The only "problem" I've seen is the ability of the no gain cathodyne PI to drive the grids to B... if you let it.  
I'm running the driver triode ahead of the PI pretty high in voltage and a very little Ck and I've used as high as 68Ks on the PI.  
I like the B+, measured at the plate/ground, at around 370vdc to 380vdc.  
I've switched over to smaller 120ma Thordarson PT that has a little less VAC from the start.  
This still works out real good with either Rks and I can cheat a little by moving slightly away from hard class A.  
It seems solid in class A with anything up to near full blast drive voltage, and I can get a solid 12-15 watts out with a great tone.  
I am working on a similar version now but with BF tone section and 2 tube reverb.  

  Page 1 of 2 Next> Last Page>>