small 6y6/6w6 amp

[8wattjack]

1

[shooter]

its just this little ghost note, almost like an occilation

try gator clipping a bigger PS cap for the pre section, maybe 47uF+
try disconnecting the NFB
Try adding a bypass cap on V1b, 10uF range


does it kinda sorta sound like a tremolo effect?
do you have a scope?, If so, look at the pre cathodes, AC, look in the weeds for a coherent signal.

[shooter]

Its currently 50/50/20/20

your layout showed only 2 8uF, I typically use 100uF at B+ and screen, then 47uF at each pre node.  You can try increasing the dropping R's instead of boosting the caps but this WILL change your down-stream volts.  I'm not skooled enough to explain, but if the filter nodes are "off" they can aid in producing motorboating(?) osc.

[shooter]

Will the ez81 hold up to 100UF

I forgot you're tube rectifying, I'm not sure and don't have the datasheet yet. you strapped the "extra" 47uF on the pre tap?  if so that shouldn't mess with your rectifier, my understanding it's the 1st cap passed the rec. tube that matters most.

If 4500k is good

Have you been following the OT impedance thread by Silverfox, and obliquely mine on Circuit question.  Tons of good things on OT and proper loading.
You can ballpark what your tube wants for plate load by calculating R with ohms law using plate V and tube I.  That gives you the primary side, then do the best you can with what you have to make that happen with speaker(s) on the output side.  When you're happy, then you can decide if you want to buy or have build some "odd" value OT

[PRR]

> screens are on an ultralinear tap then through a 180k 2 watt resistor

180_K_ seems impossibly high.

And UK 180K is not what is shown in the layout post at top of thread.

> screen voltage fluctuates from about 105v to 125v

So? Not of itself a big deal.

> Would a negative voltage on the grid help?

Same-as positive 18V on cathode. Should be no-change.

> The tube data calls for 4500 ohms at higher voltage.

Which "tube data"? 6Y6 and 6W6 are not-quite the same tube. There's several suggested conditions around.

> 2500/8 output trans. It sounds better on a 16 ohm speaker which would have the primary at 5k

V/I. You have 210V across the tube and about 33mA plate current. 210V/33mA is 6.3K, a long way from 2.5K, and a 5K is a better fit. I think you want to run the tube hotter, or make your power-stage B+ drop some.

Your 210V pushes the 6Y6 ratings (although I think they are conservative).

Let's see what 10W Pdiss at 2500 Ohms looks like.

Well actually Tung-Sol 6Y6 1961 shows 200V 61mA at 135Vg2 14Vg1 (212r cathode bias) works for 2,600r load. I think you want to run the power tube hotter, or drop the power-stage B+ some.

[PRR]

.

[shooter]

but the tube runs away quickly with a lower value

You mean the current starts ramping up and running away?  on everything or just G2?

[HotBluePlates]

Okay. Good schematic. It is somewhat what I had in mind, but The OT has a screen tap. Should I do away with it? ...

Insulate the end of the screen tap and tuck it out of the way.

I have a scratch build where the OT has screen taps.  I don't use mine, either.

[shooter]

10 K

build where the OT has screen taps

I use mine , but, what is it 40% of? idle voltage drop, signal voltage swing?  I have yet to "find" the tap 40% less than anything I measure.

[HotBluePlates]

... OT has screen taps

I use mine , but, what is it 40% of? idle voltage drop, signal voltage swing?  I have yet to "find" the tap 40% less than anything I measure.

"40%" of the windings between the B+ and the Plate end.  In other words, the screen sees a varying V_G2 = 40% of the plate signal voltage swing.

You'll only measure that if you're monitoring a.c. volts at the screen tap relative to the B+ point.  And that can be a dangerous place (for the meter) to measure voltage with some meters & some amps.

[shooter]

Here is what I did.

And it sounds like?

Thanks HBP, one measurement I haven't done yet.

[shooter]

I feel better about the end result now

It sounds to me like this is just the beginning

[HotBluePlates]

... I am also reading that the data on a 6y6 is far from being the whole truth. ...

> screens are on an ultralinear tap then through a 180k 2 watt resistor

180_K_ seems impossibly high.

... It is that high to get down to the max 135v on the screens. ...

G2           163

... I'm wondering if the G2 voltage is okay. Way high according to data. ...

There are lots of data sheets out there.  Some require interpretation.  If you are worried about something they say, you really oughta link to it so we see what you see.

G.E. 6Y6GT
Tung Sol 6Y6GA
Sylvania 6Y6GA

You seem to believe the 6Y6 screen is rated for 135v maximum.  That seems reasonable from the Max Ratings on the 1st page of the G.E. sheet above.  But pages 3 & 4 of the same sheet have curves for 150v on the screen.  There has to be more to the story...

The Tung Sol 6Y6GA sheet claims (like the other TS sheet PRR posted) a Max Screen Supply Voltage of 200v, then gives a cryptic "See J5-C4" under "Max Screen Voltage".  At a minimum, this implies the screen will not be harmed if 200v is applied (at least under some condition). [There is probably a note "J5-C4" somewhere that applied to every sheet in the collection from which this taken, but it isn't included here.]

The Sylvania 6Y6GA sheet finally gives the "Rest of the Story".  Looking for Screen Voltage, the sheet direct the reader to "See Rating Chart" (on the last page).  The screen can be operated at Max Supply Voltage (200v) when it dissipates zero-power (i.e., no current flows), and this drops to 50% of Max Supply Voltage when dissipating all 1.75w of the screen's rated max dissipation.  That would happen at 1.75w/100v = 17.5mA of screen current.

135v/200v = 67.5% of screen voltage rated allowed, and the curve shows this applying when the screen dissipates ~88% of it rated maximum, or 1.54w.  1.54w/135v = 11.4mA.

Unless your 6Y6's screen is idling at 11.4mA, you don't need to be confined to 135v.  To verify, measure resistance of the 8.2kΩ (amp off), then measure d.c. volts dropped across the 8.2kΩ (amp on).  Use Ohm's Law to calculate screen current.

You can use that rating chart to figure the size your screen resistor should be.  Easiest path (if you have a DMM with a HOLD function) is to set up your meter to measure milliamps, unsolder the 8.2kΩ resistor, and attach the meter in place of the resistor.  Crank the amp as high as it will go, play as loud as you can.  Use the HOLD function to review the highest average current that happened while you played.  Calculate allowable max output screen voltage (1.75w / Average Current at Max Power), note difference between that & your idle screen voltage node, divide that difference by your idle screen current (found earlier).  You now have a screen resistor with minimum impact on output tube compression while protecting the screen.

You can always use a higher-resistance for the screen dropping resistor; only impact for a reasonable value will be increased compression at high volume and less power output.

[HotBluePlates]

... Basically (in amps for dummies terms) as long as you don't exceed the current rating for the screen, it should be fine. Correct?

Well, Power = Voltage * Current.

If you're unsure or don't want to calculate, 135vdc is likely always safe.  If you're willing to calculate/measure, you can go up to at least 200vdc (and maybe more, because tubes don't break right at even "Absolute Max" ratings).

If you calculate/measure, you can find your tube's real-world idle and max-signal screen current, and calculate a dropping resistor that will always keep the tube protected.

[shooter]

wish I could fit all the PS in one can

maybe, if you don't have anything below the can, surf for radial lead 450vdc caps, I use nichicon 100uF or 47uF, about 2" tall n 1/2" dia, hang them in parallel with your can cap to double up the farads.  you could probably get 3-4 under a standard JJ 40/20/20/20.

[shooter]

How do I quote someone properly

highlight and copy, to be quoted text, select the quote Icon, just right of the # sign icon
then paste the copied txt in the center, tween the 2 ][

[Willabe]

How do I quote someone properly??? I seem to have troubles there!!!

Hit the quote button upper right corner of post you want to quote.

Then high light part you don't want, delete. Bring this part [ / quote ] up to end of quote.

Then write your response, post. 

[purpletele]

8Watt,

I have been through a lot of speakers in the last three years.  The 10" C Golds are really nice in a pair.  The consensus with my colleagues is that it is a great speaker or pair and it cuts well and projects very well.

I have two golds and a cream in 12" and I like them quite well, however, I am more impressed with Emminence speakers from a performance and price point.

You cannot go wrong with a pair of 10" golds.