Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: TIMBO on January 22, 2016, 05:16:03 am
-
Hi guys, got a tricky one.
I have a PT that only has 12.6v CT @ 4A (not 2x6.3v taps)as its heater supply.
The amplifier that had this PT was a 100w PA using 6550 tubes.
As per the schematic of the amp, the tubes were arranged as per my top drawing. To me it seemed lopsided as the tubes current on either side of ground are not equal,but this is how the amp was wired and the total current is more than 4A.
I want to reuse this PT in another build and I took a stab on how it could be laid out. Thanks
-
I don't think the unbalanced current draw matters. It's as if you had 2 separate filament trannies, ea capable of 6.3-0 @ 4A. Here, the secondaries "happen to" share the same primary coil, and are wound out of phase so that their 0's are conveniently located in the middle, creating 6.3-0-6.3 @ 4A. So the 2 secondaries don't know or care if they have balanced current draw. (Maybe the primary will begin to care at some point, but let's assume the designers did their job well enough.) Current in excess of 4A will drop voltage, but as long as 5V remain, the tubes will be happy.
My question: is it noisy? If the heaters are quiet and free from hum, then the stock circuit is OK. But in some old PA amps, some level of hum may have been tolerated. The one-sided ground scheme at ea tube negates the benefit of humbucking by using twisted pairs of filament wires. Not so much an issue with PP power tubes, because the PP arrangement is noise & humbucking anyway. It might be an issue with other tubes. OTOH, twisting filament wiring is considered over-rated by some; and many quality vintage amps did not do so with no ill affects.
If you need or want to use twisted pairs to buck hum, there's another alternative: Make a series-pair of 2X 6.3V tubes. This sums to 12.6V for ea series pair-- BUT the filament current draw must be the same for ea of the two tubes in a series pair. Wire the series-pairs in 2X parallel strings from the tranny. Use a pair of ground reference resistors for each string.
If there's an odd tube with no mate for a series pair it can be left one-side grounded. For lowest hum, do this with a tube late in the chain and/or with a low noise tube (not an EF-86 if possible). Any heater hum that originates early will be re-amplified in later gain stages.
-
JJ has a good explanation. I looked at your schematic, and I'm curious about the 6V6 at the end of the chain. AFAIK, that should be before the preamp tubes because of current draw & noise.
-
Thanks guys, This is the amp I'm looking at cloning http://el34world.com/Forum/index.php?topic=19585.0 (http://el34world.com/Forum/index.php?topic=19585.0)
And this is the PA that the transformers came out of.
-
BTW: for more on filament wiring schemes, see Hotblue's Reply in this thread: http://el34world.com/Forum/index.php?topic=19701.msg205363#new (http://el34world.com/Forum/index.php?topic=19701.msg205363#new)
-
Thanks JJ, I have been pretty lucky with beating the hum from heater wires, I usually tuck then into the corner of the chassis when the tubes are lined along the back of the chassis.
-
see attached.
--pete
-
OOoooo, Nice one pete. I was wondering if I could go down that track. :icon_biggrin:
-
> total current is more than 4A.
z-y load: 1.6A+1.6A+0.6A= 3.8A
3.8A is less than 4A, OK
x-y load: 0.3A+0.6A= 0.9A
0.9A is less than 4A, OK
with "PA Head":
x-y load: 0.3A+0.6A+0.3A+0.3A+0.3A= 1.8A
1.8A is less than 4A, OK
I say each winding is under-loaded. One only slightly, the other is loafing.
I say you could add a half-dozen more small bottles and still be cool.
The grounded winding IS awkward for guitar-amp level circuits. It is a bit odd they did that for microphone inputs. It may have been "Gee, we got a hundreds Watts!!" rather than "We have pure-pure sound." (100W is poop-yer-pants big for 1960.)
-
Hi guys, This is the full article.
-
Hi guys, I'll try this arrangement.
-
I think the intent is that they wanted to leave one 6.3V supply available for external gear - so to allow that supply to have the max capability, they internally maxed the other 6.3V.
The EF86 was probably going to give good enough hum because the input grid is pretty well shielded from heater pins and internals, especially if using a spigoted socket, and placing the transformers a fair distance away.
-
NZ amp makers, Jansen, used to do their amps with 12.6Vac wiring. Reduces current draw for the whole heater winding. Each pair of 6.3VAC output tube filaments in series only draws what one output tube heater alone would otherwise draw (and of course the pre-amp bottles' heater current is half of what a 6.3VAC scheme is). I had occasion to rebuild one from scratch just over a year ago. I put a 100R wire-wound humdinger pot in it.
-
I'll try this arrangement (take 2)
-
I think the intent is that they wanted to leave one 6.3V supply available for external gear - so to allow that supply to have the max capability, they internally maxed the other 6.3V.
Good thinking!
-
I'll try this arrangement (take 2)
What about putting the pairs of 6550s together and the pair of 6SK7s together and the 6V6 with the EF86, and run the 12AX7s as 12V each on pins 4 and 5?
-
That was Dummyloads idea as well. Ok that seems to be a better way to go, thanks guys.
-
Hi guys, I'm having a bit of battle with doing the circuit as per DL's schem.
I checked the voltage across the winding and a nice 15v is present.
Wiring as per schem getting required volts across pins 4,5 on the 9 pin sockets and the same for the octals, at this stage no 100ohm CT resistors added.
When the CT resistors added I do not get a voltage reading across the CT resistors.
I put in a couple of valves and the heaters didn't glow.
I can't see what I missed. Thanks
-
Are you sure you have not connected the two 6.3v windings counterphase ?
Franco
-
Hi guys, I'm having a bit of battle with doing the circuit as per DL's schem.
I checked the voltage across the winding and a nice 15v is present.
Wiring as per schem getting required volts across pins 4,5 on the 9 pin sockets and the same for the octals, at this stage no 100ohm CT resistors added.
When the CT resistors added I do not get a voltage reading across the CT resistors.
I put in a couple of valves and the heaters didn't glow.
I can't see what I missed. Thanks
that plan should work unless something is mis-wired. clock-bottom/counter-top?
--pete
-
Hey Pete,
clock-bottom/counter-top? not sure what that means.
Anyway.............
MY BAD, it seems that the dope putting this amp together has over looked a simple task.
When removing the PT from the original chassis I cut the CT (four strands of wire) off a solder pad on the chassis and when tiding up the unused wires with heatsrink (CT included) I forgot to solder the strands together. WHAT A DORK :BangHead:
With all valves in a great 12.7v across the two rails. Thanks
-
pin orientation on the socket: was thinking perhaps that you may have mis-wired. :icon_biggrin:
bottom view - clockwise. clock-bottom
top view - counterclockwise. counter-top.
--pete
-
:icon_biggrin:
-
Hi guys, Seems that we are not totally done yet.
One of the local guys has brought up a good point.
The heater/cathode has a max of 200v and the voltages that were supplied to me has the cathode seeing approx. 300v it was suggested that a floating isolated supply be added.
(http://i761.photobucket.com/albums/xx254/Timbo-08/001%209_1.jpg) (http://s761.photobucket.com/user/Timbo-08/media/001%209_1.jpg.html)
http://www.altronics.com.au/p/m1000-600-ohm-600-ohm-telephone-isolation-transformer/ (http://www.altronics.com.au/p/m1000-600-ohm-600-ohm-telephone-isolation-transformer/)
Could it be something like this :dontknow:
-
that's not going to work.
use a separate 6.3VCT 500mA filament transformer & lift the CT off ground about 100V with a divider network hanging off the B+ rail. or just leave it floating.
--pete
-
Thanks Pete, so what do we do with the EF86??
-
Thanks Pete, so what do we do with the EF86??
put a resistor under it. 6.3V/200mA = 31.5R use a 33R3W part: a 2W will work but will get warm...
what were you going to use for the coupling transformer in that drawing? whatever it is, it needs to be 2:1 for best results.
--pete
-
(http://i761.photobucket.com/albums/xx254/Timbo-08/001%2010_1.jpg) (http://s761.photobucket.com/user/Timbo-08/media/001%2010_1.jpg.html)
Would adding a fuse to the circuit be needed.
-
No Tim - it would be unlikely to blow, and given the circuit is very simple, it would be highly reliable once you've checked it. Like a directly heated rectifier heater - it is hazardous.
That heater should be connected to the 6V6 cathode - don't leave heaters floating.
-
Like this
(http://i761.photobucket.com/albums/xx254/Timbo-08/001%2011_1.jpg) (http://s761.photobucket.com/user/Timbo-08/media/001%2011_1.jpg.html)
-
:smiley: