Flexible power tranny option

2nd harmonics for the masses

Flexible power tranny option

Postby elbinster » Thu Apr 15, 2010 10:37 am

After much success with the ST35 project (thanks very much for the help there!), I convinced myself to upgrade to high-sensitivity speakers, which now forces me to build a low-power amp.

Any ways, I got the impression that the 6B4 tubes would be more common and less expensive, but from the offerings I've seen, I wouldn't mind making the G*S*G upgradeable to a 6A3, 2A3 or other similar tube.

I have a plan in place, but need to check my logic here...

To use a 2A3, is there a reason I couldn't use a simple resistor-divider to cut the 6.3V down to 2 or so? Is the max amps from the power tranny the only thing preventing me from doing this? The XPWR005 would be maxed out by the 2A3 heaters, but since the 6SL7 doesn't take much, how much should I worry about overdriving the secondaries? Would something like an XPWR131 (330-260-0-260-330@175, 6.3@4A, 5@3A) and a 5xx rectifier suffice? Or would I need to overkill with an XPWR016 (350-0-70-350@450, 6.3@5A, 6.3@5A)?

Also, how 'separate' do the heaters need to be for DHT tubes? Would a 12.6V w/ a center tap be good enough? Or do they need to be completely separate windings? Would an XPWR114 (330-0-330@200mA, 6.3-0-6.3@3A 5@5A) suffice.


Since Edcor is at an 8 week lead time, would it behoove me to ask Edcor for a custom power tranny?
User avatar
elbinster
 
Posts: 81
Joined: Sat Dec 12, 2009 3:45 pm
Location: Springfield, VA

Postby battradio » Thu Apr 15, 2010 11:50 am

2A3'S DRAW 2.5 AMPS . 2.5 amps X 4 volts =10 watts per tube , thats two 20 watt resistors you would need . Heat wise you will be beter off with seperate filament transformers .
Mark
Image
User avatar
battradio
KT88
 
Posts: 927
Joined: Tue Mar 10, 2009 12:58 am
Location: near ST.Louis MO.

Postby TomMcNally » Thu Apr 15, 2010 1:47 pm

6A3's would just require a socket change, but they are very rare.
The logic behind the 6B4G is that they are in current production,
and the 6 volt filament and octal socket made for a better design.

I'd stay away from 2A3's ... not worth the trouble.

There are mods detailed in this forum to build a g*s*g with
300B's ... all it requires is a simple resistor change and of
course, UX style sockets.
User avatar
TomMcNally
Darling du Jour
 
Posts: 2729
Joined: Sat Nov 19, 2005 2:19 pm
Location: Northfield, NJ

Postby elbinster » Fri Apr 16, 2010 9:01 am

Well I can't argue with that. 20W per channel of pure heat seems excessive in a 3W amp. I might try to wire up the guts for a 300B change depending on what I find available.

Thanks for the advice!
User avatar
elbinster
 
Posts: 81
Joined: Sat Dec 12, 2009 3:45 pm
Location: Springfield, VA

Postby Shannon Parks » Sat Apr 17, 2010 6:11 am

I think If anyone wants a alternate triode from a 6B4G, the 300B is a slam-dunk and it keeps the simplicity of project. 2A3s really start to complicate things, and the old stock 2A3s are quite expensive.

Shannon
User avatar
Shannon Parks
Site Admin
 
Posts: 3764
Joined: Tue Mar 18, 2003 5:40 pm
Location: Poulsbo, Washington

Postby EWBrown » Sat Apr 17, 2010 6:45 pm

If you really want to go with 2A3s, then just use two separate 2.5V, 3A filament trannies, and run the filaments on AC. Might be a "plus" to use a 50 to 100 ohm WW "hum" pot across the filaments, with the wiper connected to the "cathode" resistor and bypass cap. Trying to drop 3.8VAC @ 5A total through resistors, just generates a lot of unwanted under-the-chassis heat.

I suppose that the really truly "elegant" (or overkill) solution would be to order a custom made power trannie from Edcor, with two 6.3VAC, and two 2.5VAC filament windings, plus the standard HV AC secondary, and perhaps a 5VAC, 3A filament winding to allow using 5AR4s or 5U4Gs.

Any new "custom" power trannie ordered through Edcor becomes a new part number, added to their already extensive list.

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !

Postby elbinster » Tue Apr 20, 2010 9:48 am

Well it certainly doesn't seem like it'd be worth the effort getting this puppy rigged up for 2A3 tubes, but shouldn't be too much of a hassle if a stockpile of 2A3s is found in 20 years.

I'll work up a schematic that lets me put in a few different sockets and see what else comes up.

From reading other posts, it seems like it might be a good idea to swap in some hum pots. Electrically, I can understand what they're doing, though I don't quite understand why they work. What's more important is knowing how to use them. Do I simply ground the inputs and twist the pot until I can't hear anything?n Or am I shooting for a balance in bias?
User avatar
elbinster
 
Posts: 81
Joined: Sat Dec 12, 2009 3:45 pm
Location: Springfield, VA

Postby lth1 » Tue Apr 20, 2010 10:41 pm

I have had two sockets (8 and 4 pin) in my GSG from the start. At first it was set up for 6B4G's and 2A3's, used a resistor to reduce filament voltage to 2.5. The GSG is now wired for the 6B4G and 6A3, saw no reason for using the more expensive 2A3. NOS 6B4G's are as good as any 2A3 IMHO, and much cheaper. I have a pair of highly modified Bottlehead Paramours (2A3) using TJ 2A3 tubes, and in my GSG I am using Tung-Sol 6B4G tubes, I much prefer the sound of the Get-set-Go amp.

My next GSG will only have a single 8 pin octal socket for the 6B4G, but two octal sockets, one for 6SL7's and 7F7's, and the other for 12SL7's.

Lee Hankins
lth1
 
Posts: 109
Joined: Sun Mar 21, 2004 12:29 am
Location: "End of the Road" Homer, Alaska

Postby EWBrown » Wed Apr 21, 2010 7:38 am

The hum pots are only necessary for AC powered filaments, the idea is to reduce (and not totally eliminate) the amount of audible 60 Hz hum, with careful adjustment, generally close t the halfway point. Bottlehead goes ths one step further, they use a 10 ohm WW pot with a 22 ohm, 5W WW resostor connected to each end, this effectively behaves like a 54 ohm pot with a physically "wider", and less "sensitive" adjustment range.

The way the hum pot works is to create a very simple resistor "bridge" circuit, with the filament and the two halves of the hum pot, and by careful adjustment, the AC voltage on the filament can be balanced out in relation to the tube's grid, which will reduce the amount of hum heard on the output

Some earlier designs used a center-tapped filament transformer, the cathode resistor and cap connected between the CT and ground buss, this works, but does not allow for "fine tuning" out of any residual hum. This approach is best used with push-pull designs, as the AC hum will cancel itself out.

J/J makes a 2.5V filament version of the 300B, which they call the 2A3-40 (2A3 with 40W plate dissipation). It would be nice if they could also design and produce a 6A3-40 (and 6B4G-40) which could efectively become a 6A3 (or 6B4G) on steroids.

/ed B
Real Radios Glow in the Dark
User avatar
EWBrown
Insulator & Iron Magnate
 
Posts: 6389
Joined: Wed Mar 19, 2003 6:03 am
Location: Now located in Clay County, NC !


Return to get*set*go

Who is online

Users browsing this forum: No registered users and 10 guests